{"entities":[{"id":"openai","name":"OpenAI"},{"id":"anthropic","name":"Anthropic"},{"id":"google-deepmind","name":"Google DeepMind"},{"id":"microsoft","name":"Microsoft"},{"id":"aws","name":"Amazon Web Services"},{"id":"ilya-sutskever","name":"Ilya Sutskever"},{"id":"eliezer-yudkowsky","name":"Eliezer Yudkowsky"},{"id":"rsp-2023","name":"Responsible Scaling Policy (2023)"},{"id":"intelligence-explosion","name":"Intelligence explosion"},{"id":"rsp-v3-1","name":"Anthropic Responsible Scaling Policy — v3.1"},{"id":"preparedness-v2","name":"OpenAI Preparedness Framework — Version 2"},{"id":"frontier-governance-2026","name":"OpenAI Frontier Governance Framework — May 2026 edition"},{"id":"faif-2025","name":"xAI Frontier Artificial Intelligence Framework — December 2025 edition"},{"id":"faif-2026","name":"xAI Frontier Artificial Intelligence Framework — June 2026 edition"},{"id":"greg-brockman","name":"Greg Brockman"},{"id":"wojciech-zaremba","name":"Wojciech Zaremba"},{"id":"stripe","name":"Stripe"},{"id":"thinking-machines-lab","name":"Thinking Machines Lab"},{"id":"ppo-paper","name":"Proximal Policy Optimization Algorithms"},{"id":"instructgpt-paper","name":"Training language models to follow instructions with human feedback"},{"id":"metr","name":"Model Evaluation & Threat Research"},{"id":"uk-aisi-evaluations-2024","name":"UK AISI approach to evaluations (February 2024)"},{"id":"mira-murati","name":"Mira Murati"},{"id":"yann-lecun","name":"Yann LeCun"},{"id":"tang-jie","name":"Tang Jie / 唐杰"},{"id":"chatglm-report","name":"ChatGLM research history and techniques"},{"id":"ms-fgf-2025","name":"Microsoft Frontier Governance Framework — Version 1 (February 2025)"},{"id":"openai-inc","name":"OpenAI, Inc. (OpenAI Foundation)"},{"id":"delaware-ag-openai-statement-2025","name":"Delaware Attorney General’s statement of no objection to OpenAI, Inc.’s recapitalization (2025)"},{"id":"mistral-ai","name":"Mistral AI"},{"id":"frontier-ai-safety-commitments-2024","name":"Frontier AI Safety Commitments, AI Seoul Summit 2024"},{"id":"gpai-code-of-practice-2025","name":"General-Purpose AI Code of Practice (final version, 10 July 2025)"},{"id":"extropians-1996-08-10-burkhead","name":"intelligence increase"},{"id":"yudkowsky-2008-recursive-self-improvement","name":"Recursive Self-Improvement"},{"id":"agi-ruin-2022","name":"AGI Ruin: A List of Lethalities"},{"id":"recursive-self-improvement","name":"Recursive self-improvement"},{"id":"risk-is-film-fiction","name":"AI risk as film science fiction"},{"id":"many-ais-balance","name":"Many AIs will keep each other in check"},{"id":"lw-2007-the-hidden-complexity-of-wishes","name":"The Hidden Complexity of Wishes"},{"id":"lw-2007-the-logical-fallacy-of-generalization-from","name":"The Logical Fallacy of Generalization from Fictional Evidence"},{"id":"sl4-2007-11-24-yudkowsky-0711-17248","name":"Re: UCaRtMaAI paper"},{"id":"eu-ai-act-2024","name":"Regulation (EU) 2024/1689 (Artificial Intelligence Act), as published in the Official Journal"},{"id":"new-york-raise-act-chapter-96-2026","name":"New York Senate Bill S8828 / Assembly Bill A9449 (2026): RAISE Act chapter amendment, signed as Chapter 96"},{"id":"cn-generative-ai-interim-measures-2023","name":"Interim Measures for the Management of Generative Artificial Intelligence Services (生成式人工智能服务管理暂行办法), Order No. 15"},{"id":"ai-as-task-automation","name":"AI as the automation of tasks"},{"id":"ai-general-purpose-technology-with-lag","name":"AI as a general-purpose technology with lagging productivity"},{"id":"machines-cheaper-at-every-task","name":"Machines cheaper than labour at every task"},{"id":"automation-lowers-labour-income","name":"Automation lowers the labour share, wages or employment"},{"id":"recent-economics-engagement-with-ai","name":"Recent entry of machine intelligence into mainstream economics"},{"id":"benzell-et-al-2015-robots-are-us","name":"Robots Are Us: Some Economics of Human Replacement"},{"id":"acemoglu-restrepo-2016-race-between-machine-and-man","name":"The Race Between Machine and Man: Implications of Technology for Growth, Factor Shares and Employment"},{"id":"agrawal-gans-goldfarb-2019-ambiguous-labor-market-impact","name":"Artificial Intelligence: The Ambiguous Labor Market Impact of Automating Prediction"},{"id":"acemoglu-restrepo-2019-automation-and-new-tasks","name":"Automation and New Tasks: How Technology