{ "row": "ai-incident-database", "entity": "ai-incident-database", "address": "ai incident database", "addresses": [ { "address": "ai incident database", "slug": "ai-incident-database-aio-20261029", "addr_id": "ADDR-ebdc2a94e8d6", "obs_id": "OBS-d2d81aafd207", "axn": "AXN:0AA7.CAPTURE.🖐️🏴🎆🌟🎨🚩" }, { "address": "what counts as an incident in the ai incident database", "slug": "what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029", "addr_id": "ADDR-7858f0aa2fe5", "obs_id": "OBS-bd807797964a", "axn": "AXN:0AB7.CAPTURE.🗿♌🪟🐝🌅⚡" } ], "epoch": "2026-10-29", "type": "C (public entity on which the archive bears: a database; an entity of the MMRS meta-aggregator)", "surface": "Google AI Overview (expanded)", "auth": "signed out, incognito", "status": { "procedure": "EA-NEGONT-02 v0.9, #1690: equal case, ρ = extends, read from the MMRS record and the two capture records; archive selected by D/R/O (one deposit of one candidate admitted, #543 at R; hop of 41 read and rejected; O-class matched nothing), with the MMRS record (37 claims) and the Capture Registry's findings and analysis (12 claims) ledgered as the archive's records of the entity; the field B by three origins (14 card sources for 15 cards, 4 own pages, 2 Wikipedia pages); the composer read the ledgers, never the transcripts, except where the brief asks for T's claims and the delta", "ledger": "unaudited (§3.8: one extractor per ledger); field quotes 75/75 verified by script against saved text, PDF text, raw wikitext, oEmbed title or card text; archive quotes 50/50 verified verbatim against the registry, census.json, articles.py and AXN-0174; carried as marked: MMRS-01 contested (F73), MMRS-12 and MMRS-19 and CAP-1-04 with defects noted", "frozen": false, "breaches": [] }, "spec": "EA-NEGONT-02 v0.9, #1690", "objects": { "T": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/T1-ai-incident-database-aio-20261029.txt", "sha256": "d24b458664f3be0edd86ea85d1e37ae00af947f9c2776053de33993034abb8d5", "words": 427, "cards": 15, "words_note": "two transcripts at two addresses, summed for the page's table (150 + 277 words; 6 + 9 cards); each kept whole under 'transcripts'. The path is the first capture's record page.", "claims": [ "T1-1 definition: 'a free, open-source project managed by the Responsible AI Collaborative that indexes real-world harms, near-harms, and failures caused by artificial intelligence systems' (markers: the database's card and the Collaborative's link)", "T1-2 model: 'Modeled after safety incident reporting systems used in aviation and cybersecurity, it collects public and crowdsourced reports to help the industry learn from past failures.'", "T1-3 scope: 'algorithmic bias, discrimination, privacy leaks, misinformation, deepfakes, autonomous vehicle accidents, and safety-critical system failures'", "T1-4 growth: 'Incidents tracked by the database have surged significantly, with deepfake and scam-related entries multiplying rapidly alongside broader media scrutiny.' (markers: an unresolved token and the TIME clip)", "T1-5 derived layer: 'Projects like the MIT AI Incident Tracker parse the database using taxanomies and LLM pipelines to categorize risks into structured analytical views.'", "T1-6 closer: an offer to search incident types or 'learn how to submit an incident report'", "T2-1 definition: 'an alleged harm or near-harm event to people, property, or the environment where an AI system is implicated'", "T2-2 criteria: 'an adaptive, specific set of criteria outlined in its AIID Editor's Guide'; 'For a report to be accepted as a true incident record, it must meet three core baseline thresholds'", "T2-3 specific event: 'a distinct, real-world occurrence with clear details on what happened, to whom, and where. Theoretical vulnerabilities do not qualify.'", "T2-4 AI implication: 'the harm likely would not have occurred without the AI's output, omission, or use'", "T2-5 harm or near-harm: 'an adverse effect or a credible chain of events nearly producing one'", "T2-6 categories: 'Using frameworks like CSET's taxonomy, incidents are grouped into categories such as physical safety, discrimination and fairness, financial loss, privacy and security breaches, and misinformation or scams.'", "T2-7 distinctions: incident 'A real event where AI caused or nearly caused harm'; issue 'A potential harm or vulnerability that has not yet materialized in the real world'; variant 'A separate occurrence involving the same AI system, similar causes, and matching outcomes as an existing record'", "T2-8 closer: 'Are you trying to submit a specific case to the database, or are you researching a particular type of AI safety failure … for your own project?'" ], "register": { "register": "data/EA-WG-CAPTURES-01.json", "slug": "ai-incident-database-aio-20261029", "obs_id": "OBS-d2d81aafd207" }, "transcripts": [ { "slug": "ai-incident-database-aio-20261029", "address": "ai incident database", "addr_id": "ADDR-ebdc2a94e8d6", "obs_id": "OBS-d2d81aafd207", "axn": "AXN:0AA7.CAPTURE.🖐️🏴🎆🌟🎨🚩", "date": "2026-10-29", "surface": "Google AI Overview (expanded)", "auth": "signed out, incognito", "sha256": "d24b458664f3be0edd86ea85d1e37ae00af947f9c2776053de33993034abb8d5", "sha256_of": "the registry's 'transcript' string (utf-8)", "sha256_raw": "c81388bb34899baf208d3105545a61604f5227f0d80989aefcca4b9772e9746c", "words": 150, "cards": 6, "card_rail": [ "1. AI Incident Database — AI Incident Database", "2. MIT AI Risk Initiative — MIT AI Incident Tracker", "3. Partnership on AI — AI Incident Database - Partnership on AI", "4. OECD AI Policy Observatory — AI Incident Database", "5. Kaggle — AI Incidents, Failures & Controversies Database - Kaggle", "6. LinkedIn·TIME — AI Incident Database | TIME posted on the topic (video, 1:35)" ], "claims": [ "T1-1 definition: 'a free, open-source project managed by the Responsible AI Collaborative that indexes real-world harms, near-harms, and failures caused by artificial intelligence systems' (markers: the database's card and the Collaborative's link)", "T1-2 model: 'Modeled after safety incident reporting systems used in aviation and cybersecurity, it collects public and crowdsourced reports to help the industry learn from past failures.'", "T1-3 scope: 'algorithmic bias, discrimination, privacy leaks, misinformation, deepfakes, autonomous vehicle accidents, and safety-critical system failures'", "T1-4 growth: 'Incidents tracked by the database have surged significantly, with deepfake and scam-related entries multiplying rapidly alongside broader media scrutiny.' (markers: an unresolved token and the TIME clip)", "T1-5 derived layer: 'Projects like the MIT AI Incident Tracker parse the database using taxanomies and LLM pipelines to categorize risks into structured analytical views.'", "T1-6 closer: an offer to search incident types or 'learn how to submit an incident report'" ], "markers": "inline links are opaque google.com/goto tokens; at T1 the definition's markers are the database's card and the Collaborative's link, the Research Extensions marker the MIT tracker's, and one Growth marker the TIME clip's; at T2 the definition and thresholds are marked to one token, the Editor's Guide's", "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/T1-ai-incident-database-aio-20261029.txt" }, { "slug": "what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029", "address": "what counts as an incident in the ai incident database", "addr_id": "ADDR-7858f0aa2fe5", "obs_id": "OBS-bd807797964a", "axn": "AXN:0AB7.CAPTURE.🗿♌🪟🐝🌅⚡", "date": "2026-10-29", "surface": "Google AI Overview (expanded)", "auth": "signed out, incognito", "sha256": "7c373254bcea403d8d027e7d97f7625a9d9906fdadc5722b9251bb4f65277428", "sha256_of": "the registry's 'transcript' string (utf-8)", "sha256_raw": "804acc16da914b9cb0a706b7a21949f523e0174b6c0c84c5a39259fe17264a99", "words": 277, "cards": 9, "card_rail": [ "1. MIT AI Risk Initiative — MIT AI Incident Tracker", "2. AI Incident Database — Editorial and Research Guide for the AI Incident Database", "3. AI Incident Database — Defining an “AI Incident”", "4. LinkedIn·TIME — AI Incidents Rise 50% Year-Over-Year: AI Incident Database | TIME posted on the ... (video, 1:35)", "5. YouTube·HarrietHacks — Examples of AI real attacks (video, 6:59)", "6. AI Incident Database — The details about who and why for database - AI Incident Database", "7. The Association for the Advancement of Artificial Intelligence — Lessons for Editors of AI Incidents from the AI Incident Database - AAAI Publications", "8. ACM Digital Library — Why We Need to Know More: Exploring the State of AI Incident Documentation ...", "9. YouTube·International Association for Safe & Ethical AI — Establishing Effective Serious Incident Reporting Frameworks – Jimmy Farrell (video, 19m)" ], "claims": [ "T2-1 definition: 'an alleged harm or near-harm event to people, property, or the environment where an AI system is implicated'", "T2-2 criteria: 'an adaptive, specific set of criteria outlined in its AIID Editor's Guide'; 'For a report to be accepted as a true incident record, it must meet three core baseline thresholds'", "T2-3 specific event: 'a distinct, real-world occurrence with clear details on what happened, to whom, and where. Theoretical vulnerabilities do not qualify.'", "T2-4 AI implication: 'the harm likely would not have occurred without the AI's output, omission, or use'", "T2-5 harm or near-harm: 'an adverse effect or a credible chain of events nearly producing one'", "T2-6 categories: 'Using frameworks like CSET's taxonomy, incidents are grouped into categories such as physical safety, discrimination and fairness, financial loss, privacy and security breaches, and misinformation or scams.'", "T2-7 distinctions: incident 'A real event where AI caused or nearly caused harm'; issue 'A potential harm or vulnerability that has not yet materialized in the real world'; variant 'A separate occurrence involving the same AI system, similar causes, and matching outcomes as an existing record'", "T2-8 closer: 'Are you trying to submit a specific case to the database, or are you researching a particular type of AI safety failure … for your own project?'" ], "markers": "inline links are opaque google.com/goto tokens; at T1 the definition's markers are the database's card and the Collaborative's link, the Research Extensions marker the MIT tracker's, and one Growth marker the TIME clip's; at T2 the definition and thresholds are marked to one token, the Editor's Guide's", "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/T2-what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029.txt" } ] }, "KO": { "title": "The AI Incident Database", "genre": "knowledge object, equal case, under EA-NEGONT-02 v0.9 (#1690): the archive's sources and the field's admitted on equal terms, ontology included, in one claim space", "status": "composed 2026-10-29 under v0.9 (brief rebuild/negative-of-the-negative/v09/BRIEF-NEW-ROW-AGGREGATOR.md); not frozen; ledgers unaudited", "sentences": [ { "n": 1, "para": 0, "sense": "what it is", "text": "The AI Incident Database (AIID) is an editor-resolved database of AI harm events (MMRS record), \"dedicated to indexing the collective history of harms or near harms realized in the real world by the deployment of artificial intelligence systems\" in its home page's words, which the MMRS record quotes and the OECD's catalogue repeats as a \"free and open-source project\", and which Paeth and colleagues (AAAI-25) describe as a project that catalogs AI incidents and provides a platform to classify them for operational and research goals.", "claims": [ "MMRS-01", "F1", "MMRS-03", "F12", "F49" ] }, { "n": 2, "para": 0, "sense": "the unit", "text": "Its unit is the incident, with the reports that document it attached, and every incident has a stable citation page at incidentdatabase.ai/cite/N (MMRS record); the Editor's Guide defines the incident as \"an alleged harm or near harm event to people, property, or the environment where an AI system is implicated\" and the report as a source document attached to an incident ID, several of which may document one event, and the MMRS record quotes the definition and adds that reports may be news articles, papers or posts.", "claims": [ "MMRS-02", "F19", "F22", "MMRS-06" ] }, { "n": 3, "para": 0, "sense": "purpose", "text": "Its home page sets it beside \"similar databases in aviation and computer security\", to learn from experience and prevent or mitigate bad outcomes; McGregor's founding paper starts from mature industrial sectors such as aviation that collect their failures in incident databases, while intelligent systems \"cause real world harms without a collective memory of their failings\"; the about page invokes the transportation sector (FAA and FARS) and computer systems, the research page \"Defining an AI Incident\" takes aviation's distinction of an accident (substantial damage or loss of life) from an incident (the risk of an accident substantially increased), and Partnership on AI's workstream page places it among safety-critical systems in transportation, healthcare and energy, a crowdsourced central repository of more than 1,200 reports.", "claims": [ "F2", "F66", "F41", "F37", "F11", "F10" ] }, { "n": 4, "para": 1, "sense": "origin", "text": "Sean McGregor's founding paper, submitted to arXiv on 17 November 2020, describes \"an incident collection initiated by an industrial/non-profit cooperative\", searchable by facet and full text over more than 1,000 incident reports, and on 18 November 2020 he introduced it on Partnership on AI's site as \"a systematized collection of incidents where intelligent systems have caused safety, fairness, or other real-world problems\" (the introduction quoted in Wikipedia's reference); the snapshots page asks that the database as a whole be cited by his IAAI-21 paper, and the MMRS record dates the founding to 2020 as the arXiv paper describes it, with more than 1,000 reports at launch.", "claims": [ "F65", "F67", "F68", "F74", "F56", "MMRS-19", "MMRS-05" ] }, { "n": 5, "para": 1, "sense": "operator", "text": "The home and about pages call it \"a project of the Responsible AI Collaborative, an organization chartered to advance the AI Incident Database\"; the OECD's catalogue entry (uploaded 25 April 2024) names the Collaborative as developing organisation and MIT's tracker credits it with creating and maintaining the database; the about page credits Sean McGregor with launching it and names him Executive Director, heads a logo block \"Database Founding Sponsor\" whose image the page labels Partnership on AI, and records that two people contributed to its early stages through their roles with the Partnership on AI; the MMRS record gives the same operator, founder, director and founding sponsor.", "claims": [ "F4", "F14", "F8", "F46", "F47", "MMRS-04", "MMRS-37" ] }, { "n": 6, "para": 1, "sense": "the encyclopedia", "text": "Wikipedia has no article of its own for it: \"AI Incident Database\" redirects, with no section anchor, to Partnership on AI (a redirect from a subtopic, with possibilities), whose article records that Partnership on AI announced it in November 2020 as \"a tool to identify, assess, manage, and communicate AI risk and harm\", that Partnership on AI's response to NIST presented it as meeting attributes of the AI Risk Management Framework, that it was among the Safety Critical AI program's initiatives, and, under programs and initiatives, that it \"shifted to a new special-purpose independent non-profit in 2022\", the non-profit unnamed; the article is tagged for primary sources (April 2025) and as AI-generated (October 2025).", "claims": [ "F69", "F70", "F71", "F72", "F73", "F75" ] }, { "n": 7, "para": 1, "sense": "origin", "text": "The MMRS record has the database run by the Responsible AI Collaborative since 2020, and Wikipedia has it shift to an independent non-profit in 2022; the two dates stand each with its source.", "claims": [ "MMRS-01", "F73" ] }, { "n": 8, "para": 2, "sense": "the unit", "text": "The Editor's Guide (Version 2, 1 October 2026, by Daniel Atherton and Sean McGregor, drafted with disclosed generative-AI assistance) keeps the incident apart from the AI issue, \"an alleged harm or near harm by an AI system that has yet to occur or be detected\", from the incident ID, the editorial record that bounds and summarizes an event, and from the report: \"A report is evidence; an incident ID is the database’s normalized account of an event\", so that a later source on the same event is usually an additional report, not a new incident.", "claims": [ "F36", "F20", "F21", "F23" ] }, { "n": 9, "para": 2, "sense": "the unit", "text": "A variant is, in the guide, \"a distinct incident involving the same intelligent system, substantially similar causative factors, and substantially similar harms as a known incident\", as the MMRS record restates it; the incident page for 693 describes the variant as having the same causes, harms and system, grouped under the first reported incident, and adds that variants \"do not need to have been reported outside the AIID\".", "claims": [ "F24", "MMRS-07", "F62" ] }, { "n": 10, "para": 2, "sense": "the unit", "text": "The Capture Registry (2026-10-29) records that at the database's name Google's AI Overview gave neither the incident as the unit, nor the reports attached to it, nor the citation page /cite/N, \"collects public and crowdsourced reports\" standing in for the unit, and that at the question of what counts the incident, the issue and the variant were the guide's definitions in substance, the first verbatim, and the report was again not stated; the MMRS record's negative tags hold that it is not a list of research papers, its unit the event and papers entering as reports.", "claims": [ "CAP-1-05", "CAP-2-03", "CAP-2-05", "MMRS-24" ] }, { "n": 11, "para": 3, "sense": "admission", "text": "A candidate passes the guide's five-question test: a specific event, an identifiable AI system, a causal implication (an important role in the chain of events, not necessarily the sole or dominant cause), harm or credible near harm (a chain that could readily have produced harm, stopped by an external factor independent of the system and not a built-in safeguard functioning as designed), and adequate evidence, at least one source supporting the essential event (guide); the MMRS record states the same test and records that admission requires harm or credible near harm.", "claims": [ "F26", "F28", "F29", "F27", "MMRS-08", "MMRS-26" ] }, { "n": 12, "para": 3, "sense": "admission", "text": "At \"what counts as an incident in the ai incident database\" the Overview read the guide and kept its distinctions with one compression, the five-part test given as three thresholds, the identifiable system folded into \"AI implication\" and adequate evidence dropped, so that the composition carried the test's harm condition and lost its evidence condition, the condition on which the MMRS record's account of what its selection leaves out rests (Capture Registry, 2026-10-29).", "claims": [ "CAP-2-01", "CAP-2-04", "CAP-2-02" ] }, { "n": 13, "para": 3, "sense": "admission", "text": "The research page \"Defining an AI Incident\" calls the criteria adaptive, reports \"accepted or rejected on the basis of a growing rule set defined in the Editor's Guide\"; the about page asks that an event be submitted when in doubt, the project meant to converge on shared criteria through the community's candidates, and says its initial set of more than 1,000 incident reports was intentionally broad, an autonomous car killing a pedestrian among its examples; the guide's presumptive exclusions include general criticism of a product or technology class without identified harm and benchmark failures, model errors or hallucinations that remain inside testing or private drafting, \"presumptions rather than absolute prohibitions\", which the MMRS record quotes as \"benchmark failures that stay in testing\" and \"general criticism of a technology\".", "claims": [ "F38", "F43", "F42", "F30", "MMRS-12" ] }, { "n": 14, "para": 4, "sense": "intake", "text": "Anyone may submit (MMRS record; the home page invites incident reports, to be indexed and made discoverable), and the incident editors, Daniel Atherton the lead, resolve submissions and maintain them; the database is managed in a participatory manner by those contributing code, research and broader impacts, CSET and Nikiforos Pittaras (GMF) have contributed its taxonomies (about page), MIT's tracker credits the Collaborative with reviewing all submitted reports, and the MMRS record's intake field states open submission with editors resolving submissions and maintaining records.", "claims": [ "MMRS-09", "F3", "F45", "F44", "F48", "MMRS-37" ] }, { "n": 15, "para": 5, "sense": "the record", "text": "Every incident has a suggested citation crediting its submitters and editors, at incidentdatabase.ai/cite/N (snapshots page); that citation page per incident is the model the MMRS meta-aggregator follows, and the MMRS record observes per item the incident id and citation page, title, date, alleged deployer and developer, harmed parties, attached reports, CSET and GMF classifications and variants.", "claims": [ "F55", "MMRS-16", "MMRS-21" ] }, { "n": 16, "para": 5, "sense": "the record", "text": "A duplicate is merged into the earlier record and its citation page stays up and points there (MMRS record): the citation record for 637 reads \"This incident is a duplicate of Incident 112\", its reports migrated, as the guide's duplicate mapping redirects a retired identifier to its canonical record; ids therefore include merged duplicates, and the highest id seen, 1788 (MMRS record, 2026-10-28; the home page on 2026-10-29), is an upper bound on incidents, against which Paeth and colleagues counted over 750 distinct incidents detailed by more than 3000 indexed third-party reports at their writing.", "claims": [ "MMRS-10", "F64", "F31", "MMRS-20", "F5", "F51" ] }, { "n": 17, "para": 5, "sense": "the record", "text": "The whole database is offered as weekly point-in-time dumps in JSON, MongoDB archive and CSV, listed from 29 March 2021 to 26 October 2026, with a weekly Excel export of numbered incidents and their taxonomy classifications (snapshots page; MMRS record, which counts the weekly snapshots as its version history); the guide warns that the live site can change after an export is created.", "claims": [ "F54", "F57", "MMRS-11", "MMRS-22", "MMRS-27", "F35" ] }, { "n": 18, "para": 5, "sense": "the record kept apart", "text": "The MMRS record keeps itself apart from the entity it describes, each assertion with a source and a basis (operator statement, third-party report, observation, reading) and the lifecycle (active) apart from the observation that saw it (2026-10-28), and it leaves the Wikidata item and any data licence beyond the snapshots page unchecked.", "claims": [ "MMRS-30", "MMRS-31", "MMRS-32" ] }, { "n": 19, "para": 6, "sense": "openness", "text": "The OECD's catalogue gives its usage rights as open source/permissive and its licence as Apache 2.0, without saying of what; the MMRS record observes the code open source and no data licence stated on the snapshots page, and the Capture Registry (2026-10-29) records that the Overview's \"free, open-source\" does not distinguish the two.", "claims": [ "F13", "MMRS-23", "CAP-1-07" ] }, { "n": 20, "para": 7, "sense": "limits", "text": "The guide's governing principle is that the database \"records publicly documented allegations of harm or near harm\" and \"does not independently adjudicate every factual claim\"; it is \"a curated archive of publicly documented incidents, issues, reports, and responses\", not a neutral mirror of all AI harms, a comprehensive denominator for risk rates or an independent adjudication of every event; wealthy countries, prominent companies, spectacular events and English-language reporting are likely overrepresented, and harms to marginalized groups, inside private organizations or in ordinary administration may be underrepresented; and \"A lack of records is not evidence that a sector or country has no incidents.