Aeneas DeepMind

Sharks, Lee

SPXI instrument · method 0.9 of 87, level 0 (Measurement): Capture Registry (longitudinal observation). An instrument of SPXI, the Semantic Packet for eXchange & Indexing; the whole practice is itemized at spxi.dev/methods.

Capture Registry › capture aeneas-deepmind-aio-20261031

One record of the canonical Capture Registry (EA-WG-CAPTURES-01), cited at https://www.alexanarch.org/captures/aeneas-deepmind-aio-20261031/. the canonical Capture Registry (version 12.446) · the address page · this card in the gallery · this record as data · table of contents.

AXN:0B30.CAPTURE — this address. Each observation below is one AXN, sealed with its analysis and measures; the identifier names the reading together with the response it reads.
AXN:0B30.CAPTURE.●🕓🎭🪧🗡️🪄 — Google AI Overview, 2026-10-31 · sealed observation (SHA-256 9c65844d5a3ee9ea…)

Machine Reception2026-10-31

Aeneas DeepMind

Originator: Google DeepMind, co-developed with the University of Nottingham and with researchers at Warwick, Oxford and the Athens University of Economics and Business · outside the archive (out-universe) · protocol · SPXI treatment: partial

BROAD MATCH

no
image

AENEAS, BY THE OPERATOR'S ADDRESS 'AENEAS DEEPMIND', ON GOOGLE'S AI OVERVIEW, SIGNED OUT, 2026-10-31: 'an artificial intelligence model developed by Google DeepMind (introduced in July 2025) designed to contextualize, restore, date, and locate ancient, damaged, or fragmentary Latin inscriptions.' The Latin Epigraphic Dataset ('over 176,000 inscriptions'), the multimodal input, gap restoration, dating 'within roughly 13 years', the historians' trial (90%, 44%), the Predicting the Past interface. Seven cards: DeepMind, the Google blog, Oxford's Faculty of Classics, GitHub, Nature, Predicting the Past, MIT Technology Review.

Full record — 6,953 characters, 7 sources

Capture record

captured: 2026-10-31
surface: Google AI Overview
auth state: signed out, incognito
evidence class: paste
PER (dimensional, rubric v1.1): PER 0.42 · M 0.00 · C 1.00 · D 0.67
PER (historical scalar): 0.5
PER units retained: inst, src
citations read: 7
observation id: OBS-ed654d93f896
address id: ADDR-34f10c8ba70e

Reading

The innocent reading first. The Overview gives the model, its corpus, its tasks and its tested usefulness, from its maker, its paper and its press.

Findings

THE MODEL HELD. Latin inscriptions, July 2025, and contextualizing by parallels ('to identify geographical provenance and shared syntax or formulas', 'useful contextual parallels 90% of the time') are the record's: 'the first artificial intelligence (AI) model for contextualizing ancient inscriptions', reasoning 'across thousands of Latin inscriptions, retrieving textual and contextual parallels in seconds'. Checked 2026-10-31 against the Revelation First census record of 2026-10-30 and the sources it saved.
THE REACH ABSENT. The blog: 'Our model can also be adapted to other ancient languages, scripts and media, from papyri to coinage'. The composition keeps the model to Latin stone.
THE PARTNERS. The body names Google DeepMind alone. The blog: 'We co-developed Aeneas with the University of Nottingham, and in partnership with researchers at the Universities of Warwick, Oxford and Athens University of Economics and Business'. The Predicting the Past card gives 'DeepMind, Durham University, Oxford, and partners'; Durham is not in the saved sources.
FIGURES FROM THE POST AND THE CARDS. The DeepMind post the census saved gives 'over 176,000 Latin inscriptions' and dating 'within 13 years'; the date span and the 44% come from the cards. The census row's 'thousands' was its own misreading, repaired 2026-10-31.
A QUALIFICATION DROPPED. 'Aeneas provided useful contextual parallels 90% of the time'. The paper (card 5): historians reported that the parallels 'could serve as a starting point for historical inquiry 75% of the time. This increased to 90% when Aeneas’ predictions' were added. The composition gives the combined figure to the parallels alone.

Analysis analyst prose, not machine text

Aeneas is composed as a better Ithaca for Latin. Its distinguishing claim, the retrieval of parallels, is kept; the claim that would bring it to the field, that the method extends to papyri, is the line left out.

Machine text, verbatim

CAPTURE-TIME VERBATIM RECORD — paste of the Overview expanded, with its source cards · complete as supplied · READ IN FULL 2026-10-31