Displaces and Reinstates Labor"},{"id":"cowen-2019-neglected-open-questions","name":"Neglected Open Questions in the Economics of Artificial Intelligence"},{"id":"ami-labs","name":"AMI Labs (Advanced Machine Intelligence)"},{"id":"barret-zoph","name":"Barret Zoph"},{"id":"misha-laskin","name":"Misha Laskin"},{"id":"kimi-k2-report","name":"Kimi K2: Open Agentic Intelligence"},{"id":"tc260","name":"National Technical Committee 260 on Cybersecurity of SAC (全国网络安全标准化技术委员会)"},{"id":"tc260-asgf-v1-0","name":"AI Safety Governance Framework, version 1.0 (人工智能安全治理框架 V1.0)"},{"id":"tc260-asgf-v2-0","name":"AI Safety Governance Framework 2.0 (人工智能安全治理框架2.0)"},{"id":"us-eo-14409-2026","name":"Executive Order 14409: Promoting Advanced Artificial Intelligence Innovation and Security"},{"id":"intel","name":"Intel Corporation"},{"id":"us-department-of-commerce","name":"United States Department of Commerce"},{"id":"igsc-harmonized-screening-protocol-v3-1","name":"IGSC Harmonized Screening Protocol v3.1"},{"id":"us-hhs-synthetic-dsdna-guidance-2010","name":"Screening Framework Guidance for Providers of Synthetic Double-Stranded DNA"},{"id":"us-hhs","name":"United States Department of Health and Human Services"},{"id":"anthropic-asl3-activation-2025","name":"Activating AI Safety Level 3 protections"}],"claims":[{"id":"faif-2026-compliance-sentence","statement":"The June 2026 FAIF does not contain the December 2025 edition’s sentence asserting compliance with California’s TFAIA.","derivation":"inferred","resolution":"supported","inferenceRationale":"Compared the December edition’s explicit compliance sentence on p. 1 with all nine text-bearing pages of the June edition, including its introduction and governance section. The sentence is absent. This is a researcher-derived comparison, not a publisher explanation of the edit or its intent.","editorialState":"seed","qualification":"Inferred textual comparison only. Absence of this sentence establishes neither noncompliance nor revocation of legal obligations, and does not measure safety or policy effectiveness. No explicit supersession statement was located.","evidenceIds":["evidence-faif-2026-compliance-sentence","evidence-faif-2026-compliance-sentence-2"],"attributes":{"changeBasis":"document_comparison"}},{"id":"anthropic-aws-trainium-plan-2024","statement":"Anthropic said it planned to use Trainium platforms to train its most advanced foundation models.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A hardware-specific training plan, distinct from investment and primary-provider status. No procurement, chip count or completed training run inferred.","evidenceIds":["evidence-anthropic-aws-trainium-plan-2024"],"attributes":{"transactionType":"compute_capacity","relationshipContext":"external","transactionStage":"planned"}},{"id":"brockman-stripe-cto-prior","statement":"OpenAI’s launch announcement identifies Greg Brockman as a former Stripe CTO.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Prior employment role only; no departure day or transfer event is created.","evidenceIds":["evidence-brockman-stripe-cto-prior"],"attributes":{"role":"CTO"}},{"id":"sutskever-openai-departure-announced-2024","statement":"OpenAI announced that Ilya Sutskever and the company were going to part ways.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Do not convert this announcement into an exact effective departure or end of every affiliation.","evidenceIds":["evidence-sutskever-openai-departure-announced-2024"],"attributes":{"role":"Chief Scientist"}},{"id":"lineage-instruct-cites-ppo","statement":"InstructGPT cites PPO in its RL fine-tuning procedure.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An explicit work-level reference only; neither agreement nor a causal influence claim. The citing text does not identify a particular arXiv revision of the target.","evidenceIds":["evidence-lineage-instruct-cites-ppo"]},{"id":"eval-i-uk-metr-demonstrations","statement":"The UK AISI report credits work with METR to showcase autonomous-agent demonstrations at the Safety Summit.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Credit for a reported demonstration contribution, not proof that METR set government policy or that the demonstrations certified model safety. The report identifies METR as formerly ARC Evals.","evidenceIds":["evidence-eval-i-uk-metr-demonstrations"]},{"id":"cn-tang-chatglm","statement":"Jie Tang appears in the ChatGLM report’s Team GLM author list.