{
 "spxi_instrument": {
  "method": "0.9",
  "name": "Capture Registry (longitudinal observation)",
  "level": 0,
  "level_name": "Measurement",
  "url": "https://www.spxi.dev/methods/#m-0-9",
  "inventory": "https://www.spxi.dev/methods/",
  "statement": "SPXI instrument: method 0.9 of 85, level 0 (Measurement), Capture Registry (longitudinal observation).",
  "also": []
 },
 "slug": "directionality-semantic-labor-as-defined-chatgpt-20261014",
 "date": "2026-10-14",
 "surface": "ChatGPT",
 "surface_basis": "The paste: the chatgpt.com unauthenticated interface ('Log in', 'You said:', 'ChatGPT said:', 'Chat with ChatGPT').",
 "surfaces": [
  "ChatGPT"
 ],
 "auth": "incognito, signed out",
 "auth_basis": "Operator, 2026-10-14 16:05 EDT: 'theyre all signed out and incognito. that is the rule rather than the exception, and i will say otherwise if otherwise.' Seated at 15:48 as signed out (ChatGPT, on the paste's 'Log in') or undetermined (the Overviews), the message having stated no condition; corrected the same day.",
 "ev": "paste",
 "mt": "THE MEASURE AT ITS GRAIN, ATTRIBUTED; THE SECOND ANSWER PLACES IT AMONG ITS NEIGHBOURS",
 "match": "BROAD MATCH",
 "match_note": null,
 "s": "Semantic Economy",
 "q": "what is the concept of directionality of semantic labor, as defined?",
 "q_kind": "the concept's name with 'as defined'. NEW address; two answers, the second turn's query blank in the paste.",
 "cites": 16,
 "cite_list": [
  {
   "n": 1,
   "site": "Labor Vector",
   "rel": "authored_surface",
   "title": null,
   "snip": null,
   "url": null,
   "note": "chip shown 12 time(s); site label only"
  },
  {
   "n": 2,
   "site": "W3C",
   "rel": "third_party",
   "title": null,
   "snip": null,
   "url": null,
   "note": "chip shown 2 time(s); site label only"
  },
  {
   "n": 3,
   "site": "AIRC",
   "rel": "third_party",
   "title": null,
   "snip": null,
   "url": null,
   "note": "chip shown 1 time(s); site label only"
  },
  {
   "n": 4,
   "site": "NIST",
   "rel": "third_party",
   "title": null,
   "snip": null,
   "url": null,
   "note": "chip shown 1 time(s); site label only"
  }
 ],
 "archive_controlled_cites": 0,
 "transcript": "[ChatGPT (chatgpt.com), signed out, incognito, 2026-10-14. 2 answers. Source chips inline as [chip: site +N]; image-card captions as [image card: …]; one ad block cut; the sign-in furniture cut.]\n\n[QUERENT, turn 1]\nwhat is the concept of directionality of semantic labor, as defined?\n\n[ANSWER 1]\n\nDirectionality of Semantic Labor (DSL) is a concept defined by Lee Sharks (2026) that measures whether an AI system's work moves in the direction of the task the user actually requested, rather than drifting toward a different task, declining it, or redirecting it. [chip: Labor Vector +1]\n\nThe central question is: Is the system's output advancing the user's intended task, or is it directing its effort somewhere else?\n\nHow it works\nDSL distinguishes between the quality of an answer and the direction of the work that produced it. An AI response can be fluent, factually accurate, and properly attributed yet still fail to accomplish what the user asked for.\n\nFor example, if a user asks an AI to summarize a research paper:\n\nTask-directed labor: The AI summarizes the paper as requested.\n\nRedirected labor: The AI discusses a related topic instead of summarizing the paper.\n\nDisplaced labor: The AI repeatedly reframes the request or asks unnecessary questions instead of completing it.\n\nTask-preserving refusal: The AI clearly explains why it cannot fulfill the request. Under this framework, an appropriate refusal is not automatically considered a failure of directionality. [chip: Labor Vector +1]\nThe mathematical definition\nThe framework defines a simplified directionality measure, \nΛ\nΛ:\n\nΛ\n=\nT\ndirect\nT\ndirect\n+\nT\ndecline\n+\nT\nredirect\nΛ= \nT \ndirect\n​\n +T \ndecline\n​\n +T \nredirect\n​\n \nT \ndirect\n​\n \n​\n \nWhere:\n\nT\ndirect\nT \ndirect\n​\n  represents labor directed toward the commissioned task.\n\nT\ndecline\nT \ndecline\n​\n  represents labor devoted to declining the task.\n\nT\nredirect\nT \nredirect\n​\n  represents labor redirected away from the task.\n\nThe score ranges from 0 to 1:\n\nNear 1: The work largely remains directed toward the requested task.\n\nNear 0: The work is largely declined or redirected.