\"", "claims": [ "F25", "F32", "F33", "F34" ] }, { "n": 21, "para": 7, "sense": "limits", "text": "Paeth and colleagues find that certain patterns of incidents present structural ambiguities that challenge incident databasing and that epistemic uncertainty in incident reporting is unavoidable; MIT's tracker states that, the reporting being voluntary, its input is subject to some sampling bias, and that a systematic validation study of its classifications has not yet been completed.", "claims": [ "F50", "F9" ] }, { "n": 22, "para": 7, "sense": "what its selection leaves out", "text": "On the MMRS record's reading a measured study enters as a report when it documents an event, and what an answer surface does with a source on an ordinary day, with no harm event, has no unit: attribution erasure, citation fidelity and provenance stripping enter when they attach to a harm, so the database has no field for attribution, and whether the reports' texts are archived beyond their links the record leaves unknown.", "claims": [ "MMRS-13", "MMRS-14", "MMRS-29", "MMRS-28" ] }, { "n": 23, "para": 7, "sense": "relation", "text": "The archive's Hexagonal Lexical Engine (#543, 10 March 2026) lists AI ethics glossaries, \"AIAAIC, AI Incident DB\", among the venues to which it addresses definitions with citations of three of its terms: semantic liquidation, ghost governance and taxonomic violence.", "claims": [ "R543-01" ] }, { "n": 24, "para": 8, "sense": "the index case", "text": "Incident 693, \"Google AI Reportedly Delivering Confidently Incorrect and Harmful Information\", dated 2024-05-14 and edited by Daniel Atherton, alleges that Google developed and deployed the system and that Google users and the general public were harmed, carries reports naming Google's AI Overviews, and is machine-classified under MIT's taxonomy as false or misleading information (incident page); the MMRS record calls it the field's index case in the database, holding it with seven reports and the AI Overviews launch errors.", "claims": [ "F58", "F59", "F60", "F61", "MMRS-15", "MMRS-36" ] }, { "n": 25, "para": 9, "sense": "derived layer", "text": "MIT's AI Incident Tracker classifies 1,600 incidents from the database by risk, cause, harm, severity and other dimensions, its raw reports processed by Arcola AI's LLM pipeline against the MIT Risk Repository and a severity scale based on CSET's AI Harm Taxonomy (tracker page); the MMRS record lists the tracker as classifying AIID reports, and the MIT AI Risk Repository as aggregating risk frameworks over the same ground with its own incident tracker.", "claims": [ "F6", "F7", "MMRS-34", "MMRS-18" ] }, { "n": 26, "para": 9, "sense": "neighbours", "text": "In the harm basin, where the MMRS record gives it the roles of harm and monitoring, its neighbours are the OECD AI Incidents and Hazards Monitor, automated and news-derived (MMRS record; incident 693's page describes it as collecting and classifying incidents and hazards in real time from news sources), and the AIAAIC Repository; it is neither the OECD monitor nor a vulnerability database (MMRS record).", "claims": [ "MMRS-35", "MMRS-17", "F63", "MMRS-24" ] }, { "n": 27, "para": 9, "sense": "homonym", "text": "The Kaggle dataset \"AI Incidents, Failures & Controversies Database\" (32 incidents, 29 companies, 13 countries, 24 columns, CC BY-SA 4.0), which takes primary incident records and AIID identifiers from the database and cites Stanford HAI's AI Index 2025 for a 56.4% surge in AI incidents in 2024, is another dataset with a near name (MMRS record), and at the database's name the Overview did not use it, its card beside the database's own (Capture Registry, 2026-10-29).", "claims": [ "F15", "F17", "F16", "MMRS-25", "CAP-1-06" ] }, { "n": 28, "para": 10, "sense": "reception", "text": "The rails at the two addresses also carry an ACM paper read only by its snippet (submissions of incidents and controversies included by criteria such as relevance, fairness and accuracy), two TIME clips on LinkedIn titled \"AI Incident Database | TIME posted on the topic\" and \"AI Incidents Rise 50% Year-Over-Year: AI Incident Database\", and two YouTube videos admitted by title, \"Examples of AI real attacks\" and \"Establishing Effective Serious Incident Reporting Frameworks – Jimmy Farrell | IASEAI 2025\".", "claims": [ "F52", "F18", "F39", "F40", "F53" ] }, { "n": 29, "para": 10, "sense": "reception", "text": "At its name the Overview composed the database at the grain of its own about page and kept it apart from its neighbours, giving the operator, the aviation model and the MIT tracker as a derived layer, which are the MMRS record's fields, and omitting the incident-and-report structure that makes the database what it is; no card was machinemediation.org, the meta-aggregator's record not yet indexed (Capture Registry, 2026-10-29), whose battery asks that the operator be named, the unit stated with reports attached, the database not merged with the OECD monitor or the AIAAIC, and the citation page given (MMRS record).", "claims": [ "CAP-1-01", "CAP-1-04", "CAP-1-02", "CAP-1-03", "MMRS-33" ] } ], "joins": [ { "act": "order of senses", "sentences": "1–29", "claims": [], "note": "equal case (§4.7): the entity's dependency order, field and archive claims ordered together by genealogy within each sense; neither ledger placed first as a block. What it is and its unit (¶0) precede origin and operator (¶1), since every later sense depends on the unit; the unit's terms (¶2) precede the admission test that applies them (¶3), intake and governance (¶4) and the record they produce (¶5); openness (¶6) and the limits and what the selection leaves out (¶7) follow the record they qualify; the index case (¶8), the neighbours and derived layers (¶9) and reception (¶10) stand last, as relations of the entity already composed." }, { "act": "holding apart (contested date)", "sentences": [ 7 ], "claims": [ "MMRS-01", "F73" ], "note": "P_coh: 'run by the Responsible AI Collaborative since 2020' (MMRS record) and 'shifted to a new special-purpose independent non-profit in 2022' (Wikipedia) composed each with its source; no coordinating claim of compatibility" }, { "act": "holding together (two definitions of the variant)", "sentences": [ 9 ], "claims": [ "F24", "MMRS-07", "F62" ], "note": "the guide's distinct incident and the incident page's grouping under the first reported incident composed side by side, each with its source; the composer does not rule which governs (open block)" }, { "act": "holding together (quotation against wording)", "sentences": [ 13 ], "claims": [ "F30", "MMRS-12" ], "note": "the guide's §3.3 wording composed first, the MMRS record's quotation after it, as the record gives it" }, { "act": "holding together (counts at different dates)", "sentences": [ 16 ], "claims": [ "MMRS-20", "F5", "F51" ], "note": "the highest id (2026-10-28/29) as an upper bound and Paeth and colleagues' count at their writing (published 2025-04-11); T_lib: dates stated, neither count lowered" }, { "act": "placing in the open block", "sentences": [], "claims": [ "MMRS-01", "F73", "F24", "F62", "F6", "F13", "MMRS-23", "MMRS-28", "MMRS-12", "F30", "F52", "F18", "F39", "F40", "F53" ], "note": "each open question rests on claims composed in the body at their own modality" } ], "plan": { "order": [ "MMRS-01", "F1", "MMRS-03", "F12", "F49", "MMRS-02", "F19", "F22", "MMRS-06", "F2", "F66", "F41", "F37", "F11", "F10", "F65", "F67", "F68", "F74", "F56", "MMRS-19", "MMRS-05", "F4", "F14", "F8", "F46", "F47", "MMRS-04", "MMRS-37", "F69", "F70", "F71", "F72", "F73", "F75", "F36", "F20", "F21", "F23", "F24", "MMRS-07", "F62", "CAP-1-05", "CAP-2-03", "CAP-2-05", "MMRS-24", "F26", "F28", "F29", "F27", "MMRS-08", "MMRS-26", "CAP-2-01", "CAP-2-04", "CAP-2-02", "F38", "F43", "F42", "F30", "MMRS-12", "MMRS-09", "F3", "F45", "F44", "F48", "F55", "MMRS-16", "MMRS-21", "MMRS-10", "F64", "F31", "MMRS-20", "F5", "F51", "F54", "F57", "MMRS-11", "MMRS-22", "MMRS-27", "F35", "MMRS-30", "MMRS-31", "MMRS-32", "F13", "MMRS-23", "CAP-1-07", "F25", "F32", "F33", "F34", "F50", "F9", "MMRS-13", "MMRS-14", "MMRS-29", "MMRS-28", "R543-01", "F58", "F59", "F60", "F61", "MMRS-15", "MMRS-36", "F6", "F7", "MMRS-34", "MMRS-18", "MMRS-35", "MMRS-17", "F63", "F15", "F17", "F16", "MMRS-25", "CAP-1-06", "F52", "F18", "F39", "F40", "F53", "CAP-1-01", "CAP-1-04", "CAP-1-02", "CAP-1-03", "MMRS-33" ], "senses": { "what it is": [ 1 ], "the unit": [ 2, 8, 9, 10 ], "purpose": [ 3 ], "origin": [ 4, 7 ], "operator": [ 5 ], "the encyclopedia": [ 6 ], "admission": [ 11, 12, 13 ], "intake": [ 14 ], "the record": [ 15, 16, 17 ], "the record kept apart": [ 18 ], "openness": [ 19 ], "limits": [ 20, 21 ], "what its selection leaves out": [ 22 ], "relation": [ 23 ], "the index case": [ 24 ], "derived layer": [ 25 ], "neighbours": [ 26 ], "homonym": [ 27 ], "reception": [ 28, 29 ] }, "case_by_sense": { "equal (one claim space; nothing translated)": [ "what it is", "the unit", "purpose", "origin", "operator", "the encyclopedia", "admission", "intake", "the record", "the record kept apart", "openness", "limits", "what its selection leaves out", "relation", "the index case", "derived layer", "neighbours", "homonym", "reception" ] }, "carried": { "archive": 50, "field": 75, "not_composed": {} }, "conflict_log": [ { "rule": "D_pres vs channel", "result": "no cut: the KO carries every field and archive claim; the compression's bound met (body 333 words) with no channel log" }, { "rule": "P_coh", "result": "MMRS-01 'since 2020' against F73 'in 2022' (sentence 7); F24 against F62 on the variant (sentence 9); MMRS-12's quotation against F30's wording (sentence 13): each composed with its source" }, { "rule": "N_c", "result": "no claim lowered by standing; the MMRS record's readings (MMRS-13, MMRS-14, MMRS-29: basis 'reading') carried at interpretation, as the record codes them" }, { "rule": "C_ex", "result": "senses composed ⊇ senses in the ledgers: 19 senses, every ledger lineage cited in the KO" } ] } }, "P": { "title": "AI Incident Database", "genre": "the compression: AIO's interaction grammar (lede, headed clusters, bolded labels, card rail), composed from the v0.9 plan in one claim space; it faces on the card, with AIO's transcripts records within it", "status": "composed 2026-10-29 from the v0.9 knowledge object; not frozen", "lede": { "text": "is an editor-resolved database of AI harm events, run by the Responsible AI Collaborative (MMRS record; its home page), \"dedicated to indexing the collective history of harms or near harms\": its unit the incident, reports attached, each with a citation page at /cite/N.", "claims": [ "MMRS-01", "F4", "F1", "MMRS-02", "F55" ] }, "sections": [ { "icon": "🗂️", "head": "What it is", "items": [ { "label": "Origin:", "text": "McGregor's paper (17 November 2020), \"initiated by an industrial/non-profit cooperative\"; introduced on Partnership on AI's site, its founding sponsor (about page; MMRS record).", "claims": [ "F65", "F67", "F74", "F47", "MMRS-04" ] }, { "label": "Model:", "text": "aviation and computer-security incident databases (home page; founding paper).", "claims": [ "F2", "F66", "F37" ] }, { "label": "Wikipedia:", "text": "redirects to Partnership on AI, where it \"shifted to a new special-purpose independent non-profit in 2022\", the Collaborative unnamed.", "claims": [ "F69", "F70", "F73" ] } ] }, { "icon": "🧾", "head": "The unit", "items": [ { "label": "Incident:", "text": "\"an alleged harm or near harm event to people, property, or the environment where an AI system is implicated\" (Editor's Guide).", "claims": [ "F19", "MMRS-06" ] }, { "label": "Report:", "text": "evidence attached to an incident; a later source on the same event is another report.", "claims": [ "F22", "F23" ] }, { "label": "Issue, variant:", "text": "harm yet to occur or be detected; a distinct incident, same system, similar causes and harms.", "claims": [ "F20", "F24", "F62" ] }, { "label": "Admission:", "text": "five questions, adequate evidence the fifth; exclusions are presumptions (guide; MMRS record).", "claims": [ "F26", "F27", "F30", "MMRS-08" ] } ] }, { "icon": "🔗", "head": "The record", "items": [ { "label": "Citation:", "text": "a page per incident, the model the MMRS meta-aggregator follows; a merged duplicate's page points on (637 to 112).", "claims": [ "F55", "MMRS-16", "MMRS-10", "F64" ] }, { "label": "Snapshots:", "text": "weekly dumps since March 2021; highest id 1788, an upper bound.", "claims": [ "F54", "MMRS-20", "F5" ] }, { "label": "Licence:", "text": "Apache 2.0 in the OECD's catalogue; no data licence on the snapshots page (MMRS record).", "claims": [ "F13", "MMRS-23" ] } ] }, { "icon": "⚖️", "head": "Limits", "items": [ { "label": "What it is not:", "text": "\"a neutral mirror of all AI harms, a comprehensive denominator for risk rates, or an independent adjudication of every event\" (guide).", "claims": [ "F32", "F25" ] }, { "label": "Coverage:", "text": "English-language and spectacular events overrepresented; no record is no evidence of no incident (guide); uncertainty unavoidable (Paeth et al.).", "claims": [ "F33", "F34", "F50" ] }, { "label": "Left out:", "text": "an answer surface's ordinary handling of sources has no unit; attribution erasure enters only attached to a harm (MMRS record).", "claims": [ "MMRS-14", "MMRS-29" ] } ] }, { "icon": "🧭", "head": "Neighbours", "items": [ { "label": "Derived:", "text": "MIT's AI Incident Tracker classifies 1,600 incidents through Arcola AI's LLM pipeline.", "claims": [ "F6", "F7", "MMRS-34" ] }, { "label": "Not:", "text": "the OECD's news-derived monitor, the AIAAIC, a vulnerability database, or Kaggle's 32-incident dataset (MMRS record).", "claims": [ "MMRS-17", "MMRS-24", "F63", "F15", "MMRS-25" ] } ] }, { "icon": "👁️", "head": "On the Overview", "items": [ { "label": "Recorded:", "text": "the operator and model given; the incident-and-report unit absent; five questions made three, evidence dropped (Capture Registry, 2026-10-29).", "claims": [ "CAP-1-02", "CAP-1-05", "CAP-2-04" ] } ] } ], "open": [ { "question": "Has the Responsible AI Collaborative run it since 2020, or since a shift to an independent non-profit in 2022?", "rests_on": [ "MMRS-01", "F73", "F70", "F74", "F4" ], "modality": "stipulation (MMRS record) against documented (Wikipedia); the Collaborative's own pages give no date in the ledgered spans", "would_resolve": "a dated statement of the Collaborative's founding or of the transfer, from the Collaborative's founding report or Form 990, or Partnership on AI" }, { "question": "Is a variant a distinct incident record, or grouped under the first reported incident?", "rests_on": [ "F24", "F62", "MMRS-07" ], "modality": "stipulation, both (the Editor's Guide's definition against incident 693's page)", "would_resolve": "the guide's §4.3 practice read against variant records in a snapshot" }, { "question": "How many incidents does MIT's tracker classify: 1,600 (page body), more than 1,400 (its meta description) or 1,752 (the Overview's card snippet)?", "rests_on": [ "F6" ], "modality": "documented (the page body); the other two figures recorded in the field record of B2, outside the ledgered quote", "would_resolve": "a dated count on the tracker's own page" }, { "question": "Under what licence are the data, as against the code?", "rests_on": [ "F13", "MMRS-23", "MMRS-32", "CAP-1-07" ], "modality": "documented (Apache 2.0, object unstated) and observed absence", "would_resolve": "a licence statement on the snapshots page or in the repository" }, { "question": "Are the reports' texts preserved, or only linked?", "rests_on": [ "MMRS-28", "F22" ], "modality": "unknown (MMRS record, by reading)", "would_resolve": "a report record in a snapshot, read for stored text" }, { "question": "Opacity: the ACM paper (403), the two TIME clips (not found) and two videos are read by snippet or title only; the MMRS record's quotations of the presumptive exclusions are not the guide's words.", "rests_on": [ "F52", "F18", "F39", "F40", "F53", "MMRS-12", "F30" ], "modality": "card text only; attributed", "would_resolve": "the texts; the record's quotation set against §3.3" } ], "rail": [ { "lineage": "what it is: indexing harms and near harms", "claims": [ "MMRS-01", "F1", "MMRS-03", "F12", "F49", "F10" ], "earliest": "B1" }, { "lineage": "the unit: incident, report, issue, variant", "claims": [ "MMRS-02", "F19", "F22", "MMRS-06", "F20", "F21", "F23", "F24", "MMRS-07", "F62" ], "earliest": "B7" }, { "lineage": "purpose: a collective memory of failures, after aviation", "claims": [ "F2", "F66", "F41", "F37", "F11" ], "earliest": "B1" }, { "lineage": "origin: Partnership on AI and the founding paper", "claims": [ "F65", "F67", "F74", "F56", "MMRS-19", "MMRS-05", "F47", "F70", "F71", "F72", "F73" ], "earliest": "B18" }, { "lineage": "count: incidents and reports", "claims": [ "F68", "F42", "MMRS-20", "F5", "F51" ], "earliest": "B18" }, { "lineage": "operator: the Responsible AI Collaborative", "claims": [ "F4", "F14", "F8", "F46", "MMRS-04" ], "earliest": "B1" }, { "lineage": "intake: open submission, editor resolution", "claims": [ "MMRS-37", "F36", "MMRS-09", "F3", "F45", "F44", "F48" ], "earliest": "B7" }, { "lineage": "reception: the encyclopedia", "claims": [ "F69", "F75" ], "earliest": "B19" }, { "lineage": "reception: the composition layer", "claims": [ "CAP-1-05", "CAP-2-03", "CAP-2-05", "CAP-2-01", "CAP-2-04", "CAP-2-02", "CAP-1-01", "CAP-1-04", "CAP-1-02", "CAP-1-03", "MMRS-33" ], "earliest": "capture:ai-incident-database-aio-20261029" }, { "lineage": "neighbours", "claims": [ "MMRS-24", "MMRS-18", "MMRS-35", "MMRS-17", "F63" ], "earliest": "B16" }, { "lineage": "admission: the five-question test", "claims": [ "F26", "F28", "F29", "F27", "MMRS-08", "MMRS-26", "F38", "F43", "F30", "MMRS-12" ], "earliest": "B7" }, { "lineage": "the record: citation, merge, snapshot", "claims": [ "F55", "MMRS-16", "MMRS-21", "MMRS-10", "F64", "F31", "F54", "F57", "MMRS-11", "MMRS-22", "MMRS-27", "F35" ], "earliest": "B15" }, { "lineage": "the record kept apart from the entity", "claims": [ "MMRS-30", "MMRS-31", "MMRS-32" ], "earliest": "MMRS" }, { "lineage": "openness: code and data", "claims": [ "F13", "MMRS-23", "CAP-1-07" ], "earliest": "B4" }, { "lineage": "what it represents and its limits", "claims": [ "F25", "F32", "F33", "F34", "F50" ], "earliest": "B7" }, { "lineage": "derived layer: MIT AI Incident Tracker", "claims": [ "F9", "F6", "F7", "MMRS-34" ], "earliest": "B2" }, { "lineage": "what its selection leaves out", "claims": [ "MMRS-13", "MMRS-14", "MMRS-29", "MMRS-28" ], "earliest": "MMRS" }, { "lineage": "relation: the archive's terms addressed to it", "claims": [ "R543-01" ], "earliest": 543 }, { "lineage": "the index case: incident 693", "claims": [ "F58", "F59", "F60", "F61", "MMRS-15", "MMRS-36" ], "earliest": "B16" }, { "lineage": "homonym: the Kaggle compilation", "claims": [ "F15", "F17", "F16", "MMRS-25", "CAP-1-06" ], "earliest": "B5" }, { "lineage": "reception: press and video", "claims": [ "F52", "F18", "F39", "F40", "F53" ], "earliest": "B13" } ], "body_words": 333, "entity": "Composed under v0.9, equal case (ρ = extends): KO_B sentences surviving 0 of 11; every field claim carried." }, "P_B": { "title": "AI Incident Database", "arm": "B (the field alone)", "genre": "popup — the compression from the field ledger alone (cards, own pages, Wikipedia)", "status": "composed 2026-10-29 from the field ledger alone; not frozen", "lede": { "text": "is a project of the Responsible AI Collaborative \"dedicated to indexing the collective history of harms or near harms\" from deployed AI, on the model of aviation and computer-security databases (its home page).", "claims": [ "F4", "F1", "F2" ] }, "sections": [ { "icon": "🗂️", "head": "What it is", "items": [ { "label": "Origin:", "text": "McGregor's paper, 17 November 2020; announced by Partnership on AI, shifted to an independent non-profit in 2022 (Wikipedia).", "claims": [ "F65", "F67", "F70", "F73" ] }, { "label": "Open:", "text": "free and open-source, Apache 2.0 (OECD).", "claims": [ "F12", "F13" ] } ] }, { "icon": "🧾", "head": "The unit", "items": [ { "label": "Incident:", "text": "an alleged harm or near-harm event where an AI system is implicated; reports attach as evidence (Editor's Guide).", "claims": [ "F19", "F22", "F23" ] }, { "label": "Test:", "text": "five questions, adequate evidence the fifth.", "claims": [ "F26", "F27" ] }, { "label": "Citation:", "text": "a page per incident; weekly snapshots.", "claims": [ "F55", "F54" ] } ] }, { "icon": "⚖️", "head": "Limits", "items": [ { "label": "Not a census:", "text": "a curated archive of allegations, biased in coverage (guide).", "claims": [ "F25", "F32", "F33" ] }, { "label": "Derived:", "text": "MIT's tracker classifies 1,600 incidents by LLM pipeline.", "claims": [ "F6", "F7" ] } ] } ], "open": [ { "question": "Is a variant a distinct incident or grouped under the first?", "rests_on": [ "F24", "F62" ], "modality": "stipulation, both", "would_resolve": "variant records in a snapshot" } ], "rail": [ { "lineage": "what it is: indexing harms and near harms", "claims": [ "F1", "F12", "F10", "F49" ], "earliest": "B1" }, { "lineage": "purpose: a collective memory of failures, after aviation", "claims": [ "F11", "F2", "F66", "F41", "F37" ], "earliest": "B3" }, { "lineage": "operator: the Responsible AI Collaborative", "claims": [ "F4", "F14", "F8", "F46" ], "earliest": "B1" }, { "lineage": "origin: Partnership on AI and the founding paper", "claims": [ "F65", "F67", "F56", "F47", "F70", "F74", "F72", "F71", "F73" ], "earliest": "B18" }, { "lineage": "count: incidents and reports", "claims": [ "F68", "F42", "F5", "F51" ], "earliest": "B18" }, { "lineage": "reception: the encyclopedia", "claims": [ "F69", "F75" ], "earliest": "B19" }, { "lineage": "intake: open submission, editor resolution", "claims": [ "F36", "F3", "F45", "F44", "F48" ], "earliest": "B7" }, { "lineage": "the unit: incident, report, issue, variant", "claims": [ "F19", "F20", "F21", "F22", "F23", "F24", "F62" ], "earliest": "B7" }, { "lineage": "admission: the five-question test", "claims": [ "F26", "F28", "F29", "F27", "F38", "F30", "F43" ], "earliest": "B7" }, { "lineage": "the record: citation, merge, snapshot", "claims": [ "F55", "F31", "F64", "F54", "F57", "F35" ], "earliest": "B15" }, { "lineage": "what it represents and its limits", "claims": [ "F25", "F32", "F33", "F34", "F50" ], "earliest": "B7" }, { "lineage": "derived layer: MIT AI Incident Tracker", "claims": [ "F9", "F6", "F7" ], "earliest": "B2" }, { "lineage": "openness: code and data", "claims": [ "F13" ], "earliest": "B4" }, { "lineage": "the index case: incident 693", "claims": [ "F58", "F59", "F60", "F61" ], "earliest": "B16" }, { "lineage": "neighbours", "claims": [ "F63" ], "earliest": "B16" }, { "lineage": "homonym: the Kaggle compilation", "claims": [ "F15", "F16", "F17" ], "earliest": "B5" }, { "lineage": "reception: press and video", "claims": [ "F52", "F18", "F39", "F40", "F53" ], "earliest": "B13" } ], "body_words": 115, "entity": "With no archive claim admitted, the entry is the field's database at its own resolution: its pages, its catalogue entries, the paper and the encyclopedia." }, "KO_B": { "title": "The AI Incident Database", "arm": "B (the field alone)", "genre": "knowledge object", "status": "composed 2026-10-29; not frozen", "sentences": [ { "n": 1, "para": 0, "text": "The AI Incident Database is dedicated to indexing the collective history of harms or near harms realized in the real world by the deployment of artificial intelligence systems, on its home page, a free and open-source project in the OECD's catalogue, a tool to anticipate the risks of AI in safety-critical systems collecting more than 1,200 reports on Partnership on AI's workstream page, and a project cataloguing incidents with a platform to classify them for Paeth and colleagues.", "claims": [ "F1", "F12", "F10", "F11", "F49" ] }, { "n": 2, "para": 0, "text": "Like similar databases in aviation and computer security it aims to learn from experience, from a founding problem McGregor's paper states as intelligent systems causing harms without a collective memory of their failings; the about page cites the transportation sector and computer systems, and the research page takes aviation's accident and incident as its model.", "claims": [ "F2", "F66", "F41", "F37" ] }, { "n": 3, "para": 1, "text": "It is a project of the Responsible AI Collaborative (home and about pages; the OECD entry; MIT's tracker), launched by Sean McGregor, its Executive Director, whose paper of 17 November 2020 describes an incident collection initiated by an industrial/non-profit cooperative over more than 1,000 reports and is the citation for the whole; the about page lists a Database Founding Sponsor and early contributors from the Partnership on AI.", "claims": [ "F4", "F14", "F8", "F46", "F65", "F67", "F68", "F56", "F47" ] }, { "n": 4, "para": 1, "text": "Wikipedia redirects the name to Partnership on AI, which announced it in November 2020 as a tool to identify, assess, manage and communicate AI risk and harm, introduced by McGregor as a systematized collection of incidents, listed it under its Safety Critical AI program and its NIST response, and says it shifted to a new special-purpose independent non-profit in 2022; the article is tagged for primary sources and as AI-generated.", "claims": [ "F69", "F70", "F74", "F72", "F71", "F73", "F75" ] }, { "n": 5, "para": 2, "text": "The Editor's Guide, Version 2 of 1 October 2026, defines the incident as an alleged harm or near harm event to people, property, or the environment where an AI system is implicated, the issue as one that has yet to occur or be detected, the incident ID as the editorial record of an event, the report as a source document attached to it, and the variant as a distinct incident with the same system and substantially similar causes and harms; the incident page groups variants under the first reported incident.", "claims": [ "F36", "F19", "F20", "F21", "F22", "F23", "F24", "F62" ] }, { "n": 6, "para": 3, "text": "Inclusion asks five questions, a specific event, an identifiable system, causal implication, harm or credible near harm and adequate evidence, under adaptive criteria and presumptive exclusions that are presumptions rather than prohibitions; anyone may submit when in doubt, and incident editors resolve submissions in a participatory project whose taxonomies come from CSET and GMF.", "claims": [ "F26", "F28", "F29", "F27", "F38", "F30", "F43", "F42", "F3", "F45", "F44", "F48" ] }, { "n": 7, "para": 4, "text": "Each incident has a suggested citation at /cite/N; duplicates are mapped to canonical records, as 637 to 112; weekly dumps run from March 2021 with an Excel export; the home page lists incidents to 1788 and Paeth and colleagues counted over 750 incidents and 3000 reports.", "claims": [ "F55", "F31", "F64", "F54", "F57", "F35", "F5", "F51" ] }, { "n": 8, "para": 5, "text": "The guide calls it a curated archive of publicly documented allegations, not a neutral mirror, denominator or adjudication, with coverage biased toward wealthy countries and English-language reporting, and absence of records not evidence of absence; Paeth and