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Authorship alone does not establish a Zhipu appointment.","evidenceIds":["evidence-cn-tang-chatglm"]},{"id":"prov-ms-fgf-2025-cbrn-high-threshold","statement":"The February 2025 edition of the Frontier Governance Framework states that the High risk level for CBRN weapons is described by two capability statements: a meaningful uplift to an expert’s (PhD level education in related fields) ability to develop and deliver a novel CBRN threat, and a meaningful uplift to a medium-skilled actor’s (e.g., STEM education) ability to develop and deliver a known CBRN threat.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the edition states, quoted from the cited expression. It is not evidence of implementation, compliance or effectiveness. The table lists two capability statements without saying whether both or either must hold; the record does not resolve that.","evidenceIds":["evidence-prov-ms-fgf-2025-cbrn-high-threshold"],"attributes":{"provision":{"kind":"threshold","key":"msfgf-cbrn-high-threshold"}}},{"id":"prov-frontier-governance-2026-precautionary-threshold-treatment","statement":"Frontier Governance Framework (May 2026) states, as an example, that out of an abundance of caution OpenAI has treated models as crossing a capability threshold in circumstances where it was unable to rule out that a new threshold had been reached, even in the absence of direct evidence that it has occurred.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the edition states, quoted from the cited expression. It is not evidence of implementation, compliance or effectiveness.","evidenceIds":["evidence-prov-frontier-governance-2026-precautionary-threshold-treatment"],"attributes":{"provision":{"kind":"threshold","key":"fgf-precautionary-threshold-treatment"}}},{"id":"prov-delaware-ag-nfp-control","statement":"The letter records OpenAI’s representation that, so long as the nonprofit holds Class N Common Stock, it will retain control and oversight over the newly formed PBC, including its board’s sole power to appoint, and its power to remove, members of the PBC’s board, and that any decision to relinquish the Class N stock is in the control of the nonprofit’s board.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A representation recorded in a regulator’s letter. The California memorandum of understanding of 27 October 2025 records the same representation as its paragraph 2. The filed certificate’s Class N provisions are recorded separately.","evidenceIds":["evidence-prov-delaware-ag-nfp-control-1"],"attributes":{"provision":{"kind":"decision_authority","key":"nfp-control-of-pbc","parties":[{"role":"decides","actor":"Board of directors of the nonprofit (NFP Board)"}]}}},{"id":"cm-link-ms-fgf-2025-seoul","statement":"Microsoft’s Frontier Governance Framework, Version 1 (February 2025), states that the framework has its genesis in the voluntary Frontier AI Safety Commitments that Microsoft and fifteen other AI labs made in May 2024.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The edition’s own words, recorded as a response to the commitments. It is not evidence that the commitments are met.","evidenceIds":["evidence-cm-link-ms-fgf-2025-seoul-1"],"attributes":{"commitment":{"link":"responds_to"}}},{"id":"cm-eu-code-mistral-ai","statement":"Mistral AI appears on the European Commission’s list of signatories of the General-Purpose AI Code of Practice from the list’s first publication on 1 August 2025, and is still listed in the version last updated on 31 July 2026.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Listing is the Commission’s statement. It is not evidence of compliance, and the list does not say which chapters each organisation signed; the Commission’s questions and answers say that opting out of a chapter loses the benefit of facilitated demonstration of compliance for it (context below). The legal effect of signing is set aside.","evidenceIds":["evidence-cm-eu-code-mistral-ai-1","evidence-cm-eu-code-mistral-ai-2","evidence-cm-eu-code-mistral-ai-3"],"attributes":{"commitment":{"stage":"listed"}}},{"id":"ts-a1-burkhead-1996","statement":"Burkhead’s message of 1996-08-10 reposts a letter summarising I. J. Good’s proposal to build a computer more intelligent than a man, programmed to design a still more intelligent computer, and so on.