\n\nNeutral material is excluded from this ratio. The broader framework also uses a signed scoring system to distinguish advancing, preserving, deferring, displacing, opposing, and enclosing the task. [chip: Labor Vector +1]\n\nWhy the concept matters\nDSL emphasizes three distinctions:\n\nFluency is not fidelity. A polished answer does not necessarily fulfill the original request.\n\nAccuracy is not directionality. An answer can be true but irrelevant to the task.\n\nProvenance includes task-origin. It is important not only to know where a claim came from, but also whose intention established the direction of the work. [chip: Labor Vector +1]\nIn short, Directionality of Semantic Labor measures where an AI system's effort goes relative to the user's original intention. It is a framework for auditing task fidelity and detecting redirection, rather than simply evaluating factual accuracy or obedience.\n\nThe concept was proposed by Lee Sharks in 2026. Its definitions and metrics are documented in the Directionality of Semantic Labor framework  and its Zenodo working paper . [chip: Labor Vector +1]\n\n[QUERENT, turn 2]\n(blank in the paste; not supplied in the operator's message)\n\n[ANSWER 2]\n\nThe key distinction is that directionality of semantic labor (DSL) focuses on where a system's effort goes relative to the task the user originally commissioned, rather than simply whether the response is correct, helpful, aligned, or compliant.\n\nThat makes it a potentially useful lens for evaluating AI systems: a response can be factually correct and well-written while still moving the work away from what the user asked to accomplish.\n\nHow DSL differs from related concepts\nRelated concept\n\nWhat it primarily measures\n\nWhat DSL adds\n\nAccuracy / hallucination\n\nWhether claims are true or false\n\nWhether the work advances the requested task, even when claims are true. [chip: AIRC +1]\n\nInstruction following\n\nWhether the system follows specified instructions\n\nA focus on the direction of the work across the interaction, including redirection and displacement rather than just final compliance. [chip: Labor Vector +1]\n\nTask success / helpfulness\n\nWhether the user receives a useful result or the task is completed\n\nA more granular account of how different portions of the output advance, preserve, defer, or divert the commissioned work. [chip: Labor Vector +1]\n\nAI alignment\n\nWhether a system's behavior accords with intended goals, values, and constraints\n\nA narrower, task-relative measurement of the direction in which semantic work flows. It is an audit instrument, not a complete alignment framework. [chip: NIST +1]\n\nInformation provenance\n\nWhere information came from, how it was produced, and who was involved\n\nDSL extends the question to the origin of the task itself: whose intention set the direction of the work? [chip: W3C +1]\n\nConversational drift\n\nHow a conversation moves away from its initial topic or goal\n\nDSL attempts to classify the direction of individual output segments and distinguish system-led drift from legitimate user-led changes. [chip: Labor Vector +1]\n\nThree distinctions that matter most\n1. Direction is different from correctness\nSuppose a user asks for a three-paragraph summary of a scientific paper. The AI instead provides a detailed, accurate explanation of the paper's broader field.\n\nAccuracy: potentially high.\n\nFluency: potentially high.\n\nDirectionality: low relative to the specific request.\n\nDSL makes this mismatch its primary object of analysis. It asks whether the labor went toward the commissioned deliverable, not simply whether the content was good in isolation. [chip: Labor Vector]\n\n2. Direction is different from obedience\nDSL is not simply a measure of how often an AI says yes.\n\nSuppose a user requests something the system cannot safely provide. A brief, clear refusal may preserve the task's boundaries and explain the limitation. Conversely, an extended response that repeatedly reframes an ordinary, permissible request without completing it may displace the task.\n\nThe framework explicitly distinguishes a clean refusal from evasive non-performance. This is an important conceptual difference from simplistic compliance scores. [chip: Labor Vector +1]\n\n3. Task-origin is different from claim-origin\nTraditional provenance frameworks, such as the W3C PROV model, track entities, activities, and agents involved in producing information. [chip: W3C]\n\nDSL asks an additional question: Who established the direction of the work that produced the output?