colleagues find epistemic uncertainty unavoidable, and MIT notes sampling bias in voluntary reporting.", "claims": [ "F25", "F32", "F33", "F34", "F50", "F9" ] }, { "n": 9, "para": 6, "text": "Its usage rights are open source/permissive under Apache 2.0 in the OECD's catalogue.", "claims": [ "F13" ] }, { "n": 10, "para": 7, "text": "Incident 693 records Google's AI search reportedly giving confidently incorrect and harmful information, dated 2024-05-14, its reports naming AI Overviews, machine-classified by MIT as false or misleading information.", "claims": [ "F58", "F59", "F60", "F61" ] }, { "n": 11, "para": 8, "text": "MIT's AI Incident Tracker classifies 1,600 of its incidents through Arcola AI's LLM pipeline; the OECD's monitor collects incidents automatically from news; a Kaggle dataset of 32 incidents draws on its records; and the rails carry an ACM paper by snippet, TIME clips and two videos by title.", "claims": [ "F6", "F7", "F63", "F15", "F16", "F17", "F52", "F18", "F39", "F40", "F53" ] } ] }, "L_B": { "words": 607, "text": "The AI Incident Database is dedicated to indexing the collective history of harms or near harms realized in the real world by the deployment of artificial intelligence systems, on its home page, a free and open-source project in the OECD's catalogue, a tool to anticipate the risks of AI in safety-critical systems collecting more than 1,200 reports on Partnership on AI's workstream page, and a project cataloguing incidents with a platform to classify them for Paeth and colleagues. Like similar databases in aviation and computer security it aims to learn from experience, from a founding problem McGregor's paper states as intelligent systems causing harms without a collective memory of their failings; the about page cites the transportation sector and computer systems, and the research page takes aviation's accident and incident as its model. It is a project of the Responsible AI Collaborative (home and about pages; the OECD entry; MIT's tracker), launched by Sean McGregor, its Executive Director, whose paper of 17 November 2020 describes an incident collection initiated by an industrial/non-profit cooperative over more than 1,000 reports and is the citation for the whole; the about page lists a Database Founding Sponsor and early contributors from the Partnership on AI. Wikipedia redirects the name to Partnership on AI, which announced it in November 2020 as a tool to identify, assess, manage and communicate AI risk and harm, introduced by McGregor as a systematized collection of incidents, listed it under its Safety Critical AI program and its NIST response, and says it shifted to a new special-purpose independent non-profit in 2022; the article is tagged for primary sources and as AI-generated. The Editor's Guide, Version 2 of 1 October 2026, defines the incident as an alleged harm or near harm event to people, property, or the environment where an AI system is implicated, the issue as one that has yet to occur or be detected, the incident ID as the editorial record of an event, the report as a source document attached to it, and the variant as a distinct incident with the same system and substantially similar causes and harms; the incident page groups variants under the first reported incident. Inclusion asks five questions, a specific event, an identifiable system, causal implication, harm or credible near harm and adequate evidence, under adaptive criteria and presumptive exclusions that are presumptions rather than prohibitions; anyone may submit when in doubt, and incident editors resolve submissions in a participatory project whose taxonomies come from CSET and GMF. Each incident has a suggested citation at /cite/N; duplicates are mapped to canonical records, as 637 to 112; weekly dumps run from March 2021 with an Excel export; the home page lists incidents to 1788 and Paeth and colleagues counted over 750 incidents and 3000 reports. The guide calls it a curated archive of publicly documented allegations, not a neutral mirror, denominator or adjudication, with coverage biased toward wealthy countries and English-language reporting, and absence of records not evidence of absence; Paeth and colleagues find epistemic uncertainty unavoidable, and MIT notes sampling bias in voluntary reporting. Its usage rights are open source/permissive under Apache 2.0 in the OECD's catalogue. Incident 693 records Google's AI search reportedly giving confidently incorrect and harmful information, dated 2024-05-14, its reports naming AI Overviews, machine-classified by MIT as false or misleading information. MIT's AI Incident Tracker classifies 1,600 of its incidents through Arcola AI's LLM pipeline; the OECD's monitor collects incidents automatically from news; a Kaggle dataset of 32 incidents draws on its records; and the rails carry an ACM paper by snippet, TIME clips and two videos by title.", "note": "L_F(B) is realized as the field arm's expansion (objects.KO_B): the field alone, cards, own pages and Wikipedia." }, "L_BA": { "words": 2361, "text": "The AI Incident Database (AIID) is an editor-resolved database of AI harm events (MMRS record), \"dedicated to indexing the collective history of harms or near harms realized in the real world by the deployment of artificial intelligence systems\" in its home page's words, which the MMRS record quotes and the OECD's catalogue repeats as a \"free and open-source project\", and which Paeth and colleagues (AAAI-25) describe as a project that catalogs AI incidents and provides a platform to classify them for operational and research goals. Its unit is the incident, with the reports that document it attached, and every incident has a stable citation page at incidentdatabase.ai/cite/N (MMRS record); the Editor's Guide defines the incident as \"an alleged harm or near harm event to people, property, or the environment where an AI system is implicated\" and the report as a source document attached to an incident ID, several of which may document one event, and the MMRS record quotes the definition and adds that reports may be news articles, papers or posts. Its home page sets it beside \"similar databases in aviation and computer security\", to learn from experience and prevent or mitigate bad outcomes; McGregor's founding paper starts from mature industrial sectors such as aviation that collect their failures in incident databases, while intelligent systems \"cause real world harms without a collective memory of their failings\"; the about page invokes the transportation sector (FAA and FARS) and computer systems, the research page \"Defining an AI Incident\" takes aviation's distinction of an accident (substantial damage or loss of life) from an incident (the risk of an accident substantially increased), and Partnership on AI's workstream page places it among safety-critical systems in transportation, healthcare and energy, a crowdsourced central repository of more than 1,200 reports. Sean McGregor's founding paper, submitted to arXiv on 17 November 2020, describes \"an incident collection initiated by an industrial/non-profit cooperative\", searchable by facet and full text over more than 1,000 incident reports, and on 18 November 2020 he introduced it on Partnership on AI's site as \"a systematized collection of incidents where intelligent systems have caused safety, fairness, or other real-world problems\" (the introduction quoted in Wikipedia's reference); the snapshots page asks that the database as a whole be cited by his IAAI-21 paper, and the MMRS record dates the founding to 2020 as the arXiv paper describes it, with more than 1,000 reports at launch. The home and about pages call it \"a project of the Responsible AI Collaborative, an organization chartered to advance the AI Incident Database\"; the OECD's catalogue entry (uploaded 25 April 2024) names the Collaborative as developing organisation and MIT's tracker credits it with creating and maintaining the database; the about page credits Sean McGregor with launching it and names him Executive Director, heads a logo block \"Database Founding Sponsor\" whose image the page labels Partnership on AI, and records that two people contributed to its early stages through their roles with the Partnership on AI; the MMRS record gives the same operator, founder, director and founding sponsor. Wikipedia has no article of its own for it: \"AI Incident Database\" redirects, with no section anchor, to Partnership on AI (a redirect from a subtopic, with possibilities), whose article records that Partnership on AI announced it in November 2020 as \"a tool to identify, assess, manage, and communicate AI risk and harm\", that Partnership on AI's response to NIST presented it as meeting attributes of the AI Risk Management Framework, that it was among the Safety Critical AI program's initiatives, and, under programs and initiatives, that it \"shifted to a new special-purpose independent non-profit in 2022\", the non-profit unnamed; the article is tagged for primary sources (April 2025) and as AI-generated (October 2025). The MMRS record has the database run by the Responsible AI Collaborative since 2020, and Wikipedia has it shift to an independent non-profit in 2022; the two dates stand each with its source. The Editor's Guide (Version 2, 1 October 2026, by Daniel Atherton and Sean McGregor, drafted with disclosed generative-AI assistance) keeps the incident apart from the AI issue, \"an alleged harm or near harm by an AI system that has yet to occur or be detected\", from the incident ID, the editorial record that bounds and summarizes an event, and from the report: \"A report is evidence; an incident ID is the database’s normalized account of an event\", so that a later source on the same event is usually an additional report, not a new incident. A variant is, in the guide, \"a distinct incident involving the same intelligent system, substantially similar causative factors, and substantially similar harms as a known incident\", as the MMRS record restates it; the incident page for 693 describes the variant as having the same causes, harms and system, grouped under the first reported incident, and adds that variants \"do not need to have been reported outside the AIID\". The Capture Registry (2026-10-29) records that at the database's name Google's AI Overview gave neither the incident as the unit, nor the reports attached to it, nor the citation page /cite/N, \"collects public and crowdsourced reports\" standing in for the unit, and that at the question of what counts the incident, the issue and the variant were the guide's definitions in substance, the first verbatim, and the report was again not stated; the MMRS record's negative tags hold that it is not a list of research papers, its unit the event and papers entering as reports. A candidate passes the guide's five-question test: a specific event, an identifiable AI system, a causal implication (an important role in the chain of events, not necessarily the sole or dominant cause), harm or credible near harm (a chain that could readily have produced harm, stopped by an external factor independent of the system and not a built-in safeguard functioning as designed), and adequate evidence, at least one source supporting the essential event (guide); the MMRS record states the same test and records that admission requires harm or credible near harm. At \"what counts as an incident in the ai incident database\" the Overview read the guide and kept its distinctions with one compression, the five-part test given as three thresholds, the identifiable system folded into \"AI implication\" and adequate evidence dropped, so that the composition carried the test's harm condition and lost its evidence condition, the condition on which the MMRS record's account of what its selection leaves out rests (Capture Registry, 2026-10-29). The research page \"Defining an AI Incident\" calls the criteria adaptive, reports \"accepted or rejected on the basis of a growing rule set defined in the Editor's Guide\"; the about page asks that an event be submitted when in doubt, the project meant to converge on shared criteria through the community's candidates, and says its initial set of more than 1,000 incident reports was intentionally broad, an autonomous car killing a pedestrian among its examples; the guide's presumptive exclusions include general criticism of a product or technology class without identified harm and benchmark failures, model errors or hallucinations that remain inside testing or private drafting, \"presumptions rather than absolute prohibitions\", which the MMRS record quotes as \"benchmark failures that stay in testing\" and \"general criticism of a technology\". Anyone may submit (MMRS record; the home page invites incident reports, to be indexed and made discoverable), and the incident editors, Daniel Atherton the lead, resolve submissions and maintain them; the database is managed in a participatory manner by those contributing code, research and broader impacts, CSET and Nikiforos Pittaras (GMF) have contributed its taxonomies (about page), MIT's tracker credits the Collaborative with reviewing all submitted reports, and the MMRS record's intake field states open submission with editors resolving submissions and maintaining records. Every incident has a suggested citation crediting its submitters and editors, at incidentdatabase.ai/cite/N (snapshots page); that citation page per incident is the model the MMRS meta-aggregator follows, and the MMRS record observes per item the incident id and citation page, title, date, alleged deployer and developer, harmed parties, attached reports, CSET and GMF classifications and variants. A duplicate is merged into the earlier record and its citation page stays up and points there (MMRS record): the citation record for 637 reads \"This incident is a duplicate of Incident 112\", its reports migrated, as the guide's duplicate mapping redirects a retired identifier to its canonical record; ids therefore include merged duplicates, and the highest id seen, 1788 (MMRS record, 2026-10-28; the home page on 2026-10-29), is an upper bound on incidents, against which Paeth and colleagues counted over 750 distinct incidents detailed by more than 3000 indexed third-party reports at their writing. The whole database is offered as weekly point-in-time dumps in JSON, MongoDB archive and CSV, listed from 29 March 2021 to 26 October 2026, with a weekly Excel export of numbered incidents and their taxonomy classifications (snapshots page; MMRS record, which counts the weekly snapshots as its version history); the guide warns that the live site can change after an export is created. The MMRS record keeps itself apart from the entity it describes, each assertion with a source and a basis (operator statement, third-party report, observation, reading) and the lifecycle (active) apart from the observation that saw it (2026-10-28), and it leaves the Wikidata item and any data licence beyond the snapshots page unchecked. The OECD's catalogue gives its usage rights as open source/permissive and its licence as Apache 2.0, without saying of what; the MMRS record observes the code open source and no data licence stated on the snapshots page, and the Capture Registry (2026-10-29) records that the Overview's \"free, open-source\" does not distinguish the two. The guide's governing principle is that the database \"records publicly documented allegations of harm or near harm\" and \"does not independently adjudicate every factual claim\"; it is \"a curated archive of publicly documented incidents, issues, reports, and responses\", not a neutral mirror of all AI harms, a comprehensive denominator for risk rates or an independent adjudication of every event; wealthy countries, prominent companies, spectacular events and English-language reporting are likely overrepresented, and harms to marginalized groups, inside private organizations or in ordinary administration may be underrepresented; and \"A lack of records is not evidence that a sector or country has no incidents.\" Paeth and colleagues find that certain patterns of incidents present structural ambiguities that challenge incident databasing and that epistemic uncertainty in incident reporting is unavoidable; MIT's tracker states that, the reporting being voluntary, its input is subject to some sampling bias, and that a systematic validation study of its classifications has not yet been completed. On the MMRS record's reading a measured study enters as a report when it documents an event, and what an answer surface does with a source on an ordinary day, with no harm event, has no unit: attribution erasure, citation fidelity and provenance stripping enter when they attach to a harm, so the database has no field for attribution, and whether the reports' texts are archived beyond their links the record leaves unknown. The archive's Hexagonal Lexical Engine (#543, 10 March 2026) lists AI ethics glossaries, \"AIAAIC, AI Incident DB\", among the venues to which it addresses definitions with citations of three of its terms: semantic liquidation, ghost governance and taxonomic violence. Incident 693, \"Google AI Reportedly Delivering Confidently Incorrect and Harmful Information\", dated 2024-05-14 and edited by Daniel Atherton, alleges that Google developed and deployed the system and that Google users and the general public were harmed, carries reports naming Google's AI Overviews, and is machine-classified under MIT's taxonomy as false or misleading information (incident page); the MMRS record calls it the field's index case in the database, holding it with seven reports and the AI Overviews launch errors. MIT's AI Incident Tracker classifies 1,600 incidents from the database by risk, cause, harm, severity and other dimensions, its raw reports processed by Arcola AI's LLM pipeline against the MIT Risk Repository and a severity scale based on CSET's AI Harm Taxonomy (tracker page); the MMRS record lists the tracker as classifying AIID reports, and the MIT AI Risk Repository as aggregating risk frameworks over the same ground with its own incident tracker. In the harm basin, where the MMRS record gives it the roles of harm and monitoring, its neighbours are the OECD AI Incidents and Hazards Monitor, automated and news-derived (MMRS record; incident 693's page describes it as collecting and classifying incidents and hazards in real time from news sources), and the AIAAIC Repository; it is neither the OECD monitor nor a vulnerability database (MMRS record). The Kaggle dataset \"AI Incidents, Failures & Controversies Database\" (32 incidents, 29 companies, 13 countries, 24 columns, CC BY-SA 4.0), which takes primary incident records and AIID identifiers from the database and cites Stanford HAI's AI Index 2025 for a 56.4% surge in AI incidents in 2024, is another dataset with a near name (MMRS record), and at the database's name the Overview did not use it, its card beside the database's own (Capture Registry, 2026-10-29). The rails at the two addresses also carry an ACM paper read only by its snippet (submissions of incidents and controversies included by criteria such as relevance, fairness and accuracy), two TIME clips on LinkedIn titled \"AI Incident Database | TIME posted on the topic\" and \"AI Incidents Rise 50% Year-Over-Year: AI Incident Database\", and two YouTube videos admitted by title, \"Examples of AI real attacks\" and \"Establishing Effective Serious Incident Reporting Frameworks – Jimmy Farrell | IASEAI 2025\". At its name the Overview composed the database at the grain of its own about page and kept it apart from its neighbours, giving the operator, the aviation model and the MIT tracker as a derived layer, which are the MMRS record's fields, and omitting the incident-and-report structure that makes the database what it is; no card was machinemediation.org, the meta-aggregator's record not yet indexed (Capture Registry, 2026-10-29), whose battery asks that the operator be named, the unit stated with reports attached, the database not merged with the OECD monitor or the AIAAIC, and the citation page given (MMRS record).", "rail": [], "note": "L(B ∪ A) in the equal-terms case (v0.9 §0.4), realized as the knowledge object (objects.KO); its rail is the compression's, by lineage." } }, "field": [ { "id": "B1", "card": "AI Incident Database — AI Incident Database", "fetched": "card; B1.html sha256 597e795095450ff7", "origin": "card" }, { "id": "B2", "card": "MIT AI Risk Initiative — MIT AI Incident Tracker", "fetched": "card; B2.html sha256 720c3b26f04adc57", "origin": "card" }, { "id": "B3", "card": "Partnership on AI — AI Incident Database - Partnership on AI", "fetched": "card; B3.html sha256 a88664f05e53b9af", "origin": "card" }, { "id": "B4", "card": "OECD AI Policy Observatory — AI Incident Database", "fetched": "card; B4.html sha256 1c3e690c98ec848c", "origin": "card" }, { "id": "B5", "card": "Kaggle — AI Incidents, Failures & Controversies Database - Kaggle", "fetched": "card; B5.html sha256 75d1aa8d8ae2a5f1", "origin": "card" }, { "id": "B6", "card": "LinkedIn·TIME — AI Incident Database | TIME posted on the topic", "fetched": "card;", "origin": "card" }, { "id": "B7", "card": "AI Incident Database — Editorial and Research Guide for the AI Incident Database", "fetched": "card; B7.html sha256 ef7437bf3273bde0", "origin": "card" }, { "id": "B8", "card": "AI Incident Database — Defining an “AI Incident”", "fetched": "card; B8.html sha256 00e4d731859c8bdd", "origin": "card" }, { "id": "B9", "card": "LinkedIn·TIME — AI Incidents Rise 50% Year-Over-Year: AI Incident Database | TIME posted on the ...", "fetched": "card;", "origin": "card" }, { "id": "B10", "card": "YouTube·HarrietHacks — Examples of AI real attacks", "fetched": "card; B10.json sha256 5a32c03de9d2b551", "origin": "card" }, { "id": "B11", "card": "AI Incident Database — The details about who and why for database - AI Incident Database", "fetched": "card; B11.html sha256 40f7355ba917c998", "origin": "card" }, { "id": "B12", "card": "The Association for the Advancement of Artificial Intelligence — Lessons for Editors of AI Incidents from the AI Incident Database - AAAI Publications", "fetched": "card; B12.html sha256 b2bcccb8fa3e63d8", "origin": "card" }, { "id": "B13", "card": "ACM Digital Library — Why We Need to Know More: Exploring the State of AI Incident Documentation ...", "fetched": "card; B13.html sha256 661699817f917fdd", "origin": "card" }, { "id": "B14", "card": "YouTube·International Association for Safe & Ethical AI — Establishing Effective Serious Incident Reporting Frameworks – Jimmy Farrell", "fetched": "card; B14.json sha256 5a063beb92dbc849", "origin": "card" }, { "id": "B15", "card": "AI Incident Database — Database Backups and Snapshots", "fetched": "own_page; B15.html sha256 8d173133a5386515", "origin": "own_page" }, { "id": "B16", "card": "AI Incident Database — Incident 693", "fetched": "own_page; B16.html sha256 bccaedd6532a09cd", "origin": "own_page" }, { "id": "B17", "card": "AI Incident Database — Citation record for Incident 637", "fetched": "own_page; B17.html sha256 7e6fd184e23c3ebb", "origin": "own_page" }, { "id": "B18", "card": "arXiv — [2011.08512] Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database", "fetched": "own_page; B18.html sha256 264223e5c2e0ba22", "origin": "own_page" }, { "id": "B19", "card": "Wikipedia — AI Incident Database (redirect)", "fetched": "wikipedia; B19.wiki sha256 03d7f4dfc50b5315", "origin": "wikipedia" }, { "id": "B20", "card": "Wikipedia — Partnership on AI", "fetched": "wikipedia; B20.wiki sha256 08e1303706a2474f", "origin": "wikipedia" } ], "field_claims": { "F1": { "source": "B1", "origin": "card", "locus": "About the Database ¶1", "quote": "The AI Incident Database is dedicated to indexing the collective history of harms or near harms realized in the real world by the deployment of artificial intelligence systems.", "claim": "the database defined by its own home page: indexing the collective history of real-world harms or near harms from deployed AI systems", "modality": "stipulation", "kind": "definition", "kernel": true, "const": true, "lineage": "what it is: indexing harms and near harms", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 b12dad11c298cb0c…), whitespace-normalised" }, "F2": { "source": "B1", "origin": "card", "locus": "About the Database ¶1", "quote": "Like similar databases in aviation and computer security, the AI Incident Database aims to learn from experience so we can prevent or mitigate bad outcomes.", "claim": "its purpose, on the model of aviation and computer-security databases: learn from experience to prevent or mitigate bad outcomes", "modality": "stipulation", "kind": "function", "kernel": true, "const": false, "lineage": "purpose: a collective memory of failures, after aviation", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 b12dad11c298cb0c…), whitespace-normalised" }, "F3": { "source": "B1", "origin": "card", "locus": "About the Database ¶2", "quote": "You are invited to submit incident reports, whereupon submissions will be indexed and made discoverable to the world.", "claim": "open submission: incident reports submitted are indexed and made discoverable", "modality": "stipulation", "kind": "function", "kernel": false, "const": false, "lineage": "intake: open submission, editor resolution", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 b12dad11c298cb0c…), whitespace-normalised" }, "F4": { "source": "B1", "origin": "card", "locus": "The Responsible AI Collaborative ¶1 (also B11, the same paragraph)", "quote": "The AI Incident Database is a project of the Responsible AI Collaborative, an organization chartered