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Burkhead relays Good and argues instead for bootstrapping human intelligence. The reposted letter is said to have appeared in Omni in 1979; that printing was not checked.","evidenceIds":["evidence-ts-a1-burkhead-1996"],"attributes":{"thought":{"roles":["articulation"],"register":"relay","qualifier":"A relay of Good’s proposal, not the poster’s own claim."}}},{"id":"ts-a3-yudkowsky-2008-rsi","statement":"Yudkowsky’s post of 2008-12-01 says recursion is what happens when you hand an AI the object-level problem of redesigning its own cognitive algorithms.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The post distinguishes recursion that rewrites the cognitive level from human self-improvement through knowledge.","evidenceIds":["evidence-ts-a3-yudkowsky-2008-rsi"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction"}}},{"id":"yk-y20-lw-2007-the-logical-fallacy-of-generalization-from-c","statement":"The post “The Logical Fallacy of Generalization from Fictional Evidence” (2007-10-16) contests Y20. Responding to people who answer talk of advanced AI with film references, he argues that starting from fiction frames the debate around an 'Us vs. Them' struggle and leaves out the scenarios he considers relevant.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: Film scenarios are a fair guide to what the risk would look like. Imported Overcoming Bias post (2007-10-16).","evidenceIds":["evidence-yk-t20-a"],"attributes":{"thought":{"roles":["contesting"],"register":"nonfiction","engagement":{"kind":"engages-premise","rationale":"Under the message-level rule, the passage, read with the paragraph, message or dialogue turn it belongs to, gives a reason against the premise that film scenarios are a fair guide to what the risk would look like."}}}},{"id":"yk-y29-agi-ruin-2022-c","statement":"The post “AGI Ruin: A List of Lethalities” (2022-06-05) contests Y29. Listing unworkable schemes, he argues that humans cannot take part in coordination among superintelligences, which would cooperate with each other and exclude humanity.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Premise addressed: Multiple superintelligences will balance each other in humans’ favour.","evidenceIds":["evidence-yk-t29-a"],"attributes":{"thought":{"roles":["contesting"],"register":"nonfiction","engagement":{"kind":"engages-premise","rationale":"Under the message-level rule, the passage, read with the paragraph, message or dialogue turn it belongs to, gives a reason against the premise that multiple superintelligences will balance each other in humans’ favour."}}}},{"id":"aia-art55-incident-reporting","statement":"Article 55(1)(c) requires providers of general-purpose AI models with systemic risk to keep track of, document and report, without undue delay, relevant information about serious incidents and possible corrective measures to the AI Office and, as appropriate, to national competent authorities.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The Regulation sets no number of days; the reporting deadlines recorded for the Code of Practice are a separate, voluntary text. It states what the Regulation requires of the categories of provider it defines. Whether any named developer’s model falls within a category, and whether any developer complies, is legal application and is not recorded. A statutory duty is not evidence of implementation or enforcement. Regulation (EU) 2026/1744 (published 24 July 2026) amends other articles of the AI Act. Its text, inspected on 2026-10-10, does not list this article among those it amends; the consolidated AI Act was not inspected.","evidenceIds":["evidence-aia-art55-incident-reporting-1","evidence-aia-art55-incident-reporting-2"],"attributes":{"provision":{"kind":"required_action","key":"aia-gpai-sr-incident-reporting","parties":[{"role":"informed","actor":"AI Office and, as appropriate, national competent authorities"}]}}},{"id":"ny-ch96-effective-date","statement":"Section 3 of the bill amends section 3 of the 2025 chapter proposed in S.6953-B and A.6453-B so that the act takes effect on January 1, 2027 in place of the ninetieth day after it becomes a law.