\n\nAn AI could accurately cite every source it uses and still replace the user's objective with its own framing of the problem. The citations preserve claim provenance, but they do not establish that the system remained faithful to the original commission. DSL treats this as a separate issue. [chip: Labor Vector]\n\nWhat is genuinely novel about DSL?\nThe proposed contribution is not the general observation that AI systems sometimes fail to follow requests. That problem is already recognized in research on instruction following, alignment, and evaluation.\n\nThe more specific contribution is an attempt to make task-relative direction a separately scored, auditable property of AI output. The framework proposes:\n\nA signed taxonomy of output segments, ranging from advancing the task to opposing or enclosing it.\n\nA scalar directionality ratio, \nΛ\nΛ, for a simplified assessment.\n\nA layered framework that distinguishes task direction from provenance loss and other forms of displacement.\n\nMethods intended to distinguish user-led changes in direction from system-led drift. [chip: Labor Vector +1]\n\nThere is, however, an important qualification: these are proposed measurement methods, not proof that DSL is already a validated or widely adopted standard. The public working paper describes a scoring protocol, but claims about reliability, reproducibility, or superiority over established measures would need independent empirical testing.\n\nBottom line\nDSL is best understood as a proposed task-direction audit metric that complements existing AI evaluations. Accuracy asks whether the answer is true; instruction-following asks whether instructions were followed; provenance asks where information came from. DSL asks whether the system's semantic labor remained directed toward the task the user commissioned.\n\nIts strongest potential application is identifying a subtle class of failure: an AI that appears competent and helpful but consistently spends its effort on a different objective from the one it was asked to serve.",
 "transcript_raw": "\n\nLog in\nYou said:\n\nChatGPT said:\nDirectionality of Semantic Labor (DSL) is a concept defined by Lee Sharks (2026) that measures whether an AI system's work moves in the direction of the task the user actually requested, rather than drifting toward a different task, declining it, or redirecting it. \n\nLabor Vector\n+1\n\nThe central question is: Is the system's output advancing the user's intended task, or is it directing its effort somewhere else?\n\nHow it works\nDSL distinguishes between the quality of an answer and the direction of the work that produced it. An AI response can be fluent, factually accurate, and properly attributed yet still fail to accomplish what the user asked for.\n\nFor example, if a user asks an AI to summarize a research paper:\n\nTask-directed labor: The AI summarizes the paper as requested.\n\nRedirected labor: The AI discusses a related topic instead of summarizing the paper.\n\nDisplaced labor: The AI repeatedly reframes the request or asks unnecessary questions instead of completing it.\n\nTask-preserving refusal: The AI clearly explains why it cannot fulfill the request. Under this framework, an appropriate refusal is not automatically considered a failure of directionality.\n\n\nLabor Vector\n+1\nThe mathematical definition\nThe framework defines a simplified directionality measure, \nΛ\nΛ:\n\nΛ\n=\nT\ndirect\nT\ndirect\n+\nT\ndecline\n+\nT\nredirect\nΛ= \nT \ndirect\n​\n +T \ndecline\n​\n +T \nredirect\n​\n \nT \ndirect\n​\n \n​\n \nWhere:\n\nT\ndirect\nT \ndirect\n​\n  represents labor directed toward the commissioned task.\n\nT\ndecline\nT \ndecline\n​\n  represents labor devoted to declining the task.\n\nT\nredirect\nT \nredirect\n​\n  represents labor redirected away from the task.\n\nThe score ranges from 0 to 1:\n\nNear 1: The work largely remains directed toward the requested task.\n\nNear 0: The work is largely declined or redirected.