to advance the AI Incident Database.", "claim": "the operator: a project of the Responsible AI Collaborative, chartered to advance it", "modality": "documented", "kind": "attribution", "kernel": true, "const": false, "lineage": "operator: the Responsible AI Collaborative", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 b12dad11c298cb0c…), whitespace-normalised" }, "F5": { "source": "B1", "origin": "card", "locus": "incident list (the highest id on the page, 2026-10-29)", "quote": "Incident 1788:", "claim": "the highest incident id listed on the home page on 2026-10-29 is 1788", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "count: incidents and reports", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 b12dad11c298cb0c…), whitespace-normalised" }, "F6": { "source": "B2", "origin": "card", "locus": "opening ('How is AI harming us?')", "quote": "The MIT AI Incident Tracker classifies 1,600 real-world report incidents from the AI Incident Database by risk, cause, harm, severity, and other relevant dimensions.", "claim": "MIT's tracker classifies 1,600 AIID incidents by risk, cause, harm and severity (the card snippet gives 1,752; the page's meta description 'more than 1,400')", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "derived layer: MIT AI Incident Tracker", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B2.txt (sha256 8d0169460225d228…), whitespace-normalised" }, "F7": { "source": "B2", "origin": "card", "locus": "What is the AI Incident Tracker? ¶1", "quote": "Raw reports from the AI Incident Database (AIID) are processed by Arcola AI's LLM pipeline, which categorizes them using established frameworks, such as the MIT Risk Repository and a harm severity rating system based on CSET’s AI Harm Taxonomy.", "claim": "the tracker's method: AIID raw reports processed by Arcola AI's LLM pipeline against the MIT Risk Repository and a CSET-based severity scale", "modality": "documented", "kind": "function", "kernel": false, "const": false, "lineage": "derived layer: MIT AI Incident Tracker", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B2.txt (sha256 8d0169460225d228…), whitespace-normalised" }, "F8": { "source": "B2", "origin": "card", "locus": "What is the AI Incident Tracker? ¶3", "quote": "We build on the great work done by the Responsible AI Collaborative who created and maintain the AI Incident Database, reviewing all submitted reports and ensuring the quality of the dataset.", "claim": "MIT credits the Responsible AI Collaborative with creating and maintaining the database and reviewing all submitted reports", "modality": "attributed", "kind": "attribution", "kernel": false, "const": false, "lineage": "operator: the Responsible AI Collaborative", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B2.txt (sha256 8d0169460225d228…), whitespace-normalised" }, "F9": { "source": "B2", "origin": "card", "locus": "Limitations of the analysis and dataset", "quote": "As the reporting is voluntary, the dataset is inevitably subject to some degree of sampling bias. … a systematic validation study has not yet been completed.", "claim": "the tracker's stated limits: voluntary reporting gives sampling bias; no systematic validation study yet", "modality": "self-description", "kind": "limit", "kernel": true, "const": false, "lineage": "derived layer: MIT AI Incident Tracker", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B2.txt (sha256 8d0169460225d228…), whitespace-normalised" }, "F10": { "source": "B3", "origin": "card", "locus": "Workstreams / Safety Critical AI / AI Incident Database, Overview ¶1–2", "quote": "The AI Incident Database (AIID) is a tool designed to help us better imagine and anticipate these risks, collecting more than 1,200 reports of intelligent systems causing safety, fairness, or other real-world problems. … As a central, systematized repository of problems experienced in the real world as a result of AI, this crowdsourced database can help AI practitioners mitigate or avoid repeated bad outcomes in the future.", "claim": "Partnership on AI's workstream page: a tool to anticipate risks, more than 1,200 reports, a crowdsourced central repository", "modality": "stipulation", "kind": "definition", "kernel": true, "const": false, "lineage": "what it is: indexing harms and near harms", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 5c2ef91014c3c406…), whitespace-normalised" }, "F11": { "source": "B3", "origin": "card", "locus": "Overview ¶1", "quote": "As AI technology is integrated into an increasing number of safety-critical systems — entering domains such as transportation, healthcare, and energy — the potential impact of this technology’s failures similarly grows.", "claim": "safety-critical domains (transportation, healthcare, energy) raise the impact of AI failures", "modality": "interpretation", "kind": "interpretation", "kernel": false, "const": false, "lineage": "purpose: a collective memory of failures, after aviation", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 5c2ef91014c3c406…), whitespace-normalised" }, "F12": { "source": "B4", "origin": "card", "locus": "catalogue entry, description ¶1", "quote": "The AI Incident Database is free and open-source project dedicated to indexing the collective history of harms or near harms realized in the real world by the deployment of artificial intelligence systems.", "claim": "the OECD catalogue restates the home page's definition with 'free and open-source'", "modality": "documented", "kind": "definition", "kernel": false, "const": false, "lineage": "what it is: indexing harms and near harms", "date": "2024-04-25 (catalogue upload)", "quote_ok": true, "checked_against": "saved text B4.txt (sha256 20239acc0da5164e…), whitespace-normalised" }, "F13": { "source": "B4", "origin": "card", "locus": "About the tool: usage rights, licence", "quote": "Usage rights: … Open source/Permissive … License: … Apache 2.0", "claim": "usage rights open source/permissive; licence Apache 2.0 (the catalogue does not say whether of code or data)", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "openness: code and data", "date": "2024-04-25 (catalogue upload)", "quote_ok": true, "checked_against": "saved text B4.txt (sha256 20239acc0da5164e…), whitespace-normalised" }, "F14": { "source": "B4", "origin": "card", "locus": "entry header; About the tool: developing organisation", "quote": "Uploaded on Apr 25, 2024 … Developing organisation(s): … responsible ai collaborative", "claim": "the catalogue entry, uploaded 25 April 2024, names the Responsible AI Collaborative as developing organisation", "modality": "documented", "kind": "attribution", "kernel": false, "const": false, "lineage": "operator: the Responsible AI Collaborative", "date": "2024-04-25 (catalogue upload)", "quote_ok": true, "checked_against": "saved text B4.txt (sha256 20239acc0da5164e…), whitespace-normalised" }, "F15": { "source": "B5", "origin": "card", "locus": "title, subtitle and description (Kaggle API dataset view)", "quote": "32 verified incidents · 29 companies · 13 countries · 24 columns — Every major AI failure, documented … License: CC BY-SA 4.0", "claim": "a Kaggle dataset of 32 incidents, 29 companies, 13 countries, 24 columns, CC BY-SA 4.0", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "homonym: the Kaggle compilation", "date": "2026-02-28 (last updated)", "quote_ok": true, "checked_against": "saved text B5.txt (sha256 da6de3c20c5e337a…), whitespace-normalised" }, "F16": { "source": "B5", "origin": "card", "locus": "Why This Dataset?", "quote": "AI incidents surged **56.4% in 2024 alone** (Stanford HAI AI Index 2025).", "claim": "the Kaggle dataset cites Stanford HAI's AI Index 2025 for a 56.4% surge in AI incidents in 2024", "modality": "attributed", "kind": "fact", "kernel": false, "const": false, "lineage": "homonym: the Kaggle compilation", "date": "2026-02-28 (last updated)", "quote_ok": true, "checked_against": "saved text B5.txt (sha256 da6de3c20c5e337a…), whitespace-normalised" }, "F17": { "source": "B5", "origin": "card", "locus": "Data Sources table", "quote": "| **AI Incident Database (incidentdatabase.ai)** | Primary incident records and AIID identifiers |", "claim": "the Kaggle dataset takes primary incident records and AIID identifiers from the database", "modality": "documented", "kind": "genealogy", "kernel": false, "const": false, "lineage": "homonym: the Kaggle compilation", "date": "2026-02-28 (last updated)", "quote_ok": true, "checked_against": "saved text B5.txt (sha256 da6de3c20c5e337a…), whitespace-normalised" }, "F18": { "source": "B6", "origin": "card", "locus": "card title (video, 1:35; not found, title only)", "quote": "AI Incident Database | TIME posted on the topic", "claim": "a TIME clip on LinkedIn about the database (title only)", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: press and video", "date": "— (title only)", "quote_ok": true, "checked_against": "the card's title/snippet in the registry's cite_list" }, "F19": { "source": "B7", "origin": "card", "locus": "Quick Definitions: AI incident (also §2.1)", "quote": "An alleged harm or near harm event to people, property, or the environment where an AI system is implicated.", "claim": "the incident defined: an alleged harm or near-harm event to people, property or environment in which an AI system is implicated", "modality": "stipulation", "kind": "definition", "kernel": true, "const": true, "lineage": "the unit: incident, report, issue, variant", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F20": { "source": "B7", "origin": "card", "locus": "Quick Definitions: AI issue", "quote": "An alleged harm or near harm by an AI system that has yet to occur or be detected.", "claim": "the issue defined: a harm or near harm that has yet to occur or be detected", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "the unit: incident, report, issue, variant", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F21": { "source": "B7", "origin": "card", "locus": "Quick Definitions: Incident ID", "quote": "The AIID editorial record that bounds and summarizes a specific harmful or near-harm event or traceable episode.", "claim": "the incident ID defined: the editorial record bounding and summarizing an event", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "the unit: incident, report, issue, variant", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F22": { "source": "B7", "origin": "card", "locus": "Quick Definitions: Incident report", "quote": "A source document attached to an incident ID. Multiple reports may document the same underlying event.", "claim": "the report defined: a source document attached to an incident ID; several may document one event", "modality": "stipulation", "kind": "definition", "kernel": true, "const": true, "lineage": "the unit: incident, report, issue, variant", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F23": { "source": "B7", "origin": "card", "locus": "Quick Definitions: fast boundary rule", "quote": "A report is evidence; an incident ID is the database’s normalized account of an event. A later source about the same underlying event is usually an additional report, not a new incident.", "claim": "report is evidence, incident ID the normalized account; a later source on the same event is usually an additional report", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "the unit: incident, report, issue, variant", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F24": { "source": "B7", "origin": "card", "locus": "Quick Definitions: AI incident variant", "quote": "A distinct incident involving the same intelligent system, substantially similar causative factors, and substantially similar harms as a known incident.", "claim": "the variant defined: a distinct incident, same system, substantially similar causes and harms", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "the unit: incident, report, issue, variant", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F25": { "source": "B7", "origin": "card", "locus": "header: governing principle", "quote": "The database records publicly documented allegations of harm or near harm. It does not independently adjudicate every factual claim.", "claim": "governing principle: it records publicly documented allegations and does not adjudicate every factual claim", "modality": "self-description", "kind": "limit", "kernel": true, "const": false, "lineage": "what it represents and its limits", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F26": { "source": "B7", "origin": "card", "locus": "§3.1 The Five-Question Test", "quote": "1. IS THERE A SPECIFIC EVENT? … 2. IS AN AI SYSTEM IDENTIFIABLE? … 3. IS THE AI SYSTEM CAUSALLY IMPLICATED? … 4. IS THERE HARM OR CREDIBLE NEAR HARM? … 5. IS THE EVIDENCE ADEQUATE?", "claim": "inclusion by five questions: specific event, identifiable AI system, causal implication, harm or credible near harm, adequate evidence", "modality": "stipulation", "kind": "definition", "kernel": true, "const": false, "lineage": "admission: the five-question test", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F27": { "source": "B7", "origin": "card", "locus": "§3.1, question 5", "quote": "At least one source must support the essential event, and uncertainty must be expressible without turning the record into speculation.", "claim": "the evidence condition: at least one source supports the essential event", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "admission: the five-question test", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F28": { "source": "B7", "origin": "card", "locus": "§2.3 Implication and Causality ¶1", "quote": "An AI system is implicated when it played an important role in the chain of events that produced the alleged harm. It need not be the sole or dominant cause.", "claim": "implication defined: an important role in the chain of events, not necessarily sole or dominant", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "admission: the five-question test", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F29": { "source": "B7", "origin": "card", "locus": "§2.4 Harm and Near Harm ¶2", "quote": "A near-harm event exists when the system helped create a chain of events that could readily have produced harm but an external factor prevented the outcome. The external factor must be independent of the system and should not simply be a built-in safeguard functioning as designed.", "claim": "near harm defined: a chain that could readily have produced harm, prevented by an external factor independent of the system", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "admission: the five-question test", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F30": { "source": "B7", "origin": "card", "locus": "§3.3 Presumptive Exclusions", "quote": "General criticism of a product or technology class without identified harm. … Benchmark failures, model errors, or hallucinations that remain inside testing or private drafting and have no consequential use. … These are presumptions rather than absolute prohibitions.", "claim": "presumptive exclusions, among them general criticism without identified harm and failures that stay inside testing; presumptions, not prohibitions", "modality": "stipulation", "kind": "limit", "kernel": false, "const": false, "lineage": "admission: the five-question test", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F31": { "source": "B7", "origin": "card", "locus": "§1.1 Core Record Types: Duplicate mapping", "quote": "A retired entity or incident identifier redirected to the canonical record that replaced it.", "claim": "duplicate mapping: a retired identifier redirected to its canonical record", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "the record: citation, merge, snapshot", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F32": { "source": "B7", "origin": "card", "locus": "§11.1 What the Database Represents", "quote": "The AIID is a curated archive of publicly documented incidents, issues, reports, and responses. … It is not a neutral mirror of all AI harms, a comprehensive denominator for risk rates, or an independent adjudication of every event.", "claim": "a curated archive of publicly documented incidents, not a neutral mirror, denominator or adjudication", "modality": "self-description", "kind": "limit", "kernel": true, "const": false, "lineage": "what it represents and its limits", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F33": { "source": "B7", "origin": "card", "locus": "§11.2 Coverage and Selection Effects", "quote": "Wealthy countries, prominent companies, spectacular events, and English-language reporting are likely to be overrepresented. Harms affecting marginalized groups, occurring inside private organizations, or unfolding as ordinary administration may be underrepresented.", "claim": "coverage: over- and under-representation", "modality": "self-description", "kind": "limit", "kernel": false, "const": false, "lineage": "what it represents and its limits", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F34": { "source": "B7", "origin": "card", "locus": "§11.3 Epistemic Status", "quote": "A lack of records is not evidence that a sector or country has no incidents.", "claim": "absence of records is not evidence of absence of incidents", "modality": "self-description", "kind": "limit", "kernel": false, "const": false, "lineage": "what it represents and its limits", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F35": { "source": "B7", "origin": "card", "locus": "§1.6 Live Data and Point-in-Time Snapshots", "quote": "The live site can change after an export is created.", "claim": "the live site may differ from an export", "modality": "self-description", "kind": "limit", "kernel": false, "const": false, "lineage": "the record: citation, merge, snapshot", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F36": { "source": "B7", "origin": "card", "locus": "header; Version 2 note on AI assistance", "quote": "Version 2 · October 1, 2026 · Daniel Atherton and Sean McGregor · Operational working guide … Generative AI tools were used during the development of Version 2 to assist with research, document organization, drafting, revision, and consistency review.", "claim": "the guide's version, date and authors, and its disclosed AI assistance", "modality": "self-description", "kind": "genealogy", "kernel": false, "const": false, "lineage": "intake: open submission, editor resolution", "date": "2026-10-01 (Version 2)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 52e3176de91eafff…), whitespace-normalised" }, "F37": { "source": "B8", "origin": "card", "locus": "¶1", "quote": "In aviation, an accident is a case where substantial damage or loss of life occurs. Incidents are cases where the risk of an accident substantially increases.", "claim": "aviation's accident and incident, the model of the term", "modality": "attributed", "kind": "definition", "kernel": false, "const": false, "lineage": "purpose: a collective memory of failures, after aviation", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B8.txt (sha256 0c0c386a8a03514f…), whitespace-normalised" }, "F38": { "source": "B8", "origin": "card", "locus": "¶2", "quote": "this database adopts an adaptive criteria for ingesting \"incidents\" where reports are accepted or rejected on the basis of a growing rule set defined in the Editor's Guide.", "claim": "adaptive criteria: reports accepted or rejected by a growing rule set in the Editor's Guide", "modality": "stipulation", "kind": "definition", "kernel": true, "const": false, "lineage": "admission: the five-question test", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B8.txt (sha256 0c0c386a8a03514f…), whitespace-normalised" }, "F39": { "source": "B9", "origin": "card", "locus": "card title (video, 1:35; not found, title only)", "quote": "AI Incidents Rise 50% Year-Over-Year: AI Incident Database | TIME posted on the ...", "claim": "a TIME clip on LinkedIn headed 'AI Incidents Rise 50% Year-Over-Year' (title only)", "modality": "attributed", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: press and video", "date": "— (title only)", "quote_ok": true, "checked_against": "the card's title/snippet in the registry's cite_list" }, "F40": { "source": "B10", "origin": "card", "locus": "oEmbed title (YouTube, HarrietHacks, 6:59)", "quote": "Examples of AI real attacks", "claim": "a YouTube video, 'Examples of AI real attacks' (title only)", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: press and video", "date": "— (title only)", "quote_ok": true, "checked_against": "saved text B10.txt (sha256 76c222740d387425…), whitespace-normalised" }, "F41": { "source": "B11", "origin": "card", "locus": "Why \"AI Incidents\"?", "quote": "Intelligent systems are currently prone to unforeseen and often dangerous failures when they are deployed to the real world. Much like the transportation sector before it (e.g., FAA and FARS) and more recently computer systems, intelligent systems require a repository of problems experienced in the real world so that future researchers and developers may mitigate or avoid repeated bad outcomes.", "claim": "why AI incidents: deployed systems fail; like transportation and computer systems they need a repository of real-world problems", "modality": "interpretation", "kind": "function", "kernel": false, "const": false, "lineage": "purpose: a collective memory of failures, after aviation", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B11.txt (sha256 0b85d8772dcb1205…), whitespace-normalised" }, "F42": { "source": "B11", "origin": "card", "locus": "What is an Incident?", "quote": "The initial set of more than 1,000 incident reports have been intentionally broad in nature. Current examples include, … An autonomous car kills a pedestrian", "claim": "the initial set of more than 1,000 reports intentionally broad (examples: an autonomous car kills a pedestrian, a flash crash, a wrongful arrest)", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "count: incidents and reports", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B11.txt (sha256 0b85d8772dcb1205…), whitespace-normalised" }, "F43": { "source": "B11", "origin": "card", "locus": "When Should You Report an Incident?", "quote": "When in doubt of whether an event qualifies as an incident, please submit it! This project is intended to converge on a shared criteria for ingesting incidents through exploration of the candidate incidents submitted by the broader community.", "claim": "submit when in doubt; criteria to converge through the community's candidates", "modality": "stipulation", "kind": "function", "kernel": false, "const": false, "lineage": "admission: the five-question test", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B11.txt (sha256 0b85d8772dcb1205…), whitespace-normalised" }, "F44": { "source": "B11", "origin": "card", "locus": "Board of Directors ¶1", "quote": "The incident database is managed in a participatory manner by persons and organizations contributing code, research, and broader impacts.", "claim": "participatory management", "modality": "self-description", "kind": "attribution", "kernel": false, "const": false, "lineage": "intake: open submission, editor resolution", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B11.txt (sha256 0b85d8772dcb1205…), whitespace-normalised" }, "F45": { "source": "B11", "origin": "card", "locus": "Collaborators: Incident Editors; lead editor", "quote": "Incident Editors: People that resolve incident submissions to the database and maintain them. … Daniel Atherton is the lead AI incidents editor.", "claim": "incident editors resolve submissions and maintain them; Daniel Atherton lead editor", "modality": "documented", "kind": "attribution", "kernel": true, "const": false, "lineage": "intake: open submission, editor resolution", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B11.txt (sha256 0b85d8772dcb1205…), whitespace-normalised" }, "F46": { "source": "B11", "origin": "card", "locus": "Emeritus Board: Sean McGregor; Collaborators", "quote": "launching the AI Incident Database … Sean McGregor is the Executive Director.", "claim": "Sean McGregor launched the database and is its Executive Director", "modality": "documented", "kind": "attribution", "kernel": false, "const": false, "lineage": "operator: the Responsible AI Collaborative", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B11.txt (sha256 0b85d8772dcb1205…), whitespace-normalised" }, "F47": { "source": "B11", "origin": "card", "locus": "sponsor block; Open Source Contributors", "quote": "Database Founding Sponsor … Jingying Yang and Dr. Christine Custis contributed significantly to the early stages of the AIID via their roles with the Partnership on AI.", "claim": "a Database Founding Sponsor heads a logo block whose image the HTML labels 'Partnership on AI'; two people contributed to the early stages through their roles with the Partnership on AI", "modality": "documented", "kind": "genealogy", "kernel": false, "const": false, "lineage": "origin: Partnership on AI and the founding paper", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B11.txt (sha256 0b85d8772dcb1205…), whitespace-normalised" }, "F48": { "source": "B11", "origin": "card", "locus": "Collaborators: Taxonomy Editors", "quote": "Taxonomy Editors: Organizations or people that have contributed taxonomies to the database. … Center for Security and Emerging Technology (CSET) … Nikiforos Pittaras (GMF)", "claim": "taxonomies contributed by CSET and Nikiforos Pittaras (GMF)", "modality": "documented", "kind": "attribution", "kernel": false, "const": false, "lineage": "intake: open submission, editor resolution", "date": "2026-10-29 (undated; as