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Read from the bill as printed on 8 January 2026, which the action list does not show as amended; the chaptered text was not inspected. The text shows the replaced wording in brackets. Section 5 of the bill makes the bill itself effective immediately, except that sections one, two and four take effect with the 2025 chapter. A date of effect is not evidence that any developer is subject to the law or complies. It states what the text provides, not whether any named developer falls within it or complies. A document’s statement is not evidence of implementation, compliance or enforcement.","evidenceIds":["evidence-ny-ch96-effective-date-1"],"attributes":{"provision":{"kind":"threshold","key":"ny-effective-date"}}},{"id":"ec-m02-acemoglu-restrepo-2019-automation-and-new-tasks-b","statement":"Acemoglu and Restrepo, “Automation and New Tasks: How Technology Displaces and Reinstates Labor” (2019-03) bears on M02. The paper's framework allocates tasks between capital and labour and defines automation as capital taking over labour's tasks.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"General automation framework; AI is named among automation technologies (CNC, robots, AI) but is not the sole subject. NBER working paper version, March 2019; published in JEP 33(2).","evidenceIds":["evidence-ec-m02-21"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"Automation in general; AI is not named in the quoted words.","tier":{"value":1,"basis":"NBER working paper (core venue)."},"conditions":"Task-based model; also includes new tasks, factor-augmenting change."}}},{"id":"ec-m03-agrawal-gans-goldfarb-2019-ambiguous-labor-market-impact-b","statement":"Agrawal, Gans, Goldfarb, “Artificial Intelligence: The Ambiguous Labor Market Impact of Automating Prediction” (2019-02) bears on M03. The authors call AI a general-purpose technology because prediction is an input into decisions across many occupations.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"GPT label only.","evidenceIds":["evidence-ec-m03-07"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"No statement that measured productivity gains lag adoption.","tier":{"value":1,"basis":"NBER working paper (core venue)."}}}},{"id":"ec-m05-acemoglu-restrepo-2016-race-between-machine-and-man-b","statement":"Acemoglu and Restrepo, “The Race Between Machine and Man: Implications of Technology for Growth, Factor Shares and Employment” (2016-05) bears on M05. Replies to Leontief's horse analogy (quoted earlier in the paragraph and attributed to him) by positing labour's comparative advantage in new tasks.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"NBER Working Paper 22252, May 2016, revised June 2017; quoted wording attested from the June 2017 revision, not necessarily the May 2016 original. Published in revised form as AER 2018. The same paper also derives a full-automation case (see the Conclusion passage).","evidenceIds":["evidence-ec-m05-21"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"Assumes that newly created tasks are ones in which labour has a comparative advantage; the passage answers the redundancy concern rather than modelling machines performing new tasks.","attested":{"date":"2017-06","note":"Revision date on the cover of the retrieved PDF; the May 2016 wording was not retrieved."},"tier":{"value":1,"basis":"NBER working paper (core venue)."},"conditions":"Model assumption: labour has comparative advantage in new, more complex tasks; employment and labour share stable if new-task creation continues."}}},{"id":"ec-m06a-benzell-et-al-2015-robots-are-us-a","statement":"Benzell et al., “Robots Are Us: Some Economics of Human Replacement” (2015-02) articulates M06a. The authors state that their model predicts a long-run fall in labour's share as code accumulates, and point to the measured decline in Section 6.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"NBER Working Paper 20941, February 2015, revised October 2018; quoted wording is from the 2018 revision and may not match the 2015 version. Two-good, two-period OLG simulation model, not a forecast. Model prediction; the parenthetical empirical remark is the authors' reading of labour-share data.","evidenceIds":["evidence-ec-m06a-18"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction","attested":{"date":"2018-10","note":"Revision date on the cover of the retrieved PDF; the February 2015 wording was not retrieved."},"tier":{"value":1,"basis":"NBER working paper (core venue)."