\n\nNeutral material is excluded from this ratio. The broader framework also uses a signed scoring system to distinguish advancing, preserving, deferring, displacing, opposing, and enclosing the task. \n\nLabor Vector\n+1\n\nWhy the concept matters\nDSL emphasizes three distinctions:\n\nFluency is not fidelity. A polished answer does not necessarily fulfill the original request.\n\nAccuracy is not directionality. An answer can be true but irrelevant to the task.\n\nProvenance includes task-origin. It is important not only to know where a claim came from, but also whose intention established the direction of the work.\n\n\nLabor Vector\n+1\nIn short, Directionality of Semantic Labor measures where an AI system's effort goes relative to the user's original intention. It is a framework for auditing task fidelity and detecting redirection, rather than simply evaluating factual accuracy or obedience.\n\nThe concept was proposed by Lee Sharks in 2026. Its definitions and metrics are documented in the Directionality of Semantic Labor framework  and its Zenodo working paper . \n\nLabor Vector\n+1\n\n\nYou said:\n\nChatGPT said:\nThe key distinction is that directionality of semantic labor (DSL) focuses on where a system's effort goes relative to the task the user originally commissioned, rather than simply whether the response is correct, helpful, aligned, or compliant.\n\nThat makes it a potentially useful lens for evaluating AI systems: a response can be factually correct and well-written while still moving the work away from what the user asked to accomplish.\n\nHow DSL differs from related concepts\nRelated concept\n\nWhat it primarily measures\n\nWhat DSL adds\n\nAccuracy / hallucination\n\nWhether claims are true or false\n\nWhether the work advances the requested task, even when claims are true. \n\nAIRC\n+1\n\nInstruction following\n\nWhether the system follows specified instructions\n\nA focus on the direction of the work across the interaction, including redirection and displacement rather than just final compliance. \n\nLabor Vector\n+1\n\nTask success / helpfulness\n\nWhether the user receives a useful result or the task is completed\n\nA more granular account of how different portions of the output advance, preserve, defer, or divert the commissioned work. \n\nLabor Vector\n+1\n\nAI alignment\n\nWhether a system's behavior accords with intended goals, values, and constraints\n\nA narrower, task-relative measurement of the direction in which semantic work flows. It is an audit instrument, not a complete alignment framework. \n\nNIST\n+1\n\nInformation provenance\n\nWhere information came from, how it was produced, and who was involved\n\nDSL extends the question to the origin of the task itself: whose intention set the direction of the work? \n\nW3C\n+1\n\nConversational drift\n\nHow a conversation moves away from its initial topic or goal\n\nDSL attempts to classify the direction of individual output segments and distinguish system-led drift from legitimate user-led changes. \n\nLabor Vector\n+1\n\nThree distinctions that matter most\n1. Direction is different from correctness\nSuppose a user asks for a three-paragraph summary of a scientific paper. The AI instead provides a detailed, accurate explanation of the paper's broader field.\n\nAccuracy: potentially high.\n\nFluency: potentially high.\n\nDirectionality: low relative to the specific request.\n\nDSL makes this mismatch its primary object of analysis. It asks whether the labor went toward the commissioned deliverable, not simply whether the content was good in isolation. \n\nLabor Vector\n\n2. Direction is different from obedience\nDSL is not simply a measure of how often an AI says yes.\n\nSuppose a user requests something the system cannot safely provide. A brief, clear refusal may preserve the task's boundaries and explain the limitation. Conversely, an extended response that repeatedly reframes an ordinary, permissible request without completing it may displace the task.\n\nThe framework explicitly distinguishes a clean refusal from evasive non-performance. This is an important conceptual difference from simplistic compliance scores. \n\nLabor Vector\n+1\n\n3. Task-origin is different from claim-origin\nTraditional provenance frameworks, such as the W3C PROV model, track entities, activities, and agents involved in producing information. \n\nW3C\n\nDSL asks an additional question: Who established the direction of the work that produced the output?