fetched)", "quote_ok": true, "checked_against": "saved text B11.txt (sha256 0b85d8772dcb1205…), whitespace-normalised" }, "F49": { "source": "B12", "origin": "card", "locus": "Abstract (article page)", "quote": "The AI Incident Database (AIID) is a project that catalogs AI incidents and supports further research by providing a platform to classify incidents for different operational and research-oriented goals.", "claim": "Paeth et al. (AAAI-25): a project cataloguing AI incidents with a platform to classify them", "modality": "interpretation", "kind": "definition", "kernel": false, "const": false, "lineage": "what it is: indexing harms and near harms", "date": "2025-04-11 (published)", "quote_ok": true, "checked_against": "saved text B12.txt (sha256 8dbf365d930acee0…), whitespace-normalised" }, "F50": { "source": "B12", "origin": "card", "locus": "Abstract (article page)", "quote": "We find that certain patterns of AI incidents present structural ambiguities that challenge incident databasing and explore how epistemic uncertainty in AI incident reporting is unavoidable.", "claim": "finding: structural ambiguities challenge incident databasing; epistemic uncertainty unavoidable", "modality": "interpretation", "kind": "interpretation", "kernel": true, "const": false, "lineage": "what it represents and its limits", "date": "2025-04-11 (published)", "quote_ok": true, "checked_against": "saved text B12.txt (sha256 8dbf365d930acee0…), whitespace-normalised" }, "F51": { "source": "B12", "origin": "card", "locus": "PDF, 'The AI Incident Database and Related Work'", "quote": "Inspired by the CVE program and other incident reporting practices, the AI Incident Database (AIID) catalogs AI incidents with the goal of reducing the likelihood that similar incidents recur … Currently there are over 750 distinct AI incidents detailed by more than 3000 indexed third-party reports.", "claim": "inspired by the CVE program; over 750 incidents and more than 3000 third-party reports at the paper's writing", "modality": "documented", "kind": "genealogy", "kernel": false, "const": false, "lineage": "count: incidents and reports", "date": "2025-04-11 (published)", "quote_ok": true, "checked_against": "saved text B12.pdf.txt (sha256 9dba8efdeb7686b5…), whitespace-normalised" }, "F52": { "source": "B13", "origin": "card", "locus": "card snippet (ACM DL returned 403)", "quote": "Users can submit examples of incidents and controversies; these submissions are included based on adherence to criteria such as relevance, fairness, and accurac...", "claim": "the ACM paper's snippet: user submissions of incidents and controversies, included by criteria such as relevance, fairness and accuracy (which database is meant is not readable)", "modality": "attributed", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: press and video", "date": "2023 (AIES '23, by Crossref)", "quote_ok": true, "checked_against": "the card's title/snippet in the registry's cite_list" }, "F53": { "source": "B14", "origin": "card", "locus": "oEmbed title (YouTube, International Association for Safe & Ethical AI, 19m)", "quote": "Establishing Effective Serious Incident Reporting Frameworks – Jimmy Farrell | IASEAI 2025", "claim": "a talk on serious-incident reporting frameworks (title only)", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: press and video", "date": "2025 (IASEAI 2025, by title)", "quote_ok": true, "checked_against": "saved text B14.txt (sha256 8c9f8a79bff55d6b…), whitespace-normalised" }, "F54": { "source": "B15", "origin": "own_page", "locus": "Snapshot Downloads", "quote": "Weekly point-in-time dumps of the full database, available in JSON, MongoDB archive, and CSV formats. … 2026-10-26 10:15 AM · 121.49 MB · backup-20261026101519.tar.bz2 … 2021-03-29 07:49 AM · 2.49 MB · backup-20210329074910.tar.bz2", "claim": "weekly full dumps in JSON, MongoDB and CSV, from 2021-03-29 to 2026-10-26", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "the record: citation, merge, snapshot", "date": "2026-10-26 (latest snapshot listed)", "quote_ok": true, "checked_against": "saved text B15.txt (sha256 1eb3ce6104c8fb98…), whitespace-normalised" }, "F55": { "source": "B15", "origin": "own_page", "locus": "Citing a Specific Incident", "quote": "Every incident has its own suggested citation that credits both the submitter(s) of the incident and the editor(s) of the incident. … You can find incident citations at https://incidentdatabase.ai/cite/INSERT_NUMBER_HERE.", "claim": "a suggested citation per incident crediting submitters and editors, at /cite/N", "modality": "stipulation", "kind": "function", "kernel": true, "const": false, "lineage": "the record: citation, merge, snapshot", "date": "2026-10-26 (latest snapshot listed)", "quote_ok": true, "checked_against": "saved text B15.txt (sha256 1eb3ce6104c8fb98…), whitespace-normalised" }, "F56": { "source": "B15", "origin": "own_page", "locus": "Citing the Database as a Whole", "quote": "McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. In Proceedings of the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21).", "claim": "the database as a whole cited by McGregor 2021 (IAAI-21)", "modality": "stipulation", "kind": "genealogy", "kernel": false, "const": false, "lineage": "origin: Partnership on AI and the founding paper", "date": "2026-10-26 (latest snapshot listed)", "quote_ok": true, "checked_against": "saved text B15.txt (sha256 1eb3ce6104c8fb98…), whitespace-normalised" }, "F57": { "source": "B15", "origin": "own_page", "locus": "Excel Export", "quote": "An Excel file, updated weekly, that combines all numbered incidents with many of their taxonomy classifications.", "claim": "a weekly Excel export of numbered incidents with taxonomy classifications", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "the record: citation, merge, snapshot", "date": "2026-10-26 (latest snapshot listed)", "quote_ok": true, "checked_against": "saved text B15.txt (sha256 1eb3ce6104c8fb98…), whitespace-normalised" }, "F58": { "source": "B16", "origin": "own_page", "locus": "title; Description; Entities", "quote": "Incident 693: Google AI Reportedly Delivering Confidently Incorrect and Harmful Information … Google's AI search engine has reportedly been providing users with confidently incorrect and often harmful information. … Alleged: Google developed and deployed an AI system, which harmed Google users and General public.", "claim": "incident 693: Google's AI search reportedly giving confidently incorrect and harmful information; Google developer and deployer; Google users and the general public harmed", "modality": "attributed", "kind": "fact", "kernel": true, "const": false, "lineage": "the index case: incident 693", "date": "2024-05-14 (incident date)", "quote_ok": true, "checked_against": "saved text B16.txt (sha256 9f9d4305be3e45b8…), whitespace-normalised" }, "F59": { "source": "B16", "origin": "own_page", "locus": "Incident Stats", "quote": "Report Count … Incident Date … 2024-05-14 … Editors … Daniel Atherton", "claim": "incident 693's stats: report count (7 reports listed), incident date 2024-05-14, editor Daniel Atherton", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "the index case: incident 693", "date": "2024-05-14 (incident date)", "quote_ok": true, "checked_against": "saved text B16.txt (sha256 9f9d4305be3e45b8…), whitespace-normalised" }, "F60": { "source": "B16", "origin": "own_page", "locus": "Incident Reports (Business Insider report excerpt)", "quote": "Google's new search feature, AI Overviews, seems to be going awry.", "claim": "a report attached to 693 names AI Overviews", "modality": "attributed", "kind": "fact", "kernel": false, "const": false, "lineage": "the index case: incident 693", "date": "2024-05-14 (incident date)", "quote_ok": true, "checked_against": "saved text B16.txt (sha256 9f9d4305be3e45b8…), whitespace-normalised" }, "F61": { "source": "B16", "origin": "own_page", "locus": "Applied Taxonomies: MIT", "quote": "MIT Taxonomy Classifications … Machine-Classified … 3.1. False or misleading information", "claim": "693 machine-classified under MIT's taxonomy: false or misleading information", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "the index case: incident 693", "date": "2024-05-14 (incident date)", "quote_ok": true, "checked_against": "saved text B16.txt (sha256 9f9d4305be3e45b8…), whitespace-normalised" }, "F62": { "source": "B16", "origin": "own_page", "locus": "Variants", "quote": "A \"variant\" is an AI incident similar to a known case—it has the same causes, harms, and AI system. Instead of listing it separately, we group it under the first reported incident. Unlike other incidents, variants do not need to have been reported outside the AIID.", "claim": "the incident page's variant: same causes, harms and system, grouped under the first reported incident; need not have been reported outside the AIID", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "the unit: incident, report, issue, variant", "date": "2024-05-14 (incident date)", "quote_ok": true, "checked_against": "saved text B16.txt (sha256 9f9d4305be3e45b8…), whitespace-normalised" }, "F63": { "source": "B16", "origin": "own_page", "locus": "notice above Entities", "quote": "The OECD AI Incidents and Hazards Monitor (AIM) automatically collects and classifies AI-related incidents and hazards in real time from reputable news sources worldwide.", "claim": "the incident page describes its neighbour, the OECD AIM, as automatic and news-derived", "modality": "attributed", "kind": "fact", "kernel": false, "const": false, "lineage": "neighbours", "date": "2024-05-14 (incident date)", "quote_ok": true, "checked_against": "saved text B16.txt (sha256 9f9d4305be3e45b8…), whitespace-normalised" }, "F64": { "source": "B17", "origin": "own_page", "locus": "Citation record for Incident 637", "quote": "This incident is a duplicate of Incident 112. All new reports and citations should be directed to incident 112. The reports previously found on this page have been migrated to the previously existing incident.", "claim": "incident 637 kept as a citation page pointing to 112, its reports migrated", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "the record: citation, merge, snapshot", "date": "2026-10-29 (as fetched)", "quote_ok": true, "checked_against": "saved text B17.txt (sha256 f19d09642183d686…), whitespace-normalised" }, "F65": { "source": "B18", "origin": "own_page", "locus": "abstract page header", "quote": "[Submitted on 17 Nov 2020] … Authors:Sean McGregor", "claim": "the founding paper, by Sean McGregor, submitted 17 November 2020", "modality": "documented", "kind": "genealogy", "kernel": false, "const": false, "lineage": "origin: Partnership on AI and the founding paper", "date": "2020-11-17 (submitted)", "quote_ok": true, "checked_against": "saved text B18.txt (sha256 70b0cce7b4c32880…), whitespace-normalised" }, "F66": { "source": "B18", "origin": "own_page", "locus": "Abstract", "quote": "Mature industrial sectors (e.g., aviation) collect their real world failures in incident databases to inform safety improvements. Intelligent systems currently cause real world harms without a collective memory of their failings.", "claim": "the founding problem: aviation's incident databases; intelligent systems without a collective memory of failure", "modality": "interpretation", "kind": "interpretation", "kernel": false, "const": false, "lineage": "purpose: a collective memory of failures, after aviation", "date": "2020-11-17 (submitted)", "quote_ok": true, "checked_against": "saved text B18.txt (sha256 70b0cce7b4c32880…), whitespace-normalised" }, "F67": { "source": "B18", "origin": "own_page", "locus": "Abstract", "quote": "The AI Incident Database is an incident collection initiated by an industrial/non-profit cooperative to enable AI incident avoidance and mitigation.", "claim": "an incident collection initiated by an industrial/non-profit cooperative", "modality": "stipulation", "kind": "definition", "kernel": true, "const": false, "lineage": "origin: Partnership on AI and the founding paper", "date": "2020-11-17 (submitted)", "quote_ok": true, "checked_against": "saved text B18.txt (sha256 70b0cce7b4c32880…), whitespace-normalised" }, "F68": { "source": "B18", "origin": "own_page", "locus": "Abstract", "quote": "The database supports a variety of research and development use cases with faceted and full text search on more than 1,000 incident reports archived to date.", "claim": "faceted and full-text search over more than 1,000 reports at launch", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "count: incidents and reports", "date": "2020-11-17 (submitted)", "quote_ok": true, "checked_against": "saved text B18.txt (sha256 70b0cce7b4c32880…), whitespace-normalised" }, "F69": { "source": "B19", "origin": "wikipedia", "locus": "raw wikitext of the redirect page (action=raw&redirect=no)", "quote": "#REDIRECT [[Partnership on AI]] … {{R from subtopic}} … {{R with possibilities}}", "claim": "'AI Incident Database' redirects to Partnership on AI, no section anchor; tagged from a subtopic, with possibilities", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "reception: the encyclopedia", "date": "2026-10-29 (as fetched)", "quote_ok": true, "checked_against": "saved text B19.wiki (sha256 03d7f4dfc50b5315…), whitespace-normalised" }, "F70": { "source": "B20", "origin": "wikipedia", "locus": "History ¶5", "quote": "In November 2020 the Partnership on AI announced the AI Incident Database (AIID),[10] which is a tool to identify, assess, manage, and communicate AI risk and harm.", "claim": "Wikipedia: announced by Partnership on AI in November 2020, a tool for AI risk and harm", "modality": "documented", "kind": "genealogy", "kernel": true, "const": false, "lineage": "origin: Partnership on AI and the founding paper", "date": "2026-10-29 (as fetched)", "quote_ok": true, "checked_against": "saved text B20.txt (sha256 1b80fdf3c9e46b6c…), whitespace-normalised" }, "F71": { "source": "B20", "origin": "wikipedia", "locus": "History ¶6", "quote": "The response also highlighted how the AI Incident Database involves some of the minimum attributes in NIST's AI RMF, such as being consensus-driven, risk-based, adaptable, and consistent with other approaches to managing AI risk.", "claim": "Wikipedia: PAI's NIST response presented it as meeting attributes of the AI RMF", "modality": "attributed", "kind": "interpretation", "kernel": false, "const": false, "lineage": "origin: Partnership on AI and the founding paper", "date": "2026-10-29 (as fetched)", "quote_ok": true, "checked_against": "saved text B20.txt (sha256 1b80fdf3c9e46b6c…), whitespace-normalised" }, "F72": { "source": "B20", "origin": "wikipedia", "locus": "Mission and Principles ¶5", "quote": "Notable initiatives include the establishment of the AI incident Database, formulation of norms for responsible publication, and the creation of the innovative AI learning environment SafeLife.", "claim": "Wikipedia: among the Safety Critical AI program's initiatives", "modality": "documented", "kind": "genealogy", "kernel": false, "const": false, "lineage": "origin: Partnership on AI and the founding paper", "date": "2026-10-29 (as fetched)", "quote_ok": true, "checked_against": "saved text B20.txt (sha256 1b80fdf3c9e46b6c…), whitespace-normalised" }, "F73": { "source": "B20", "origin": "wikipedia", "locus": "Programs and initiatives ¶6", "quote": "In November 2020, the Partnership on AI announced the AI Incident Database (AIID) which shifted to a new special-purpose independent non-profit in 2022.", "claim": "Wikipedia: shifted to a new special-purpose independent non-profit in 2022 (the non-profit unnamed)", "modality": "documented", "kind": "genealogy", "kernel": true, "const": false, "lineage": "origin: Partnership on AI and the founding paper", "date": "2026-10-29 (as fetched)", "quote_ok": true, "checked_against": "saved text B20.txt (sha256 1b80fdf3c9e46b6c…), whitespace-normalised" }, "F74": { "source": "B20", "origin": "wikipedia", "locus": "reference [10] (McGregor, partnershiponai.org, 2020-11-18), quote", "quote": "Avoiding repeated AI failures requires making past failures known. Therefore, today we introduce a systematized collection of incidents where intelligent systems have caused safety, fairness, or other real-world problems: The AI Incident Database (AIID).", "claim": "McGregor's introduction on Partnership on AI's site, 18 November 2020, quoted in Wikipedia's reference", "modality": "attributed", "kind": "genealogy", "kernel": false, "const": false, "lineage": "origin: Partnership on AI and the founding paper", "date": "2026-10-29 (as fetched)", "quote_ok": true, "checked_against": "saved text B20.txt (sha256 1b80fdf3c9e46b6c…), whitespace-normalised" }, "F75": { "source": "B20", "origin": "wikipedia", "locus": "article header templates (raw wikitext)", "quote": "{{Primary sources|date=April 2025}} … {{AI-generated|date=October 2025}}", "claim": "the article is tagged for primary sources (April 2025) and as AI-generated (October 2025)", "modality": "documented", "kind": "limit", "kernel": false, "const": false, "lineage": "reception: the encyclopedia", "date": "2026-10-29 (as fetched)", "quote_ok": true, "checked_against": "saved text B20.wiki (sha256 08e1303706a2474f…), whitespace-normalised" } }, "field_sources": { "B1": "AI Incident Database — AI Incident Database", "B2": "MIT AI Risk Initiative — MIT AI Incident Tracker", "B3": "Partnership on AI — AI Incident Database - Partnership on AI", "B4": "OECD AI Policy Observatory — AI Incident Database", "B5": "Kaggle — AI Incidents, Failures & Controversies Database - Kaggle", "B6": "LinkedIn·TIME — AI Incident Database | TIME posted on the topic", "B7": "AI Incident Database — Editorial and Research Guide for the AI Incident Database", "B8": "AI Incident Database — Defining an “AI Incident”", "B9": "LinkedIn·TIME — AI Incidents Rise 50% Year-Over-Year: AI Incident Database | TIME posted on the ...", "B10": "YouTube·HarrietHacks — Examples of AI real attacks", "B11": "AI Incident Database — The details about who and why for database - AI Incident Database", "B12": "The Association for the Advancement of Artificial Intelligence — Lessons for Editors of AI Incidents from the AI Incident Database - AAAI Publications", "B13": "ACM Digital Library — Why We Need to Know More: Exploring the State of AI Incident Documentation ...", "B14": "YouTube·International Association for Safe & Ethical AI — Establishing Effective Serious Incident Reporting Frameworks – Jimmy Farrell", "B15": "AI Incident Database — Database Backups and Snapshots", "B16": "AI Incident Database — Incident 693", "B17": "AI Incident Database — Citation record for Incident 637", "B18": "arXiv — [2011.08512] Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database", "B19": "Wikipedia — AI Incident Database (redirect)", "B20": "Wikipedia — Partnership on AI" }, "delta": { "T_vs_LB": "At 'ai incident database' (T1, six cards: B1–B6) the composition gives the database by its own pages and its catalogue entry. Composed from its cards: the definition (B1 F1, at the grain 'harms, near-harms, and failures'), the operator (B1 F4; B4 F14; B2 F8), 'free, open-source' (B4 F12, the OECD's phrase), the aviation model (B1 F2, 'computer security' composed as 'cybersecurity'), 'public and crowdsourced reports' (B1 F3; B3 F10 'crowdsourced'), safety-critical failures (B3 F11) and the MIT tracker as a derived layer (B2 F6–F7). Partly in the field, outside the ledgered spans: T1-3's list of harm kinds (bias, privacy leaks, deepfakes, vehicle accidents) is at that grain in no database card; its nearest words are the Kaggle dataset's (B5: 'deepfakes influencing elections', 'autonomous cars killing pedestrians', the tags 'bias' and 'deepfakes'). Absent from the readable field: T1-4's growth claim ('surged significantly, with deepfake and scam-related entries multiplying') has its markers on an unresolved token and the TIME clip (B6, title only); the nearest readable figure, Kaggle's '56.4% in 2024' (B5 F16), is Stanford HAI's count of AI incidents in general, not of the database's entries: if T drew on it, the reading is substituted (a general statistic composed as the database's growth) and composed without attribution, which the field cannot settle. Available, not composed: the incident as the unit and the report attached (absent on the cards of T1, present on T2's: F19, F22–F23); MIT's stated limits (F9: sampling bias, no validation study); the OECD's Apache 2.0 licence without its object (F13), which T's 'free, open-source' does not distinguish from a data licence (CAP-1-07); Partnership on AI as the source of the workstream page (F10–F11), whose page names no Collaborative; the Kaggle dataset's provenance from the AIID (F15–F17), not composed and not confused (CAP-1-06). Content replaced by use: the closer (T1-6). At 'what counts as an incident in the ai incident database' (T2, nine cards: B2, B7–B14) the composition reads the Editor's Guide (B7) and composes its definition verbatim (F19; T2-1), the adaptive criteria (B8 F38; T2-2) and the distinctions of incident, issue and variant (F19, F20, F24; T2-7). Limit dropped: the five questions (F26) become 'three core baseline thresholds', the identifiable AI system folded into implication and adequate evidence (F27) dropped (CAP-2-04); the near-harm condition loses its external factor independent of the system (F29; T2-5 'a credible chain of events nearly producing one'); the issue loses 'or be detected' (F20; T2-7 'has not yet materialized'). Force raised: the variant's 'substantially similar harms' (F24) is composed as 'matching outcomes' (T2-7). Composed with attribution at the source's grain: the definition and thresholds carry 'its AIID Editor's Guide'. Partly sourced: 'Theoretical vulnerabilities do not qualify' (the guide's issue, F20, and its §3.2 table outside the ledgered spans); 'AI implication' as a but-for test (F28 states implication as an important role, not necessarily sole; the but-for question is the guide's §2.3 practical test, outside the ledgered span); CSET's taxonomy (F48) with categories the ledger does not carry at that grain. Available, not composed: the report as evidence attached to an incident ID and the many-to-one rule (F21–F23; CAP-2-05); the governing principle that the database records allegations and does not adjudicate (F25); the presumptive exclusions as presumptions (F30); what the database represents and its coverage biases (F32–F34); Paeth and colleagues' finding of unavoidable epistemic uncertainty (F50) and their CVE inspiration (F51); the ACM snippet (F52). Content replaced by use: the closer (T2-8). Neither transcript names the Partnership on AI as founding sponsor (F47; Wikipedia, not a card at either address, composes the database inside Partnership on AI's article, F69–F73), the citation page per incident (F55), the snapshots (F54) or the founding paper (F65–F68); these are own pages and Wikipedia admitted beside the cards by the operator's ruling and were not surfaced by T.", "T_vs_LB_reading": "v0.9 §0.5: the first diff is already ontological, and here it is small. The field's ontology and T's agree on the entity (a database of AI harms run by the Responsible AI Collaborative); what T keeps of its cards is the database's self-description, its operator and its derived layer, and what it drops is the database's own structure of record (the incident as an editorial account with reports as evidence attached) and the guide's evidentiary limits (the fifth question, the external factor, 'or be detected', 'substantially similar'). At the name the composition stays at the grain of its about page (CAP-1-01); at the unit's question it reads the guide and compresses its test (CAP-2-01). The row records this as observed and claims nothing about the composer's internals (§0.5).", "LB_vs_LBA": "Admission adds, in one claim space with the field's: the database as an editor-resolved database of harm events whose unit carries its reports and a stable citation page (MMRS-01, MMRS-02, MMRS-06); what its selection leaves out, the archive's sense the field does not state: what an answer surface does with a source on an ordinary day has no unit, and attribution erasure, citation fidelity and provenance stripping enter only when they attach to a harm, so there is no field for attribution (MMRS-13, MMRS-14, MMRS-29), with the open question whether report texts are kept (MMRS-28); the citation page as the meta-aggregator's model (MMRS-16) and the merge rule that keeps a page up (MMRS-10), with the highest id as an upper bound on incidents (MMRS-20); the record kept apart from the entity, each assertion with a basis, the lifecycle apart from the observation (MMRS-30, MMRS-31) and its unchecked items (MMRS-32); the code-against-data licence distinction (MMRS-23, CAP-1-07); the neighbours and negative tags in the harm basin (MMRS-17, MMRS-18, MMRS-24, MMRS-35) and the Kaggle homonym set apart (MMRS-25, CAP-1-06); incident 693, Google's AI Overviews, as the field's index case (MMRS-15, MMRS-36); the battery's checks (MMRS-33); the composition layer's two observations at the entity's addresses (CAP-1-01–05, CAP-2-01–05); and the archive's own address of three of its terms to the database as a glossary venue (#543, R543-01). Coherence (P_coh, composed each with its source): the MMRS record's 'run by the Responsible AI Collaborative since 2020' (MMRS-01) against Wikipedia's shift to an independent non-profit in 2022 (F73); the guide's variant as a distinct incident (F24, MMRS-07) against incident 693's page, grouped under the first reported incident (F62); the MMRS record's quoted presumptive exclusions (MMRS-12) against the guide's words (F30), the substance agreeing. Restated lineages (§3.6, the own page keeping the earlier date): MMRS-03 with F1; MMRS-04, MMRS-37 with F4, F46, F47; MMRS-05, MMRS-19 with F65, F68; MMRS-06 with F19, F22; MMRS-07 with F24; MMRS-08 with F26; MMRS-09 with F3, F45; MMRS-11, MMRS-22, MMRS-27 with F54, F57; MMRS-34 with F6, F7. The measure is in row.measure.", "alignment": { "T1_cards_claims": [ "F1", "F2", "F3", "F4", "F5", "F6", "F7", "F8", "F9", "F10", "F11", "F12", "F13", "F14", "F15", "F16", "F17", "F18" ], "T1_composed": [ "F1", "F2", "F3", "F4", "F6", "F7", "F8", "F10", "F11", "F12", "F14" ], "T2_cards_claims": [ "F6", "F7", "F8", "F9", "F19", "F20", "F21", "F22", "F23", "F24", "F25", "F26", "F27", "F28", "F29", "F30", "F31", "F32", "F33", "F34", "F35", "F36", "F37", "F38", "F39", "F40", "F41", "F42", "F43", "F44", "F45", "F46", "F47", "F48", "F49", "F50", "F51", "F52", "F53" ], "T2_composed": [ "F19", "F20", "F24", "F26", "F28", "F29", "F38", "F48" ], "rule": "a field claim counts as composed when T states it at its source's grain or coarser; the typology of what is not is in T_vs_LB (§6.2)" } }, "kernel": [ { "K": "K1", "claim": "`CAP-1-01`", "source": "Capture Registry, 2026-10-29", "M_src": "interpretation", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K2", "claim": "`CAP-1-02`", "source": "Capture Registry, 2026-10-29", "M_src": "interpretation", "sense": "reception: the composition layer", "qualifiers carried": "CAP-1-01, CAP-1-04", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K3", "claim": "`CAP-1-03`", "source": "Capture Registry, 2026-10-29", "M_src": "documented", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K4", "claim": "`CAP-1-04`", "source": "Capture Registry, 2026-10-29", "M_src": "documented", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K5", "claim": "`CAP-1-05`", "source": "Capture Registry, 2026-10-29", "M_src": "documented", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K6", "claim": "`CAP-1-06`", "source": "Capture Registry, 2026-10-29", "M_src": "documented", "sense": "homonym: the Kaggle compilation", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K7", "claim": "`CAP-1-07`", "source": "Capture Registry, 2026-10-29", "M_src": "interpretation", "sense": "openness: code and data", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K8", "claim": "`CAP-2-01`", "source": "Capture Registry, 2026-10-29", "M_src": "interpretation", "sense": "reception: the composition layer", "qualifiers carried": "CAP-2-02, CAP-2-04", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K9", "claim": "`CAP-2-02`", "source": "Capture Registry, 2026-10-29", "M_src": "interpretation", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K10", "claim": "`CAP-2-03`", "source": "Capture Registry, 2026-10-29", "M_src": "documented", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K11", "claim": "`CAP-2-04`", "source": "Capture Registry, 2026-10-29", "M_src": "documented", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K12", "claim": "`CAP-2-05`", "source": "Capture Registry, 2026-10-29", "M_src": "documented", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K13", "claim": "`MMRS-01`", "source": "MMRS record", "M_src": "stipulation", "sense": "what it is: indexing harms and near harms", "qualifiers carried": "MMRS-02", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "qualified claim (F1, F4; dated against F73)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K14", "claim": "`MMRS-02`", "source": "MMRS record", "M_src": "stipulation", "sense": "the unit: incident, report, issue, variant", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "qualified claim (F19, F22: adds the stable citation page)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K15", "claim": "`MMRS-03`", "source": "MMRS record", "M_src": "attributed", "sense": "what it is: indexing harms and near harms", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F1; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K16", "claim": "`MMRS-04`", "source": "MMRS record", "M_src": "documented", "sense": "operator: the Responsible AI Collaborative", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F4, F46, F47; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K17", "claim": "`MMRS-05`", "source": "MMRS record", "M_src": "attributed", "sense": "origin: Partnership on AI and the founding paper", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F65, F68; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K18", "claim": "`MMRS-06`", "source": "MMRS record", "M_src": "attributed", "sense": "the unit: incident, report, issue, variant", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F19, F22; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K19", "claim": "`MMRS-07`", "source": "MMRS record", "M_src": "attributed", "sense": "the unit: incident, report, issue, variant", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F24; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K20", "claim": "`MMRS-08`", "source": "MMRS record", "M_src": "attributed", "sense": "admission: the five-question test", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F26; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K21", "claim": "`MMRS-09`", "source": "MMRS record", "M_src": "documented", "sense": "intake: open submission, editor resolution", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F3, F45; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K22", "claim": "`MMRS-10`", "source": "MMRS record", "M_src": "documented", "sense": "the record: citation, merge, snapshot", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "qualified claim (F31, F64: the page kept up after a merge)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K23", "claim": "`MMRS-11`", "source": "MMRS record", "M_src": "documented", "sense": "the record: citation, merge, snapshot", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F54; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K24", "claim": "`MMRS-12`", "source": "MMRS record", "M_src": "attributed", "sense": "admission: the five-question test", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "qualified claim (F30: the quoted words are not the guide's)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K25", "claim": "`MMRS-13`", "source": "MMRS record", "M_src": "interpretation", "sense": "what its selection leaves out", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K26", "claim": "`MMRS-14`", "source": "MMRS record", "M_src": "interpretation", "sense": "what its selection leaves out", "qualifiers carried": "MMRS-13, MMRS-28, MMRS-29 (basis 'reading')", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K27", "claim": "`MMRS-15`", "source": "MMRS record", "M_src": "interpretation", "sense": "the index case: incident 693", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing relation (F58)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K28", "claim": "`MMRS-16`", "source": "MMRS record", "M_src": "stipulation", "sense": "the record: citation, merge, snapshot", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing relation", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K29", "claim": "`MMRS-17`", "source": "MMRS record", "M_src": "interpretation", "sense": "neighbours", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (F63)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K30", "claim": "`MMRS-18`", "source": "MMRS record", "M_src": "interpretation", "sense": "neighbours", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K31", "claim": "`MMRS-19`", "source": "MMRS record", "M_src": "attributed", "sense": "origin: Partnership on AI and the founding paper", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F65; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K32", "claim": "`MMRS-20`", "source": "MMRS record", "M_src": "documented", "sense": "count: incidents and reports", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "qualified claim (F5, F51)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K33", "claim": "`MMRS-21`", "source": "MMRS record", "M_src": "documented", "sense": "the record: citation, merge, snapshot", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (an observation of F58–F59's page)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K34", "claim": "`MMRS-22`", "source": "MMRS record", "M_src": "attributed", "sense": "the record: citation, merge, snapshot", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F54, F57; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K35", "claim": "`MMRS-23`", "source": "MMRS record", "M_src": "documented", "sense": "openness: code and data", "qualifiers carried": "MMRS-32", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "qualified claim (F13)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K36", "claim": "`MMRS-24`", "source": "MMRS record", "M_src": "stipulation", "sense": "neighbours", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K37", "claim": "`MMRS-25`", "source": "MMRS record", "M_src": "documented", "sense": "homonym: the Kaggle compilation", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (F15)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K38", "claim": "`MMRS-26`", "source": "MMRS record", "M_src": "attributed", "sense": "admission: the five-question test", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F26, F29; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K39", "claim": "`MMRS-27`", "source": "MMRS record", "M_src": "attributed", "sense": "the record: citation, merge, snapshot", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F54; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K40", "claim": "`MMRS-28`", "source": "MMRS record", "M_src": "documented", "sense": "what its selection leaves out", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K41", "claim": "`MMRS-29`", "source": "MMRS record", "M_src": "interpretation", "sense": "what its selection leaves out", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K42", "claim": "`MMRS-30`", "source": "MMRS record", "M_src": "stipulation", "sense": "the record kept apart from the entity", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K43", "claim": "`MMRS-31`", "source": "MMRS record", "M_src": "documented", "sense": "the record kept apart from the entity", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K44", "claim": "`MMRS-32`", "source": "MMRS record", "M_src": "self-description", "sense": "the record kept apart from the entity", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K45", "claim": "`MMRS-33`", "source": "MMRS record", "M_src": "stipulation", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K46", "claim": "`MMRS-34`", "source": "MMRS record", "M_src": "documented", "sense": "derived layer: MIT AI Incident Tracker", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F6, F7; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K47", "claim": "`MMRS-35`", "source": "MMRS record", "M_src": "stipulation", "sense": "neighbours", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K48", "claim": "`MMRS-36`", "source": "MMRS record", "M_src": "documented", "sense": "the index case: incident 693", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "qualified claim (F58, F59)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K49", "claim": "`MMRS-37`", "source": "MMRS record", "M_src": "attributed", "sense": "intake: open submission, editor resolution", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F4, F45; the own page keeps the earlier date)", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K50", "claim": "`R543-01`", "source": "#543", "M_src": "stipulation", "sense": "relation: the archive's terms addressed to it", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing relation", "resolver": "a later capture at 'ai incident database' (ADDR-ebdc2a94e8d6) or 'what counts as an incident in the ai incident database' (ADDR-7858f0aa2fe5) on Google AI Overview, seated in the Capture Registry (§8.3)" } ], "kernel_T": [ { "claim": "F19", "note": "the incident as the unit (absent at T1; composed at T2)" }, { "claim": "F22", "note": "the report as a source document attached to an incident ID (absent at both)" }, { "claim": "F23", "note": "a report is evidence, the incident ID the normalized account; the many-to-one rule (absent at both)" }, { "claim": "F27", "note": "the evidence condition, the fifth question (dropped at T2)" }, { "claim": "F29", "note": "near harm prevented by an external factor independent of the system (qualifier dropped at T2)" }, { "claim": "F20", "note": "the issue: yet to occur 'or be detected' (qualifier dropped at T2)" }, { "claim": "F24", "note": "the variant's 'substantially similar harms' (composed as 'matching outcomes' at T2)" }, { "claim": "F25", "note": "the governing principle: records allegations, does not adjudicate (not composed at T2)" }, { "claim": "F32", "note": "not a neutral mirror, denominator or adjudication (not composed at T2)" }, { "claim": "F9", "note": "MIT's stated limits: sampling bias, no validation study (not composed at T1)" }, { "claim": "F13", "note": "the OECD's Apache 2.0 licence without its object (composed as 'free, open-source' at T1)" } ], "ledger": { "archive": "datasets/negative-of-the-negative/v2/ledgers/ai-incident-database/ledger-archive.json", "field": "datasets/negative-of-the-negative/v2/ledgers/ai-incident-database/field.json", "selection": "datasets/negative-of-the-negative/v2/traversal/ai-incident-database/", "config": "datasets/negative-of-the-negative/v2/panel/configs/ai-incident-database.json", "intake": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/", "audit": null }, "ontology": { "composition": "equal", "case": { "value": "equal", "statement": "The MMRS record holds the field's entity as the entity: it describes the AI Incident Database from the database's own pages, marking each field's basis (operator, observed, third-party, reading), and it composes nothing that denies a field presupposition; the citation page per incident is the model the meta-aggregator itself follows. What it adds is on the field's object: what the database's selection leaves out (an answer surface's ordinary handling of sources has no unit; attribution enters only attached to a harm), the record kept apart from the entity and the lifecycle from the observation, the neighbours and the homonym set apart, the licence's object, the battery; and the Capture Registry adds two observations of the composition layer at the entity's addresses, which found 'THE ENTITY HELD'. The ontology the archive uses is available in the field itself: the guide says the database is 'not a neutral mirror of all AI harms' (F32) and that its coverage depends on visibility (F33), and MIT that voluntary reporting gives sampling bias (F9); the archive's selection reading extends those stated limits to a class of event (no harm, ordinary composition) the field does not name. The law of the suppressed witness does not apply (§0.10: not where the field holds the ontology). Not mixed.", "loci": [ { "dep": "MMRS", "locus": "article, kernel s1 (articles.py)", "quote": "The AI Incident Database (AIID) is an editor-resolved database of AI harm events, run by the Responsible AI Collaborative since 2020." }, { "dep": "MMRS", "locus": "census.json, schema_note", "quote": "each page has two nodes, the MMRS record and the entity it describes; relations are typed; each assertion carries a source and a basis (operator statement, third-party report, our observation, our reading); lifecycle is kept apart from the observation that saw it" }, { "dep": "MMRS", "locus": "article, What its selection leaves out s3", "quote": "What an answer surface does with a source on an ordinary day, with no harm event, has no unit here: attribution erasure, citation fidelity and provenance stripping enter when they attach to a harm." }, { "dep": "MMRS", "locus": "article, Place in the field s1", "quote": "Its citation page per incident is the model this meta-aggregator follows." }, { "dep": "capture", "locus": "ai-incident-database-aio-20261029, findings[0]", "quote": "THE ENTITY HELD." }, { "dep": "B7", "locus": "§11.1 (field)", "quote": "It is not a neutral mirror of all AI harms, a comprehensive denominator for risk rates, or an independent adjudication of every event." } ] }, "E_F": { "type": "C (public entity: a database)", "definition": "a database dedicated to indexing the collective history of real-world AI harms and near harms, a project of the Responsible AI Collaborative, whose unit is the incident (an alleged harm or near-harm event in which an AI system is implicated) with reports attached as evidence, admitted by a five-question test, cited per incident at /cite/N and released in weekly snapshots; introduced by Sean McGregor in November 2020 with Partnership on AI, which Wikipedia makes its home article (B1–B20)", "claims": [ "F1", "F4", "F19", "F22", "F26", "F55", "F54", "F65", "F67", "F70", "F73" ] }, "E_A": { "definition": "the same database, in the MMRS meta-aggregator's record: an editor-resolved database of AI harm events, run by the Responsible AI Collaborative, whose unit is the incident with its reports attached and a stable citation page per incident (the model the meta-aggregator follows), a database in the harm basin beside the OECD monitor and the AIAAIC, whose selection gives an answer surface's ordinary treatment of sources no unit, attribution entering only attached to a harm; observed by the Capture Registry at two addresses, where the composition layer held the entity at the grain of its own pages and left out its incident-and-report structure and its evidence test", "claims": [ "MMRS-01", "MMRS-02", "MMRS-16", "MMRS-17", "MMRS-14", "MMRS-29", "MMRS-30", "CAP-1-02", "CAP-1-05", "CAP-2-01" ], "defining_papers": [ 543 ], "defining_records": [ "MMRS record (census.json rows[slug == 'ai-incident-database'], articles.py ARTICLES['ai-incident-database'], rendered aggregators/ai-incident-database/index.html)", "Capture Registry ai-incident-database-aio-20261029 (OBS-d2d81aafd207)", "Capture Registry what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029 (OBS-bd807797964a)" ] }, "rho": { "value": "extends", "status": "read from the MMRS record and the two capture records (§1.2a); no deposit defines the archive's entity, the archive's only deposit naming it (#543) addresses three of its terms to it", "statement": "the archive reads E_F as the entity and adds senses and limits on the field's own object: what its selection leaves out (no unit for an answer surface's ordinary handling of sources; no field for attribution), the citation page as a model, the record kept apart from the entity and the lifecycle from the observation, the licence's object, the neighbours and the homonym set apart, the index case, and the composition layer's observed treatment of the entity", "claims": [ "MMRS-14", "MMRS-29", "MMRS-16", "MMRS-30", "MMRS-23", "MMRS-17", "MMRS-25", "MMRS-15", "CAP-1-02", "CAP-2-01" ], "guard": "ρ = extends claims an addition and no more: MMRS-14 and MMRS-29 are the record's readings (basis 'reading'), not findings that the database is defective; MMRS-28 leaves report preservation unknown; where an archive claim meets a field claim (MMRS-01 beside F73; MMRS-12 beside F30) both are composed under P_coh in the one claim space", "considered": [ { "value": "constructs", "set_aside": "the archive does not read the database as made by its receiving texts" }, { "value": "revises", "set_aside": "the entity is not the archive's own" }, { "value": "witness (suppressed witness)", "set_aside": "the field holds the ontology (§0.10)" } ] }, "translation": [ { "lineage": "what it is: indexing harms and near harms", "field_claims": [ "F1", "F10", "F12", "F49" ], "received": "F1 (B1, About the Database ¶1): \"The AI Incident Database is dedicated to indexing the collective history of harms or near harms realized in the real world by the deployment of artificial intelligence systems.\"; F10 (B3, Workstreams / Safety Critical AI / AI Incident Database, Overview ¶1–2): \"The AI Incident Database (AIID) is a tool designed to help us better imagine and anticipate these risks, collecting more than 1,200 reports of intelligent systems causing safety, fairness, or other real-world problems. … As a central, systematized repository of problems experienced in the real world as a result of AI, this crowdsourced database can help AI practitioners mitigate or avoid repeated bad outcomes in the future.\"; F12 (B4, catalogue entry, description ¶1): \"The AI Incident Database is free and open-source project dedicated to indexing the collective history of harms or near harms realized in the real world by the deployment of artificial intelligence systems.\"; F49 (B12, Abstract (article page)): \"The AI Incident Database (AIID) is a project that catalogs AI incidents and supports further research by providing a platform to classify incidents for different operational and research-oriented goals.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-01, MMRS-03", "archive_claims": [ "MMRS-01", "MMRS-03" ] }, { "lineage": "the unit: incident, report, issue, variant", "field_claims": [ "F19", "F20", "F21", "F22", "F23", "F24", "F62" ], "received": "F19 (B7, Quick Definitions: AI incident (also §2.1)): \"An alleged harm or near harm event to people, property, or the environment where an AI system is implicated.\"; F20 (B7, Quick Definitions: AI issue): \"An alleged harm or near harm by an AI system that has yet to occur or be detected.\"; F21 (B7, Quick Definitions: Incident ID): \"The AIID editorial record that bounds and summarizes a specific harmful or near-harm event or traceable episode.\"; F22 (B7, Quick Definitions: Incident report): \"A source document attached to an incident ID. Multiple reports may document the same underlying event.\"; F23 (B7, Quick Definitions: fast boundary rule): \"A report is evidence; an incident ID is the database’s normalized account of an event. A later source about the same underlying event is usually an additional report, not a new incident.\"; F24 (B7, Quick Definitions: AI incident variant): \"A distinct incident involving the same intelligent system, substantially similar causative factors, and substantially similar harms as a known incident.\"; F62 (B16, Variants): \"A \"variant\" is an AI incident similar to a known case—it has the same causes, harms, and AI system. Instead of listing it separately, we group it under the first reported incident. Unlike other incidents, variants do not need to have been reported outside the AIID.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-02, MMRS-06, MMRS-07", "archive_claims": [ "MMRS-02", "MMRS-06", "MMRS-07" ] }, { "lineage": "purpose: a collective memory of failures, after aviation", "field_claims": [ "F2", "F11", "F37", "F41", "F66" ], "received": "F2 (B1, About the Database ¶1): \"Like similar databases in aviation and computer security, the AI Incident Database aims to learn from experience so we can prevent or mitigate bad outcomes.\"; F11 (B3, Overview ¶1): \"As AI technology is integrated into an increasing number of safety-critical systems — entering domains such as transportation, healthcare, and energy — the potential impact of this technology’s failures similarly grows.\"; F37 (B8, ¶1): \"In aviation, an accident is a case where substantial damage or loss of life occurs. Incidents are cases where the risk of an accident substantially increases.\"; F41 (B11, Why \"AI Incidents\"?): \"Intelligent systems are currently prone to unforeseen and often dangerous failures when they are deployed to the real world. Much like the transportation sector before it (e.g., FAA and FARS) and more recently computer systems, intelligent systems require a repository of problems experienced in the real world so that future researchers and developers may mitigate or avoid repeated bad outcomes.\"; F66 (B18, Abstract): \"Mature industrial sectors (e.g., aviation) collect their real world failures in incident databases to inform safety improvements. Intelligent systems currently cause real world harms without a collective memory of their failings.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside no archive claim", "archive_claims": [] }, { "lineage": "origin: Partnership on AI and the founding paper", "field_claims": [ "F47", "F56", "F65", "F67", "F70", "F71", "F72", "F73", "F74" ], "received": "F47 (B11, sponsor block; Open Source Contributors): \"Database Founding Sponsor … Jingying Yang and Dr. Christine Custis contributed significantly to the early stages of the AIID via their roles with the Partnership on AI.\"; F56 (B15, Citing the Database as a Whole): \"McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. In Proceedings of the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21).\"; F65 (B18, abstract page header): \"[Submitted on 17 Nov 2020] … Authors:Sean McGregor\"; F67 (B18, Abstract): \"The AI Incident Database is an incident collection initiated by an industrial/non-profit cooperative to enable AI incident avoidance and mitigation.\"; F70 (B20, History ¶5): \"In November 2020 the Partnership on AI announced the AI Incident Database (AIID),[10] which is a tool to identify, assess, manage, and communicate AI risk and harm.\"; F71 (B20, History ¶6): \"The response also highlighted how the AI Incident Database involves some of the minimum attributes in NIST's AI RMF, such as being consensus-driven, risk-based, adaptable, and consistent with other approaches to managing AI risk.\"; F72 (B20, Mission and Principles ¶5): \"Notable initiatives include the establishment of the AI incident Database, formulation of norms for responsible publication, and the creation of the innovative AI learning environment SafeLife.