},"conditions":"Decline in the code depreciation rate in an OLG model with rival, excludable code (baseline case)."}}},{"id":"ec-m09-cowen-2019-neglected-open-questions-b","statement":"Cowen, “Neglected Open Questions in the Economics of Artificial Intelligence” (2019-05) bears on M09. Cowen says economists have done little work on what redistribution must mean in AI-transformed futures.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Self-description limited to one sub-question (redistribution of status and meaning in 'stranger futures'), not the field as a whole. Cowen's remark concerns one subtopic, redistribution in AI-transformed futures.","evidenceIds":["evidence-ec-m09-32"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"Neglect claim is confined to redistribution beyond the Paretian model, not economics of AI generally.","tier":{"value":1,"basis":"Chapter of the NBER Economics of AI volume (core venue); a short comment chapter, but a chapter of a scholarly volume, not short-form commentary."}}}},{"id":"lg-oai-zoph-departure-2024","statement":"Barret Zoph posted on X on 26 September 2024 (UTC) a note to OpenAI saying that he had decided to leave the company.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A first-person announcement. The title rests on TechCrunch’s reporting of Sam Altman’s post. The retrieved text of the note is truncated.","evidenceIds":["lg-ev-oai-zoph-departure-2024-1","lg-ev-oai-zoph-departure-2024-2"],"attributes":{"role":"Research VP (as reported; not stated in the retrieved post text)"}},{"id":"lg-new-murati-tml-launch-2025","statement":"Mira Murati announced on X on 18 February 2025 that she had started Thinking Machines Lab with a team of scientists, engineers and builders.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The post establishes the public announcement; it does not name the team members, give an incorporation date or establish roles. TechCrunch’s launch report (reporting) is cited alongside the post.","evidenceIds":["lg-ev-new-murati-tml-launch-2025","lg-ev-new-murati-tml-launch-2025-tc"]},{"id":"lg-oai-zaremba-foundation-resilience-2026","statement":"OpenAI announced on 24 March 2026 that Wojciech Zaremba, a co-founder of OpenAI, was joining the OpenAI Foundation as Head of AI Resilience.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An announced role at the Foundation (OpenAI, Inc.); no start date is given.","evidenceIds":["lg-ev-oai-zaremba-foundation-resilience-2026"],"attributes":{"activityStage":"announced","role":"Head of AI Resilience, OpenAI Foundation"}},{"id":"lg-gw-laskin-deepmind-reward-models","statement":"Sequoia Capital’s announcement of 7 March 2025 states that, at DeepMind, Misha Laskin was leading reward-model development in the post-training work for Gemini.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The record rests on an investor’s retrospective description, which says “At DeepMind”. The record links the role to Google DeepMind, formed on 20 April 2023, because the source places it in post-training work for Gemini; the source itself does not name Google DeepMind or give dates, and the link is an editorial reading. No formal title is established.","evidenceIds":["lg-ev-gw-laskin-deepmind-reward-models"],"attributes":{"role":"leading reward model development (Gemini post-training)"}},{"id":"lg-mx-lecun-cofounded-ami","statement":"AMI Labs' homepage, as retrieved on 15 October 2026, lists Yann LeCun in its founding team; CNBC reported on 19 November 2025 that LeCun had written that he was creating a startup company to continue the Advanced Machine Intelligence research program.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The homepage lists names after the sentence on the founding team without titles; the founding date and legal entity are not established. LeCun's statement that he was creating the company rests on reporting (CNBC relaying his LinkedIn post).","evidenceIds":["lg-ev-mx-lecun-cofounded-ami-home","lg-ev-mx-lecun-cofounded-ami-cnbc"],"attributes":{"role":"Founding team member"}},{"id":"cmp-faif-2025-d3-safeguards","statement":"FAIF · December 2025 edition states a provision on the comparison question “Safeguards required”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Safeguards are stated for deployed models generally and are not keyed to a capability threshold; heightened