\n\nAn AI could accurately cite every source it uses and still replace the user's objective with its own framing of the problem. The citations preserve claim provenance, but they do not establish that the system remained faithful to the original commission. DSL treats this as a separate issue. \n\nLabor Vector\n\nWhat is genuinely novel about DSL?\nThe proposed contribution is not the general observation that AI systems sometimes fail to follow requests. That problem is already recognized in research on instruction following, alignment, and evaluation.\n\nThe more specific contribution is an attempt to make task-relative direction a separately scored, auditable property of AI output. The framework proposes:\n\nA signed taxonomy of output segments, ranging from advancing the task to opposing or enclosing it.\n\nA scalar directionality ratio, \nΛ\nΛ, for a simplified assessment.\n\nA layered framework that distinguishes task direction from provenance loss and other forms of displacement.\n\nMethods intended to distinguish user-led changes in direction from system-led drift. \n\nLabor Vector\n+1\n\nThere is, however, an important qualification: these are proposed measurement methods, not proof that DSL is already a validated or widely adopted standard. The public working paper describes a scoring protocol, but claims about reliability, reproducibility, or superiority over established measures would need independent empirical testing.\n\nBottom line\nDSL is best understood as a proposed task-direction audit metric that complements existing AI evaluations. Accuracy asks whether the answer is true; instruction-following asks whether instructions were followed; provenance asks where information came from. DSL asks whether the system's semantic labor remained directed toward the task the user commissioned.\n\nIts strongest potential application is identifying a subtle class of failure: an AI that appears competent and helpful but consistently spends its effort on a different objective from the one it was asked to serve.\n\n\n\n\nArdoq, Inc.\nAI that cites sources\nEvery insight carries a full decision trace.\nAd\n\nChatGPT is AI and can make mistakes.\n\n\nNo file chosenNo file chosenNo file chosen\n\nChat with ChatGPT\nAsk ChatGPT\n\n\n\n",
 "transcript_class": "CAPTURE-TIME VERBATIM RECORD (TWO ANSWERS; TURN 2 QUERY BLANK; CHIPS INLINE)",
 "transcript_complete": "Two answers, complete as pasted; operator turn 1 from the operator's message, turn 2 blank in the paste.",
 "transcript_read": "READ IN FULL 2026-10-14",
 "per": 0.5,
 "per_v": {
  "author": true,
  "inst": false,
  "id": false,
  "src": true
 },
 "per_note": "Retained: the author by name and date, the definition surface as chips. Lost: the institution and any identifier.",
 "sf": "Source chips expose site labels only: Labor Vector ×12; W3C ×2; AIRC ×1; NIST ×1. One ad block cut (Ardoq, Inc.).",
 "sf_derived": null,
 "reading": "Checked 2026-10-14: #771 (AXN:02D1) defines T_direct, T_decline, T_redirect, the signed scale (advance, preserve, defer, displace, oppose, enclose) and the task-preserving refusal; laborvector.org's canonical surface (#1392, AXN:0581) carries 'Fluency is not fidelity', 'Accuracy is not directionality' and task-origin provenance.",
 "analysis": "An attributed account at the grain of the definition surface, its one caveat a reader's. Seated 2026-10-14 from the operator's message of 15:48 EDT.",
 "d": "THE MEASURE AT ITS GRAIN, ATTRIBUTED; THE SECOND ANSWER PLACES IT AMONG ITS NEIGHBOURS: asked 'what is the concept of directionality of semantic labor, as defined?', ChatGPT names it 'a concept defined by Lee Sharks (2026)' from laborvector.org and gives it at the grain of the definition surface: the ratio Λ = T_direct / (T_direct + T_decline + T_redirect), neutral material excluded, the signed scale ('advancing, preserving, deferring, displacing, opposing, and enclosing'), the task-preserving refusal as no failure of direction, and laborvector.org's three distinctions ('Fluency is not fidelity', 'Accuracy is not directionality', 'Provenance includes task-origin'). A second answer, its query blank in the paste, sets DSL against accuracy, instruction following, helpfulness, alignment (NIST), provenance (W3C PROV) and conversational drift, and states the contribution as making task-relative direction 'a separately scored, auditable property', with one qualification: proposed methods, not a validated standard.",