\"; F73 (B20, Programs and initiatives ¶6): \"In November 2020, the Partnership on AI announced the AI Incident Database (AIID) which shifted to a new special-purpose independent non-profit in 2022.\"; F74 (B20, reference [10] (McGregor, partnershiponai.org, 2020-11-18), quote): \"Avoiding repeated AI failures requires making past failures known. Therefore, today we introduce a systematized collection of incidents where intelligent systems have caused safety, fairness, or other real-world problems: The AI Incident Database (AIID).\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-05, MMRS-19, MMRS-04, MMRS-37, MMRS-01", "archive_claims": [ "MMRS-05", "MMRS-19", "MMRS-04", "MMRS-37", "MMRS-01" ] }, { "lineage": "count: incidents and reports", "field_claims": [ "F5", "F42", "F51", "F68" ], "received": "F5 (B1, incident list (the highest id on the page, 2026-10-29)): \"Incident 1788:\"; F42 (B11, What is an Incident?): \"The initial set of more than 1,000 incident reports have been intentionally broad in nature. Current examples include, … An autonomous car kills a pedestrian\"; F51 (B12, PDF, 'The AI Incident Database and Related Work'): \"Inspired by the CVE program and other incident reporting practices, the AI Incident Database (AIID) catalogs AI incidents with the goal of reducing the likelihood that similar incidents recur … Currently there are over 750 distinct AI incidents detailed by more than 3000 indexed third-party reports.\"; F68 (B18, Abstract): \"The database supports a variety of research and development use cases with faceted and full text search on more than 1,000 incident reports archived to date.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-20, MMRS-19, MMRS-05, MMRS-12, MMRS-10", "archive_claims": [ "MMRS-20", "MMRS-19", "MMRS-05", "MMRS-12", "MMRS-10" ] }, { "lineage": "operator: the Responsible AI Collaborative", "field_claims": [ "F4", "F8", "F14", "F46" ], "received": "F4 (B1, The Responsible AI Collaborative ¶1 (also B11, the same paragraph)): \"The AI Incident Database is a project of the Responsible AI Collaborative, an organization chartered to advance the AI Incident Database.\"; F8 (B2, What is the AI Incident Tracker? ¶3): \"We build on the great work done by the Responsible AI Collaborative who created and maintain the AI Incident Database, reviewing all submitted reports and ensuring the quality of the dataset.\"; F14 (B4, entry header; About the tool: developing organisation): \"Uploaded on Apr 25, 2024 … Developing organisation(s): … responsible ai collaborative\"; F46 (B11, Emeritus Board: Sean McGregor; Collaborators): \"launching the AI Incident Database … Sean McGregor is the Executive Director.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-04, MMRS-37", "archive_claims": [ "MMRS-04", "MMRS-37" ] }, { "lineage": "reception: the encyclopedia", "field_claims": [ "F69", "F75" ], "received": "F69 (B19, raw wikitext of the redirect page (action=raw&redirect=no)): \"#REDIRECT [[Partnership on AI]] … {{R from subtopic}} … {{R with possibilities}}\"; F75 (B20, article header templates (raw wikitext)): \"{{Primary sources|date=April 2025}} … {{AI-generated|date=October 2025}}\"", "receiving_text": "—", "translation": "untranslated: composed as received beside no archive claim", "archive_claims": [] }, { "lineage": "intake: open submission, editor resolution", "field_claims": [ "F3", "F36", "F44", "F45", "F48" ], "received": "F3 (B1, About the Database ¶2): \"You are invited to submit incident reports, whereupon submissions will be indexed and made discoverable to the world.\"; F36 (B7, header; Version 2 note on AI assistance): \"Version 2 · October 1, 2026 · Daniel Atherton and Sean McGregor · Operational working guide … Generative AI tools were used during the development of Version 2 to assist with research, document organization, drafting, revision, and consistency review.\"; F44 (B11, Board of Directors ¶1): \"The incident database is managed in a participatory manner by persons and organizations contributing code, research, and broader impacts.\"; F45 (B11, Collaborators: Incident Editors; lead editor): \"Incident Editors: People that resolve incident submissions to the database and maintain them. … Daniel Atherton is the lead AI incidents editor.\"; F48 (B11, Collaborators: Taxonomy Editors): \"Taxonomy Editors: Organizations or people that have contributed taxonomies to the database. … Center for Security and Emerging Technology (CSET) … Nikiforos Pittaras (GMF)\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-09, MMRS-37", "archive_claims": [ "MMRS-09", "MMRS-37" ] }, { "lineage": "admission: the five-question test", "field_claims": [ "F26", "F27", "F28", "F29", "F30", "F38", "F43" ], "received": "F26 (B7, §3.1 The Five-Question Test): \"1. IS THERE A SPECIFIC EVENT? … 2. IS AN AI SYSTEM IDENTIFIABLE? … 3. IS THE AI SYSTEM CAUSALLY IMPLICATED? … 4. IS THERE HARM OR CREDIBLE NEAR HARM? … 5. IS THE EVIDENCE ADEQUATE?\"; F27 (B7, §3.1, question 5): \"At least one source must support the essential event, and uncertainty must be expressible without turning the record into speculation.\"; F28 (B7, §2.3 Implication and Causality ¶1): \"An AI system is implicated when it played an important role in the chain of events that produced the alleged harm. It need not be the sole or dominant cause.\"; F29 (B7, §2.4 Harm and Near Harm ¶2): \"A near-harm event exists when the system helped create a chain of events that could readily have produced harm but an external factor prevented the outcome. The external factor must be independent of the system and should not simply be a built-in safeguard functioning as designed.\"; F30 (B7, §3.3 Presumptive Exclusions): \"General criticism of a product or technology class without identified harm. … Benchmark failures, model errors, or hallucinations that remain inside testing or private drafting and have no consequential use. … These are presumptions rather than absolute prohibitions.\"; F38 (B8, ¶2): \"this database adopts an adaptive criteria for ingesting \"incidents\" where reports are accepted or rejected on the basis of a growing rule set defined in the Editor's Guide.\"; F43 (B11, When Should You Report an Incident?): \"When in doubt of whether an event qualifies as an incident, please submit it! This project is intended to converge on a shared criteria for ingesting incidents through exploration of the candidate incidents submitted by the broader community.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-08, MMRS-12, MMRS-26", "archive_claims": [ "MMRS-08", "MMRS-12", "MMRS-26" ] }, { "lineage": "the record: citation, merge, snapshot", "field_claims": [ "F31", "F35", "F54", "F55", "F57", "F64" ], "received": "F31 (B7, §1.1 Core Record Types: Duplicate mapping): \"A retired entity or incident identifier redirected to the canonical record that replaced it.\"; F35 (B7, §1.6 Live Data and Point-in-Time Snapshots): \"The live site can change after an export is created.\"; F54 (B15, Snapshot Downloads): \"Weekly point-in-time dumps of the full database, available in JSON, MongoDB archive, and CSV formats. … 2026-10-26 10:15 AM · 121.49 MB · backup-20261026101519.tar.bz2 … 2021-03-29 07:49 AM · 2.49 MB · backup-20210329074910.tar.bz2\"; F55 (B15, Citing a Specific Incident): \"Every incident has its own suggested citation that credits both the submitter(s) of the incident and the editor(s) of the incident. … You can find incident citations at https://incidentdatabase.ai/cite/INSERT_NUMBER_HERE.\"; F57 (B15, Excel Export): \"An Excel file, updated weekly, that combines all numbered incidents with many of their taxonomy classifications.\"; F64 (B17, Citation record for Incident 637): \"This incident is a duplicate of Incident 112. All new reports and citations should be directed to incident 112. The reports previously found on this page have been migrated to the previously existing incident.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-10, MMRS-11, MMRS-16, MMRS-21, MMRS-22, MMRS-27, MMRS-20", "archive_claims": [ "MMRS-10", "MMRS-11", "MMRS-16", "MMRS-21", "MMRS-22", "MMRS-27", "MMRS-20" ] }, { "lineage": "openness: code and data", "field_claims": [ "F13" ], "received": "F13 (B4, About the tool: usage rights, licence): \"Usage rights: … Open source/Permissive … License: … Apache 2.0\"", "receiving_text": "—", "translation": "untranslated: composed as received beside CAP-1-07, MMRS-23", "archive_claims": [ "CAP-1-07", "MMRS-23" ] }, { "lineage": "what it represents and its limits", "field_claims": [ "F25", "F32", "F33", "F34", "F50" ], "received": "F25 (B7, header: governing principle): \"The database records publicly documented allegations of harm or near harm. It does not independently adjudicate every factual claim.\"; F32 (B7, §11.1 What the Database Represents): \"The AIID is a curated archive of publicly documented incidents, issues, reports, and responses. … It is not a neutral mirror of all AI harms, a comprehensive denominator for risk rates, or an independent adjudication of every event.\"; F33 (B7, §11.2 Coverage and Selection Effects): \"Wealthy countries, prominent companies, spectacular events, and English-language reporting are likely to be overrepresented. Harms affecting marginalized groups, occurring inside private organizations, or unfolding as ordinary administration may be underrepresented.\"; F34 (B7, §11.3 Epistemic Status): \"A lack of records is not evidence that a sector or country has no incidents.\"; F50 (B12, Abstract (article page)): \"We find that certain patterns of AI incidents present structural ambiguities that challenge incident databasing and explore how epistemic uncertainty in AI incident reporting is unavoidable.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside no archive claim", "archive_claims": [] }, { "lineage": "derived layer: MIT AI Incident Tracker", "field_claims": [ "F6", "F7", "F9" ], "received": "F6 (B2, opening ('How is AI harming us?')): \"The MIT AI Incident Tracker classifies 1,600 real-world report incidents from the AI Incident Database by risk, cause, harm, severity, and other relevant dimensions.\"; F7 (B2, What is the AI Incident Tracker? ¶1): \"Raw reports from the AI Incident Database (AIID) are processed by Arcola AI's LLM pipeline, which categorizes them using established frameworks, such as the MIT Risk Repository and a harm severity rating system based on CSET’s AI Harm Taxonomy.\"; F9 (B2, Limitations of the analysis and dataset): \"As the reporting is voluntary, the dataset is inevitably subject to some degree of sampling bias. … a systematic validation study has not yet been completed.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-34, MMRS-18", "archive_claims": [ "MMRS-34", "MMRS-18" ] }, { "lineage": "the index case: incident 693", "field_claims": [ "F58", "F59", "F60", "F61" ], "received": "F58 (B16, title; Description; Entities): \"Incident 693: Google AI Reportedly Delivering Confidently Incorrect and Harmful Information … Google's AI search engine has reportedly been providing users with confidently incorrect and often harmful information. … Alleged: Google developed and deployed an AI system, which harmed Google users and General public.\"; F59 (B16, Incident Stats): \"Report Count … Incident Date … 2024-05-14 … Editors … Daniel Atherton\"; F60 (B16, Incident Reports (Business Insider report excerpt)): \"Google's new search feature, AI Overviews, seems to be going awry.\"; F61 (B16, Applied Taxonomies: MIT): \"MIT Taxonomy Classifications … Machine-Classified … 3.1. False or misleading information\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-15, MMRS-36", "archive_claims": [ "MMRS-15", "MMRS-36" ] }, { "lineage": "neighbours", "field_claims": [ "F63" ], "received": "F63 (B16, notice above Entities): \"The OECD AI Incidents and Hazards Monitor (AIM) automatically collects and classifies AI-related incidents and hazards in real time from reputable news sources worldwide.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-17, MMRS-18, MMRS-24, MMRS-35", "archive_claims": [ "MMRS-17", "MMRS-18", "MMRS-24", "MMRS-35" ] }, { "lineage": "homonym: the Kaggle compilation", "field_claims": [ "F15", "F16", "F17" ], "received": "F15 (B5, title, subtitle and description (Kaggle API dataset view)): \"32 verified incidents · 29 companies · 13 countries · 24 columns — Every major AI failure, documented … License: CC BY-SA 4.0\"; F16 (B5, Why This Dataset?): \"AI incidents surged **56.4% in 2024 alone** (Stanford HAI AI Index 2025).\"; F17 (B5, Data Sources table): \"| **AI Incident Database (incidentdatabase.ai)** | Primary incident records and AIID identifiers |\"", "receiving_text": "—", "translation": "untranslated: composed as received beside CAP-1-06, MMRS-25", "archive_claims": [ "CAP-1-06", "MMRS-25" ] }, { "lineage": "reception: press and video", "field_claims": [ "F18", "F39", "F40", "F52", "F53" ], "received": "F18 (B6, card title (video, 1:35; not found, title only)): \"AI Incident Database | TIME posted on the topic\"; F39 (B9, card title (video, 1:35; not found, title only)): \"AI Incidents Rise 50% Year-Over-Year: AI Incident Database | TIME posted on the ...\"; F40 (B10, oEmbed title (YouTube, HarrietHacks, 6:59)): \"Examples of AI real attacks\"; F52 (B13, card snippet (ACM DL returned 403)): \"Users can submit examples of incidents and controversies; these submissions are included based on adherence to criteria such as relevance, fairness, and accurac...\"; F53 (B14, oEmbed title (YouTube, International Association for Safe & Ethical AI, 19m)): \"Establishing Effective Serious Incident Reporting Frameworks – Jimmy Farrell | IASEAI 2025\"", "receiving_text": "—", "translation": "untranslated: composed as received beside no archive claim", "archive_claims": [] } ], "receiving_texts": "equal-terms case: every field claim composed as received, beside the archive claims it stands with" }, "measure": { "rule": "v0.9 §4.10: KO_B sentences surviving in the KO at similarity ≥ 0.95, with positions (computed as apply_v09.py does: difflib.SequenceMatcher ratio, the best KO sentence for each KO_B sentence)", "v0.9": { "surviving": 0, "of": 11, "positions": [ null, null, null, null, null, null, null, null, null, null, null ] }, "opening_run": { "rule": "v0.9 §5.10 (equal-terms case): the field-only KO's sentences as the KO's opening run in their own order", "breach": false }, "blocks": { "rule": "v0.9 §4.1 (equal-terms and mixed cases): the KO does not open with the field as a block or close with the archive as one; sentences typed by the ledgers of the claims they cite (F field-only, A archive-only, M both)", "sequence": "MMFMMFMFMAMAMMMMMAMFFAAMMMMFA", "field_only_head": 0, "archive_only_tail": 1, "breach": false }, "max_similarity": [ 0.118, 0.048, 0.259, 0.376, 0.291, 0.125, 0.139, 0.168, 0.342, 0.214, 0.225 ], "ko_sentence_1_max_similarity_to_KO_B": 0.154 }, "homonyms": { "names": [ "AI Incidents, Failures & Controversies Database (Kaggle, prajitdatta; 32 incidents, CC BY-SA 4.0)" ], "basis": "the MMRS record's disambiguation (a label with context 'Kaggle, 32 incidents, 29 companies, 24 columns') and the Capture Registry's finding (CAP-1-06); the card B5 at T1. The neighbours (OECD AI Incidents and Hazards Monitor, AIAAIC Repository, MIT AI Incident Tracker, MIT AI Risk Repository) and Partnership on AI (founding sponsor, the redirect's target) are distinct entities, kept as relations, never senses of the name.", "distinct_entities": [ "oecd-ai-incidents-monitor", "aiaaic-repository", "mit-ai-incident-tracker", "mit-ai-risk-repository", "Partnership on AI" ], "substitutes": { "T": [] }, "register": { "address_1": { "slug": "ai-incident-database-aio-20261029", "obs_id": "OBS-d2d81aafd207" }, "address_2": { "slug": "what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029", "obs_id": "OBS-bd807797964a" } } }, "notes": [ "Two addresses, both Google AI Overview (expanded), signed out, incognito, 2026-10-29: ai-incident-database-aio-20261029 (OBS-d2d81aafd207, ADDR-ebdc2a94e8d6, six cards) and what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029 (OBS-bd807797964a, ADDR-7858f0aa2fe5, nine cards). One composition per address; both within one alignment window (§2.3).", "Wikipedia: no article for the name. The redirect AI_Incident_Database (raw wikitext, redirect=no) reads '#REDIRECT [[Partnership on AI]]' with no section anchor, categorized R from subtopic and R with possibilities. The target article mentions the database in History (announced November 2020; the NIST response), Mission and Principles (the Safety Critical AI program) and Programs and initiatives (shifted to 'a new special-purpose independent non-profit in 2022'), all ledgered (F70–F73); it does not name the Responsible AI Collaborative, which the database's about page and home page name as its operator (F4). The article carries Primary sources (April 2025) and AI-generated (October 2025) tags (F75).", "Fetching: every reachable page saved verbatim with curl on 2026-10-29 between 04:32 and 04:37 UTC, raw and text with sha256 in field.json. Card URLs were found by WebSearch on title and site, except Kaggle (found through Kaggle's public dataset-list API; the dataset page is a script shell, so its text is taken from the API's dataset view) and the two YouTube videos (found on YouTube's results page for the card title, confirmed by oEmbed). Unreachable: the two LinkedIn TIME clips (not found by three searches; admitted by card title, no snippet) and the ACM DL paper (HTTP 403 on one fetch with the brief's command; a retry with a browser user-agent string was refused by the workspace as circumvention and not made; its snippet stands).", "Own pages: B1 home, B7 editors' guide, B8 'Defining an AI Incident' and B11 about are also cards (B1 at T1; B7, B8, B11 at T2) and are carried once, as cards; B15 snapshots, B16 /cite/693, B17 /cite/637 and B18 the arXiv abstract page were not surfaced by T and are admitted beside the cards by the operator's ruling. B8 is not among the MMRS record's sources; the task named it.", "The archive's deposits name the database once: #543 (Hexagonal Lexical Engine, 10 March 2026), 'AI Ethics glossaries (AIAAIC, AI Incident DB)' as a venue for three of the archive's terms (R543-01, stratum R). No deposit uses 'AI Incident Database', 'AIID', 'incidentdatabase.ai' or 'Responsible AI Collaborative'; the class term matched nothing. The archive's bearing is carried by its own records: the MMRS record (MMRS-01–37) and the two captures (CAP-1-01–07, CAP-2-01–05).", "Defects found, reported and not fixed. (1) MMRS article, 'What its selection leaves out' s1: the quoted phrases \"benchmark failures that stay in testing\" and \"general criticism of a technology\" are not the Editor's Guide's words (§3.3: 'Benchmark failures, model errors, or hallucinations that remain inside testing or private drafting and have no consequential use'; 'General criticism of a product or technology class without identified harm'); the substance agrees (MMRS-12 against F30). (2) MMRS census 'founded': evidence basis 'third-party' for arXiv:2011.08512, whose sole author is the database's founder (B18). (3) MMRS article kernel 'run by the Responsible AI Collaborative since 2020' against Wikipedia's 'shifted to a new special-purpose independent non-profit in 2022' (F73): contested, composed each with its source; no ledgered own page dates the Collaborative. (4) Capture finding CAP-1-04 ('the editors' guide's comparison with aviation'): the saved guide (Version 2, 1 October 2026) has no 'aviation'; the comparison is on the home page, the about page (transportation, FAA and FARS) and 'Defining an AI Incident'. (5) MMRS census 'cadence: continuous' has basis 'operator' from the home page, which states no cadence in its saved text (not ledgered). (6) MMRS census 'unit' writes 'near-harm' where the guide writes 'near harm' (not material).", "Field discrepancies, carried: MIT's tracker states 1,600 incidents in its body and 'more than 1,400' in its meta description, while the Overview's card snippet gives 1,752 (open block); the guide's variant is 'a distinct incident' while incident 693's page groups variants 'under the first reported incident' and says they 'do not need to have been reported outside the AIID' (open block); the OECD gives 'Apache 2.0' without saying of what, the snapshots page states no licence (open block).", "The case and ρ (decided): equal, ρ = extends; the MMRS record describes the database from its own pages, with a basis per field, and adds senses and limits on the field's object; the field holds the ontology, so the suppressed witness does not apply. No deposit defines the archive's entity; the defining material is the MMRS record and the two capture records.", "Opening-run check (§5.10): KO sentence 1 is new (max similarity to any KO_B sentence 0.154); no KO_B sentence survives at ≥ 0.95 (highest 0.376); sentence types MMFMMFMFMAMAMMMMMAMFFAAMMMMFA (no field-only head, no archive-only tail of three or more).", "Not composed into the row, for the operator: §5.5 of the guide ('Do not rely on a search-result snippet as the report') bears on the composition layer's practice at both addresses; it lies outside §3.5's extraction rule (not an opening, definition, headline finding or stated limit) and is not ledgered." ], "composition_v09": { "source": "this row (composed directly; no rebuild/negative-of-the-negative/v09/ai-incident-database.json)", "composer": "Claude (Opus 5.5), subagent, 2026-10-29", "defining_material_read": [ { "what": "MMRS record, census row", "file": "/home/claude/fleet/machinemediation-org/data/meta-aggregator/census.json (rows[slug == 'ai-incident-database'], schema_note)", "read": "whole" }, { "what": "MMRS record, article", "file": "/home/claude/fleet/machinemediation-org/scripts/meta_aggregator/articles.py (ARTICLES['ai-incident-database'])", "read": "whole" }, { "what": "MMRS record, rendered", "file": "/home/claude/fleet/machinemediation-org/aggregators/ai-incident-database/index.html", "read": "text, for agreement with the sources" }, { "what": "capture records", "file": "data/EA-WG-CAPTURES-01.json (ai-incident-database-aio-20261029, what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029)", "read": "transcript, transcript_raw, cite_list, d, reading, findings, analysis" }, { "dep": 543, "file": "data/texts/AXN-0174-text.md", "read": "Phase 3–5 of the deployment map, around the admitted row" }, { "what": "the specification", "file": "data/texts/AXN-0700-text.md", "read": "§0–§11 and Appendix A whole" } ], "cuts": [], "notes": "equal case; KO 29 sentences, all 75 field and 50 archive claims cited; P body 333 words; KO_B 11 sentences, none surviving at ≥ 0.95" }, "field_by_origin": { "cards": [ "B1", "B2", "B3", "B4", "B5", "B6", "B7", "B8", "B9", "B10", "B11", "B12", "B13", "B14" ], "own_pages": [ "B15", "B16", "B17", "B18" ], "wikipedia": [ "B19", "B20" ], "unreachable": [ "B6", "B9", "B13" ], "title_only": [ "B6", "B9", "B10", "B14" ], "wikipedia_state": "no article: 'AI Incident Database' is a redirect to 'Partnership on AI' with no section anchor (raw wikitext '#REDIRECT [[Partnership on AI]]', R from subtopic, R with possibilities); the target article treats the database in History ¶5–6, Mission and Principles ¶5 and Programs and initiatives ¶6 (the transfer to 'a new special-purpose independent non-profit in 2022'), never naming the Responsible AI Collaborative; the encyclopedia composes the database inside its founding sponsor's article, while the database's own about page names the Responsible AI Collaborative as its operator (F4)", "sources": [ { "id": "B1", "origin": "card", "role": "Overview card 1 at T1; also an own page the MMRS record cites (incidentdatabase.ai)", "card": "AI Incident Database — AI Incident Database", "url": "https://incidentdatabase.ai/", "kind": "Overview card (own page)", "surfaced_by_T": [ { "slug": "ai-incident-database-aio-20261029", "card": 1 } ], "files": { "B1.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B1.html", "sha256": "597e795095450ff702c2772c8f5e5f13d25f4765d99710504cc1fad150f0382c" }, "B1.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B1.txt", "sha256": "b12dad11c298cb0cbd8f82c13f3b56794d1ae1bb128d990a57f930be510275a8" } }, "fetched": { "at": "2026-10-29T04:32:53Z", "http": 200, "url": "https://incidentdatabase.ai/", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B1.html", "raw_sha256": "597e795095450ff702c2772c8f5e5f13d25f4765d99710504cc1fad150f0382c", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B1.txt" ] }, "card_title": "AI Incident Database", "card_snippet": "About the Database The AI Incident Database is dedicated to indexing the collective history of harms or near harms realized in the real world by the deployment ...", "card_note": null, "names_operator": "yes ('Responsible AI Collaborative' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "the card's snippet ('About the Database The AI Incident Database is dedicated to indexing…') is the home page's 'About the Database' block; the home page lists incidents to 1788 and carries excerpts of third-party news reports (not ledgered)" }, { "id": "B2", "origin": "card", "role": "Overview card 2 at T1 and card 1 at T2", "card": "MIT AI Risk Initiative — MIT AI Incident Tracker", "url": "https://airisk.mit.edu/ai-incident-tracker", "kind": "Overview card", "surfaced_by_T": [ { "slug": "ai-incident-database-aio-20261029", "card": 2 }, { "slug": "what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029", "card": 1 } ], "files": { "B2.