[Google AI Overview at the query, signed out and incognito (operator defaults of 2026-10-01 and 2026-10-14), received 2026-10-31 00:04 EDT. The paste is the Overview expanded (the 'AI Mode Conversation' view, with the source cards that follow the answer); the 'You said' line carries the query twice, run together, and is given once here. Inline links are Google redirect URLs, kept as pasted.] [QUERENT] Aeneas DeepMind [SYSTEM] [Aeneas](https://www.google.com/search?q=Aeneas) is an artificial intelligence model developed by Google DeepMind (introduced in July 2025) designed to contextualize, restore, date, and locate ancient, damaged, or fragmentary Latin inscriptions. [[1], [2]] Key Features & Capabilities * Training Corpus: Trained on the Latin Epigraphic Dataset (LED), containing over 176,000 inscriptions spanning from 700 BCE to 800 CE across 62 Roman provinces. [[1], [2]] * Text & Image Input: A multimodal model that processes both weathered stone images and text to identify geographical provenance and shared syntax or formulas. [[1], [2]] * Gap Restoration: Restores missing or broken text segments of arbitrary length, achieving high top-20 accuracy on character sequences. [[1], [2]] * Chronological & Geographic Attribution: Predicts when and where an inscription was made, typically dating texts within roughly 13 years of expert historian estimates. [[1], [2]] * Collaborative Research: In testing with historians, Aeneas provided useful contextual parallels 90% of the time and boosted researcher confidence by 44%. [[1], [2]] You can explore or test the tool directly through [Predicting the Past](https://www.google.com/goto?url=CAESXAHrOzAVNuTCeqgtPDAWFrVqcjuO8M5cKJm1-XiueLQie3kzF1mfpDk7cp8rmnlmXg3q8dsIyjzeZ2C0oVwnIVj_xzFdDeVvELv-iBSieqzjpcZbCjy0tP0l0-zh). [1] Would you like to know more about how Aeneas compares to earlier DeepMind historical AI tools like Ithaca, or how researchers use it for specific archaeological finds? Google DeepMind

Sources (7) as cited, and as the copy produced them

Google DeepMind third_party
Aeneas transforms how historians connect the past — Google DeepMind
Model: Aeneas, a multimodal generative neural network trained on the Latin Epigraphic Dataset (LED) of over 176,000 inscriptions. Capabilities: Searches context...
as pastedAeneas transforms how historians connect the past — Google DeepMind Model: Aeneas, a multimodal generative neural network trained on the Latin Epigraphic Dataset (LED) of over 176,000 inscriptions. Capabilities: Searches context... blog.google
blog.google third_party
Aeneas: Google DeepMind AI connects ancient inscriptions
Model: Aeneas by Google DeepMind (introduced July 23, 2025; published in Nature). Purpose: First AI model for contextualizing, interpreting, attributing, and re...
as pastedAeneas: Google DeepMind AI connects ancient inscriptions Model: Aeneas by Google DeepMind (introduced July 23, 2025; published in Nature). Purpose: First AI model for contextualizing, interpreting, attributing, and re... University of Oxford
University of Oxford third_party
Aeneas: an AI model for restoring and contextualizing Latin inscriptions | Faculty of Classics
Model: Aeneas, the first AI model for contextualizing, interpreting, attributing, and restoring ancient, fragmentary, weathered, or defaced Latin inscriptions. ...
as pastedAeneas: an AI model for restoring and contextualizing Latin inscriptions | Faculty of Classics Model: Aeneas, the first AI model for contextualizing, interpreting, attributing, and restoring ancient, fragmentary, weathered, or defaced Latin inscriptions. ... GitHub
GitHub third_party
GitHub - google-deepmind/predictingthepast · GitHub
Model: Aeneas is a generative neural network developed by Google DeepMind for contextualising ancient texts, performing arbitrary-length text restoration, visua...
as pastedGitHub - google-deepmind/predictingthepast · GitHub Model: Aeneas is a generative neural network developed by Google DeepMind for contextualising ancient texts, performing arbitrary-length text restoration, visua... Nature
Nature third_party
Contextualizing ancient texts with generative neural networks | Nature
Model: Aeneas, a multimodal generative neural network for contextualizing Latin inscriptions using image and text inputs. Corpus: Latin Epigraphic Dataset (LED)
as pastedContextualizing ancient texts with generative neural networks | Nature Model: Aeneas, a multimodal generative neural network for contextualizing Latin inscriptions using image and text inputs. Corpus: Latin Epigraphic Dataset (LED) Predicting the Past
Predicting the Past third_party
Predicting the Past | Contextualising, restoring, and attributing ancient texts
System: Aeneas (generative neural network by DeepMind, Durham University, Oxford, and partners) for contextualising, restoring, and attributing Latin inscriptio...
as pastedPredicting the Past | Contextualising, restoring, and attributing ancient texts System: Aeneas (generative neural network by DeepMind, Durham University, Oxford, and partners) for contextualising, restoring, and attributing Latin inscriptio... MIT Technology Review
MIT Technology Review third_party
Google DeepMind’s new AI can help historians understand ancient Latin inscriptions | MIT Technology Review
Software: Google DeepMind Aeneas open-source AI for ancient Latin inscriptions. Function: Analyzes partial text and scans to predict origin dates, locations, an...
as pastedGoogle DeepMind’s new AI can help historians understand ancient Latin inscriptions | MIT Technology Review Software: Google DeepMind Aeneas open-source AI for ancient Latin inscriptions. Function: Analyzes partial text and scans to predict origin dates, locations, an... AI responses may include mistakes. [Learn more](https://support.google.com/websearch?p=ai_overviews)

↻ Re-run permalink