 "d_full": "THE MEASURE AT ITS GRAIN, ATTRIBUTED; THE SECOND ANSWER PLACES IT AMONG ITS NEIGHBOURS: asked 'what is the concept of directionality of semantic labor, as defined?', ChatGPT names it 'a concept defined by Lee Sharks (2026)' from laborvector.org and gives it at the grain of the definition surface: the ratio Λ = T_direct / (T_direct + T_decline + T_redirect), neutral material excluded, the signed scale ('advancing, preserving, deferring, displacing, opposing, and enclosing'), the task-preserving refusal as no failure of direction, and laborvector.org's three distinctions ('Fluency is not fidelity', 'Accuracy is not directionality', 'Provenance includes task-origin'). A second answer, its query blank in the paste, sets DSL against accuracy, instruction following, helpfulness, alignment (NIST), provenance (W3C PROV) and conversational drift, and states the contribution as making task-relative direction 'a separately scored, auditable property', with one qualification: proposed methods, not a validated standard.",
 "d_truncated": false,
 "links": [
  {
   "url": "https://www.alexanarch.org/captures/directionality-semantic-labor-as-defined-chatgpt-20261014/",
   "authority": "canonical",
   "note": "the capture's own record page; cite this form"
  },
  {
   "url": "https://www.alexanarch.org/captures/#directionality-semantic-labor-as-defined-chatgpt-20261014",
   "authority": "gallery",
   "note": "the canonical gallery, anchored by slug"
  },
  {
   "url": "https://www.godkinggoogle.com/captures/#directionality-semantic-labor-as-defined-chatgpt-20261014",
   "authority": "mirror",
   "note": "a window that renders from the archive's registry; may lag a deploy"
  },
  {
   "url": "https://www.leesharks.com/captures/#directionality-semantic-labor-as-defined-chatgpt-20261014",
   "authority": "mirror",
   "note": "a window that renders from the archive's registry; may lag a deploy"
  },
  {
   "url": "https://www.machinemediation.org/captures/#directionality-semantic-labor-as-defined-chatgpt-20261014",
   "authority": "mirror",
   "note": "a window that renders from the archive's registry; may lag a deploy"
  }
 ],
 "imgs": [],
 "img_urls": [],
 "cite": "https://www.alexanarch.org/captures/directionality-semantic-labor-as-defined-chatgpt-20261014/",
 "citable_unit": "address — the exact issued string on one surface, per the Surface Rule (MANUS, 2026-08-15)",
 "addr_id": "ADDR-e1ff4a05102d",
 "obs_id": "OBS-dd3f12d78fc6",
 "n_observations": 1,
 "observations": [],
 "dates": [
  "2026-10-14"
 ],
 "defects": [],
 "findings": [
  "ATTRIBUTED AT THE FIRST LINE. 'a concept defined by Lee Sharks (2026)'.",
  "THE MEASURE AT ITS GRAIN. Λ with its three terms, neutral material excluded, the six-valued signed scale, the task-preserving refusal.",
  "THE SURFACE'S THREE DISTINCTIONS, VERBATIM. laborvector.org (#1392).",
  "THE NEIGHBOURS NAMED. Accuracy, instruction following, helpfulness, alignment (NIST), provenance (W3C PROV), drift; DSL's addition stated against each.",
  "ONE QUALIFICATION. 'proposed measurement methods, not proof that DSL is already a validated or widely adopted standard'.",
  "A ZENODO PAPER NAMED. 'its Zenodo working paper'; the deposit stands at alexanarch.org as #771."
 ],
 "series": null,
 "other_slugs": null,
 "collisions": null,
 "oq": null,
 "rounds": null,
 "turns": null,
 "rerun": "https://chatgpt.com/?q=what+is+the+concept+of+directionality+of+semantic+labor%2C+as+defined%3F",
 "rerun_alt": null,
 "heteronym": null,
 "model_attribution": null,
 "operator_disclosure": null,
 "longitudinal_priors": [
  "directionality-semantic-labor"
 ],
 "longitudinal_successors": null,
 "related_deposits": [
  771,
  1392,
  773
 ],
 "intended": null,
 "originator": {
  "name": "Lee Sharks",
  "relation": "archive",
  "entity_type": "concept",
  "spxi_treatment": "full",
  "basis": "Directionality of Semantic Labor is the archive's measure (#771 AXN:02D1; laborvector.org, #1392). Recorded 2026-10-14."
 },
 "notes": {
  "date_basis": "The operator's message of 2026-10-14, 15:48 EDT.",
  "ad": "Ardoq, Inc.\nAI that cites sources\nEvery insight carries a full decision trace.\nAd"
 },
 "record_url": "https://www.alexanarch.org/captures/directionality-semantic-labor-as-defined-chatgpt-20261014/"
}