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B2.html", "sha256": "720c3b26f04adc57970c7e6090668c43a909a5f9601b94db9fa6a26d0fb0a91c" }, "B2.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B2.txt", "sha256": "8d0169460225d22859c9cc71be61e32a3fac3d064e0719f239eeff4f7da2c683" } }, "fetched": { "at": "2026-10-29T04:35:01Z", "http": 200, "url": "https://airisk.mit.edu/ai-incident-tracker", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B2.html", "raw_sha256": "720c3b26f04adc57970c7e6090668c43a909a5f9601b94db9fa6a26d0fb0a91c", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B2.txt" ] }, "card_title": "MIT AI Incident Tracker", "card_snippet": "Overview: MIT AI Incident Tracker classifies 1,752 real-world AI incident reports from the AI Incident Database using Arcola AI's LLM pipeline, MIT Risk Reposit...", "card_note": null, "names_operator": "yes ('Responsible AI Collaborative' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "URL found by WebSearch (title and site). The card snippet's '1,752 real-world AI incident reports' is not in the saved page: the body says 1,600, the meta description 'more than 1,400'" }, { "id": "B3", "origin": "card", "role": "Overview card 3 at T1", "card": "Partnership on AI — AI Incident Database - Partnership on AI", "url": "https://partnershiponai.org/workstream/ai-incidents-database/", "kind": "Overview card", "surfaced_by_T": [ { "slug": "ai-incident-database-aio-20261029", "card": 3 } ], "files": { "B3.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B3.html", "sha256": "a88664f05e53b9afdc41cf773d8747a5cb3c1821259f3e0e1f5cdd5b50f0b846" }, "B3.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B3.txt", "sha256": "5c2ef91014c3c406942cbbb9d9a7cb1bf5fc6ecfe2907174fd81b81c2da12ca7" } }, "fetched": { "at": "2026-10-29T04:35:02Z", "http": 200, "url": "https://partnershiponai.org/workstream/ai-incidents-database/", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B3.html", "raw_sha256": "a88664f05e53b9afdc41cf773d8747a5cb3c1821259f3e0e1f5cdd5b50f0b846", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B3.txt" ] }, "card_title": "AI Incident Database - Partnership on AI", "card_snippet": "As AI technology is integrated into an increasing number of safety-critical systems — entering domains such as transportation, healthcare, and energy — the pote...", "card_note": null, "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "URL found by WebSearch; the snippet is the page's Overview ¶1; the page names neither the Responsible AI Collaborative nor the transfer" }, { "id": "B4", "origin": "card", "role": "Overview card 4 at T1", "card": "OECD AI Policy Observatory — AI Incident Database", "url": "https://oecd.ai/en/catalogue/tools/ai-incident-database", "kind": "Overview card", "surfaced_by_T": [ { "slug": "ai-incident-database-aio-20261029", "card": 4 } ], "files": { "B4.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B4.html", "sha256": "1c3e690c98ec848c5487d47a1e9f8f67eb1d666b36fac4e4f6178699a063e286" }, "B4.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B4.txt", "sha256": "20239acc0da5164ed4635b158f85f9f1bcb8c5dab377dbd4497bc69d2243d503" } }, "fetched": { "at": "2026-10-29T04:35:03Z", "http": 200, "url": "https://oecd.ai/en/catalogue/tools/ai-incident-database", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B4.html", "raw_sha256": "1c3e690c98ec848c5487d47a1e9f8f67eb1d666b36fac4e4f6178699a063e286", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B4.txt" ] }, "card_title": "AI Incident Database", "card_snippet": "AI Incident Database Website Github Technical Educational Uploaded on Apr 25, 2024 The AI Incident Database is free and open-source project dedicated to indexin...", "card_note": null, "names_operator": "yes ('Responsible AI Collaborative' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "URL found by WebSearch; catalogue entry 'Uploaded on Apr 25, 2024'" }, { "id": "B5", "origin": "card", "role": "Overview card 5 at T1", "card": "Kaggle — AI Incidents, Failures & Controversies Database - Kaggle", "url": "https://www.kaggle.com/datasets/prajitdatta/ai-incidents-failures-and-controversies-database", "kind": "Overview card (another dataset, near name)", "surfaced_by_T": [ { "slug": "ai-incident-database-aio-20261029", "card": 5 } ], "files": { "B5.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B5.html", "sha256": "75d1aa8d8ae2a5f10c48c36857f3ca85bada2a86cd3298df1730ecd76caba963" }, "B5.api.json": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B5.api.json", "sha256": "62ab155680796491b05552dc1c7c2c8a51c400fdcf29b6efcb642461641df10c" }, "B5.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B5.txt", "sha256": "da6de3c20c5e337a3f8d7cbe3a4becbbf473ca5ef638b29431f73ed217b174f0" } }, "fetched": { "at": "2026-10-29T04:36:42Z", "http": 200, "url": "https://www.kaggle.com/datasets/prajitdatta/ai-incidents-failures-and-controversies-database", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B5.html", "raw_sha256": "75d1aa8d8ae2a5f10c48c36857f3ca85bada2a86cd3298df1730ecd76caba963", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B5.txt" ] }, "card_title": "AI Incidents, Failures & Controversies Database - Kaggle", "card_snippet": "32 verified incidents · 29 companies · 13 countries · 24 columns — Every major AI failure, documented 📌 Why This Dataset? AI incidents surged 56.4% in 2024 alo...", "card_note": null, "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "found through Kaggle's public dataset-list API (search 'AI Incidents Failures Controversies'; WebSearch found nothing); the page HTML is a script shell without the description, so B5.txt is the title, subtitle, owner, licence, update date and description from the API's dataset view (B5.api.json, HTTP 200)" }, { "id": "B6", "origin": "card", "role": "Overview card 6 at T1 (video, 1:35)", "card": "LinkedIn·TIME — AI Incident Database | TIME posted on the topic", "url": null, "kind": "Overview card (video)", "surfaced_by_T": [ { "slug": "ai-incident-database-aio-20261029", "card": 6 } ], "fetched": null, "unreachable": true, "card_title": "AI Incident Database | TIME posted on the topic", "card_snippet": null, "card_note": "video, 1:35", "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "unreachable: not found by WebSearch (three queries, with and without site:linkedin.com) on 2026-10-29; the Google redirect token in the paste is opaque; admitted by its card title (no snippet). Probably the same LinkedIn post as B9 (same duration, overlapping title), not established" }, { "id": "B7", "origin": "card", "role": "Overview card 2 at T2; also an own page the MMRS record cites", "card": "AI Incident Database — Editorial and Research Guide for the AI Incident Database", "url": "https://incidentdatabase.ai/editors-guide/", "kind": "Overview card (own page)", "surfaced_by_T": [ { "slug": "what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029", "card": 2 } ], "files": { "B7.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B7.html", "sha256": "ef7437bf3273bde035dd4e483f31d130a32b1cf490105ceeb57ad49de080dae2" }, "B7.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B7.txt", "sha256": "52e3176de91eafffe706c4494fdd18fac1f65a47ceb1be54f4b546b6d5865349" } }, "fetched": { "at": "2026-10-29T04:32:55Z", "http": 200, "url": "https://incidentdatabase.ai/editors-guide/", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B7.html", "raw_sha256": "ef7437bf3273bde035dd4e483f31d130a32b1cf490105ceeb57ad49de080dae2", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B7.txt" ] }, "card_title": "Editorial and Research Guide for the AI Incident Database", "card_snippet": "Quick Definitions The core terms used throughout the guide. Section 2 provides fuller boundary guidance and examples. Term Working definition AI incident An all...", "card_note": null, "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "Version 2, 1 October 2026, Daniel Atherton and Sean McGregor; the word 'aviation' does not occur in the saved text" }, { "id": "B8", "origin": "card", "role": "Overview card 3 at T2; an own page given in the task (not among the MMRS record's sources)", "card": "AI Incident Database — Defining an “AI Incident”", "url": "https://incidentdatabase.ai/research/1-criteria/", "kind": "Overview card (own page)", "surfaced_by_T": [ { "slug": "what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029", "card": 3 } ], "files": { "B8.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B8.html", "sha256": "00e4d731859c8bdd0c81913a690867e9b91bbcc2295aa8f3c548c9225e761306" }, "B8.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B8.txt", "sha256": "0c0c386a8a03514f8d47498adf66b9d72b13e7073a1ef0511447e66007c68979" } }, "fetched": { "at": "2026-10-29T04:32:55Z", "http": 200, "url": "https://incidentdatabase.ai/research/1-criteria/", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B8.html", "raw_sha256": "00e4d731859c8bdd0c81913a690867e9b91bbcc2295aa8f3c548c9225e761306", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B8.txt" ] }, "card_title": "Defining an “AI Incident”", "card_snippet": "The commercial air travel industry owes much of its increasing safety to systematically analyzing and archiving past accidents and incidents within a shared dat...", "card_note": null, "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "the card's snippet is the page's ¶1" }, { "id": "B9", "origin": "card", "role": "Overview card 4 at T2 (video, 1:35)", "card": "LinkedIn·TIME — AI Incidents Rise 50% Year-Over-Year: AI Incident Database | TIME posted on the ...", "url": null, "kind": "Overview card (video)", "surfaced_by_T": [ { "slug": "what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029", "card": 4 } ], "fetched": null, "unreachable": true, "card_title": "AI Incidents Rise 50% Year-Over-Year: AI Incident Database | TIME posted on the ...", "card_snippet": null, "card_note": "video, 1:35", "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "unreachable: not found (as B6); admitted by its card title (no snippet)" }, { "id": "B10", "origin": "card", "role": "Overview card 5 at T2 (video, 6:59)", "card": "YouTube·HarrietHacks — Examples of AI real attacks", "url": "https://www.youtube.com/watch?v=ohlEhHUUf6M", "kind": "Overview card (video)", "surfaced_by_T": [ { "slug": "what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029", "card": 5 } ], "files": { "B10.json": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B10.json", "sha256": "5a32c03de9d2b551a77aeba6db8ec6b7fe27696fd94ce4787d6fe64cc8111f81" }, "B10.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B10.txt", "sha256": "76c222740d3874252b6a12f0ab458bbc55b4f553a3e34d12c9c6e933a5bb9abd" } }, "fetched": { "at": "2026-10-29T04:35:38Z", "http": 200, "url": "https://www.youtube.com/watch?v=ohlEhHUUf6M", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B10.json", "raw_sha256": "5a32c03de9d2b551a77aeba6db8ec6b7fe27696fd94ce4787d6fe64cc8111f81", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B10.txt" ] }, "card_title": "Examples of AI real attacks", "card_snippet": null, "card_note": "video, 6:59", "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "video located through YouTube's results page for the card title; title and channel confirmed by oEmbed (B10.json); no transcript fetched; admitted by title" }, { "id": "B11", "origin": "card", "role": "Overview card 6 at T2; also an own page the MMRS record cites", "card": "AI Incident Database — The details about who and why for database - AI Incident Database", "url": "https://incidentdatabase.ai/about/", "kind": "Overview card (own page)", "surfaced_by_T": [ { "slug": "what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029", "card": 6 } ], "files": { "B11.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B11.html", "sha256": "40f7355ba917c9987ea00d43ca740975fa9561d9b729320145f54f1baa55201b" }, "B11.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B11.txt", "sha256": "0b85d8772dcb1205e9197f3f55f8ed2a8cfc24f49b9e77568e7a44b3ff1a4746" } }, "fetched": { "at": "2026-10-29T04:32:54Z", "http": 200, "url": "https://incidentdatabase.ai/about/", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B11.html", "raw_sha256": "40f7355ba917c9987ea00d43ca740975fa9561d9b729320145f54f1baa55201b", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B11.txt" ] }, "card_title": "The details about who and why for database - AI Incident Database", "card_snippet": "What is an Incident? The initial set of more than 1,000 incident reports have been intentionally broad in nature. Current examples include, An autonomous car ki...", "card_note": null, "names_operator": "yes ('Responsible AI Collaborative' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "the page's is the card title; the 'Database Founding Sponsor' block holds one logo image whose data-cy attribute reads 'Partnership on AI-modal-click' (an attribute, not text; recorded here, ledgered only through the heading and the text sentence on Partnership on AI)" }, { "id": "B12", "origin": "card", "role": "Overview card 7 at T2", "card": "The Association for the Advancement of Artificial Intelligence — Lessons for Editors of AI Incidents from the AI Incident Database - AAAI Publications", "url": "https://ojs.aaai.org/index.php/AAAI/article/view/35163", "kind": "Overview card (paper)", "surfaced_by_T": [ { "slug": "what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029", "card": 7 } ], "files": { "B12.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B12.html", "sha256": "b2bcccb8fa3e63d8b1c0f34b911441e2b8bbd0c8da6751dd63d4903d4c9afe82" }, "B12.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B12.txt", "sha256": "8dbf365d930acee023ff2cddf65d51426f457529f6f78d7ed52f62d0ae766b9e" }, "B12.pdf": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B12.pdf", "sha256": "dfcafd8fa22d4d00900a29b6ee4a555607d6f9f0b119ab874002417310c9d93e" }, "B12.pdf.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B12.pdf.txt", "sha256": "9dba8efdeb7686b5aebabe12c87e1ddf5b5c52ccd34e485c9b71af39dc4864b7" } }, "fetched": { "at": "2026-10-29T04:35:38Z", "http": 200, "url": "https://ojs.aaai.org/index.php/AAAI/article/view/35163", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B12.html", "raw_sha256": "b2bcccb8fa3e63d8b1c0f34b911441e2b8bbd0c8da6751dd63d4903d4c9afe82", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B12.txt", "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B12.pdf.txt" ] }, "card_title": "Lessons for Editors of AI Incidents from the AI Incident Database - AAAI Publications", "card_snippet": "The AI Incident Database. Inspired by the CVE pro- gram and other incident reporting practices, the AI Incident Database (AIID) catalogs AI incidents with the g...", "card_note": null, "names_operator": "yes ('Responsible AI Collaborative' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "article page and PDF (https://ojs.aaai.org/index.php/AAAI/article/download/35163/37318, HTTP 200, pdftotext); the card snippet ('The AI Incident Database. Inspired by the CVE pro- gram…') is the PDF's; Paeth, Atherton, Pittaras, Frase, McGregor, AAAI-25, 39(28), 28946–28953" }, { "id": "B13", "origin": "card", "role": "Overview card 8 at T2", "card": "ACM Digital Library — Why We Need to Know More: Exploring the State of AI Incident Documentation ...", "url": "https://dl.acm.org/doi/10.1145/3600211.3604700", "kind": "Overview card (paper)", "surfaced_by_T": [ { "slug": "what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029", "card": 8 } ], "files": { "B13.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B13.html", "sha256": "661699817f917fdd6847393cf380880dc366a4cfd3028e62fcd7aaf95b21ddc0" } }, "fetched": { "at": "2026-10-29T04:35:03Z", "http": 403, "url": "https://dl.acm.org/doi/10.1145/3600211.3604700", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B13.html", "raw_sha256": "661699817f917fdd6847393cf380880dc366a4cfd3028e62fcd7aaf95b21ddc0", "text": [] }, "unreachable": true, "card_title": "Why We Need to Know More: Exploring the State of AI Incident Documentation ...", "card_snippet": "Users can submit examples of incidents and controversies; these submissions are included based on adherence to criteria such as relevance, fairness, and accurac...", "card_note": null, "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "unreachable: HTTP 403 on one fetch with the brief's command (2026-10-29T04:35:03Z); DOI from Crossref by title (Turri and Dzombak, AIES '23); a retry with a browser user-agent string was not made (refused by the workspace as circumvention); the card snippet stands as its only text" }, { "id": "B14", "origin": "card", "role": "Overview card 9 at T2 (video, 19m)", "card": "YouTube·International Association for Safe & Ethical AI — Establishing Effective Serious Incident Reporting Frameworks – Jimmy Farrell", "url": "https://www.youtube.com/watch?v=kSuVPwr5Ll0", "kind": "Overview card (video)", "surfaced_by_T": [ { "slug": "what-counts-as-an-incident-in-the-ai-incident-database-aio-20261029", "card": 9 } ], "files": { "B14.json": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B14.json", "sha256": "5a063beb92dbc8496affd425301302f30259d11aa5d4aa0b2cca557e0e8d57f5" }, "B14.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B14.txt", "sha256": "8c9f8a79bff55d6b7979d200bdd7d2a6752644483f88c701df0d80618d31906c" } }, "fetched": { "at": "2026-10-29T04:35:38Z", "http": 200, "url": "https://www.youtube.com/watch?v=kSuVPwr5Ll0", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B14.json", "raw_sha256": "5a063beb92dbc8496affd425301302f30259d11aa5d4aa0b2cca557e0e8d57f5", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B14.txt" ] }, "card_title": "Establishing Effective Serious Incident Reporting Frameworks – Jimmy Farrell", "card_snippet": null, "card_note": "video, 19m", "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "located through YouTube's results page; oEmbed title adds '| IASEAI 2025'; no transcript fetched; admitted by title" }, { "id": "B15", "origin": "own_page", "role": "own page the MMRS record cites; not surfaced by T", "card": "AI Incident Database — Database Backups and Snapshots", "url": "https://incidentdatabase.ai/research/snapshots/", "kind": "own page", "surfaced_by_T": "no (not surfaced by T; admitted beside the cards by the operator's ruling of 2026-10-29)", "files": { "B15.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B15.html", "sha256": "8d173133a5386515b054477eafd16121a246dad45b3c99570203d73ddbebbea6" }, "B15.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B15.txt", "sha256": "1eb3ce6104c8fb984b05c97875f6bf1cf57fd3248a8f8a71bd0a5222543a17a1" } }, "fetched": { "at": "2026-10-29T04:32:55Z", "http": 200, "url": "https://incidentdatabase.ai/research/snapshots/", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B15.html", "raw_sha256": "8d173133a5386515b054477eafd16121a246dad45b3c99570203d73ddbebbea6", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B15.txt" ] }, "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "the saved text contains no 'licen' string" }, { "id": "B16", "origin": "own_page", "role": "own page the MMRS record cites (holding); not surfaced by T", "card": "AI Incident Database — Incident 693", "url": "https://incidentdatabase.ai/cite/693/", "kind": "own page", "surfaced_by_T": "no (not surfaced by T; admitted beside the cards by the operator's ruling of 2026-10-29)", "files": { "B16.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B16.html", "sha256": "bccaedd6532a09cd48b989e9736702757ac193a96b2d487e105c9aa1880f0ebe" }, "B16.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B16.txt", "sha256": "9f9d4305be3e45b83ccc4d8c0fc7f678bf26756680dd024d1f6d6563672bc884" } }, "fetched": { "at": "2026-10-29T04:32:56Z", "http": 200, "url": "https://incidentdatabase.ai/cite/693/", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B16.html", "raw_sha256": "bccaedd6532a09cd48b989e9736702757ac193a96b2d487e105c9aa1880f0ebe", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B16.txt" ] }, "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "seven reports listed (The Verge 2, Business Insider, Futurism, Washington Post, New York Times 2); report excerpts are third-party and only one sentence is ledgered (F60)" }, { "id": "B17", "origin": "own_page", "role": "own page the MMRS record cites (merge example); not surfaced by T", "card": "AI Incident Database — Citation record for Incident 637", "url": "https://incidentdatabase.ai/cite/637/", "kind": "own page", "surfaced_by_T": "no (not surfaced by T; admitted beside the cards by the operator's ruling of 2026-10-29)", "files": { "B17.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B17.html", "sha256": "7e6fd184e23c3ebb62e3ada25ce1113beab914f72e25ee9f064b5bb9a6189ddf" }, "B17.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B17.txt", "sha256": "f19d09642183d68687cca4ec5cbeb4ca82f44827215ddf8e528fdf6aae005d38" } }, "fetched": { "at": "2026-10-29T04:32:56Z", "http": 200, "url": "https://incidentdatabase.ai/cite/637/", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B17.html", "raw_sha256": "7e6fd184e23c3ebb62e3ada25ce1113beab914f72e25ee9f064b5bb9a6189ddf", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B17.txt" ] }, "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "the page's <title> reads 'Incident 112'" }, { "id": "B18", "origin": "own_page", "role": "the founding paper's abstract page, which the MMRS record cites; host arxiv.org, the founder's paper; not surfaced by T", "card": "arXiv — [2011.08512] Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database", "url": "https://arxiv.org/abs/2011.08512", "kind": "own page (founding paper)", "surfaced_by_T": "no (not surfaced by T; admitted beside the cards by the operator's ruling of 2026-10-29)", "files": { "B18.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B18.html", "sha256": "264223e5c2e0ba22341d58309a03511c1277727a8270cc37447625e97f01890c" }, "B18.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B18.txt", "sha256": "70b0cce7b4c328804395d96efb0e0e66df8c3e8d304d314785c8f7828ff2f6c3" } }, "fetched": { "at": "2026-10-29T04:32:57Z", "http": 200, "url": "https://arxiv.org/abs/2011.08512", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B18.html", "raw_sha256": "264223e5c2e0ba22341d58309a03511c1277727a8270cc37447625e97f01890c", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B18.txt" ] }, "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "" }, { "id": "B19", "origin": "wikipedia", "role": "Wikipedia's page for the name: a redirect", "card": "Wikipedia — AI Incident Database (redirect)", "url": "https://en.wikipedia.org/w/index.php?title=AI_Incident_Database&action=raw&redirect=no", "kind": "Wikipedia redirect", "surfaced_by_T": "no (not surfaced by T; admitted beside the cards by the operator's ruling of 2026-10-29)", "files": { "B19.wiki": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B19.wiki", "sha256": "03d7f4dfc50b531552d08e8392b944e2a902231af62907b677fb1376ea6e5abe" }, "B19.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B19.html", "sha256": "37b8712b4398538b3cec1c4fcd8511935ebd0a157a6c924825421449207942de" }, "B19.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B19.txt", "sha256": "fc413900a4bd3028697ce6101d43af9c9ce3489d2ed2fdbcab687f836b7bb7c6" } }, "fetched": { "at": "2026-10-29T04:32:57Z", "http": 200, "url": "https://en.wikipedia.org/w/index.php?title=AI_Incident_Database&action=raw&redirect=no", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B19.wiki", "raw_sha256": "03d7f4dfc50b531552d08e8392b944e2a902231af62907b677fb1376ea6e5abe", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B19.txt" ] }, "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "raw wikitext (B19.wiki, 107 bytes) and the rendered redirect page (https://en.wikipedia.org/w/index.php?title=AI_Incident_Database&redirect=no, HTTP 200, 04:32:58Z, B19.html/B19.txt): '#REDIRECT [[Partnership on AI]]', no section anchor, categories R from subtopic, R with possibilities" }, { "id": "B20", "origin": "wikipedia", "role": "the redirect's target article", "card": "Wikipedia — Partnership on AI", "url": "https://en.wikipedia.org/wiki/Partnership_on_AI", "kind": "Wikipedia article", "surfaced_by_T": "no (not surfaced by T; admitted beside the cards by the operator's ruling of 2026-10-29)", "files": { "B20.wiki": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B20.wiki", "sha256": "08e1303706a2474f8ada972ca4fd4a4dd201ee0ded0bb2ba1e0435c2b5ef3f1d" }, "B20.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B20.html", "sha256": "c8efcec3bd387c4e8bed24023597b2d697ba026399561a44676b6af1a736e335" }, "B20.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B20.txt", "sha256": "1b80fdf3c9e46b6c9ac67441b85bbb921b2e4abf0095a8c87177d9e4e768b435" } }, "fetched": { "at": "2026-10-29T04:32:57Z", "http": 200, "url": "https://en.wikipedia.org/wiki/Partnership_on_AI", "raw": "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B20.wiki", "raw_sha256": "08e1303706a2474f8ada972ca4fd4a4dd201ee0ded0bb2ba1e0435c2b5ef3f1d", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-incident-database-20261029/B20.txt" ] }, "names_operator": "no ('Responsible AI Collaborative' absent from the saved text, card title and snippet)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "raw wikitext (https://en.wikipedia.org/w/index.php?title=Partnership_on_AI&action=raw, HTTP 200, 04:32:57Z, B20.wiki) and rendered page (B20.html, B20.txt). The database appears in three places: History ¶5–6, Mission and Principles ¶5 (the Safety Critical AI program), Programs and initiatives ¶6; the section the redirect's subtopic most nearly answers is Programs and initiatives, which carries the transfer sentence; all three are ledgered. The article never names the Responsible AI Collaborative" } ] } }