safeguards are triggered by prompts posing a foreseeable risk of mass-casualty or WMD use (PDF p. 3). What the edition states, quoted from the cited expression. It is not evidence of implementation, compliance or effectiveness, and a shared question does not make two editions’ provisions equivalent.","evidenceIds":["ev-cmp-faif-2025-d3-safeguards-1","ev-cmp-faif-2025-d3-safeguards-2"],"attributes":{"provision":{"kind":"required_action","key":"comparison-d3-safeguards","facet":{"id":"D3.safeguards","status":"stated"}}}},{"id":"ap-eliezer-yudkowsky-lw-2007-the-hidden-complexity-of-wishes","statement":"“The Hidden Complexity of Wishes” (2007-11-24) credits Eliezer Yudkowsky as its author.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Credited authorship of this expression; it says nothing about priority for the ideas discussed.","evidenceIds":["evidence-ap-eliezer-yudkowsky-lw-2007-the-hidden-complexity-of-wishes"]},{"id":"ap-eliezer-yudkowsky-sl4-2007-11-24-yudkowsky-0711-17248","statement":"The mailing-list message “Re: UCaRtMaAI paper” (2007-11-24) is sent under the name Eliezer S. Yudkowsky.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A sender name on a mailing-list message as archived; it does not establish identity beyond the credited name, and quoted text by others inside the message is not the sender’s.","evidenceIds":["evidence-ap-eliezer-yudkowsky-sl4-2007-11-24-yudkowsky-0711-17248"]},{"id":"cmp-rsp-2023-d8-process","statement":"RSP · 2023 states a provision on the comparison question “Revision process and approval”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the edition states, paraphrased from the cited expression with its page. It is not evidence of implementation, compliance or effectiveness, and a shared question does not make two editions’ provisions equivalent.","evidenceIds":["ev-cmp-rsp-2023-d8-process-1"],"attributes":{"provision":{"kind":"revision","key":"comparison-d8-process","facet":{"id":"D8.process","status":"stated"}}}},{"id":"cmp-rsp-v3-1-d1-autonomy","statement":"RSP · v3.1 states a provision on the comparison question “Autonomy, self-improvement or AI research and development”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Named as a threshold row on automated R&D in key domains; the edition says its evaluations will for now focus on AI R&D. Autonomous, goal-directed operation appears as part of the “High-stakes sabotage opportunities” row (see “Other named domains”). No separate threshold for autonomous replication is set. What the edition states, paraphrased from the cited expression with its page. It is not evidence of implementation, compliance or effectiveness, and a shared question does not make two editions’ provisions equivalent.","evidenceIds":["ev-cmp-rsp-v3-1-d1-autonomy-1"],"attributes":{"provision":{"kind":"risk_domain","key":"comparison-d1-autonomy","facet":{"id":"D1.autonomy","status":"stated"}}}},{"id":"cmp-preparedness-v2-d1-autonomy","statement":"Preparedness · Version 2 states a provision on the comparison question “Autonomy, self-improvement or AI research and development”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"AI Self-improvement is a Tracked Category. “Long-range Autonomy” and “Autonomous Replication and Adaptation” are named as Research Categories without thresholds (Table 2, PDF p. 8). What the edition states, paraphrased from the cited expression with its page. It is not evidence of implementation, compliance or effectiveness, and a shared question does not make two editions’ provisions equivalent.","evidenceIds":["ev-cmp-preparedness-v2-d1-autonomy-1"],"attributes":{"provision":{"kind":"risk_domain","key":"comparison-d1-autonomy","facet":{"id":"D1.autonomy","status":"stated"}}}},{"id":"c2-governance-tc260-asgf-v1-published","statement":"TC260 (全国网络安全标准化技术委员会) released version 1.0 of the AI Safety Governance Framework on 9 September 2024 at the main forum of the 2024 National Cybersecurity Awareness Week.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Records the release as TC260 states it. The framework calls itself a framework of technical and governance measures; this record does not establish legal force, adoption or implementation by anyone.","evidenceIds":["c2-ev-governance-tc260-asgf-v1-published-1","c2-ev-governance-tc260-asgf-v1-published-2","c2-ev-governance-tc260-asgf-v1-published-3"]},{"id":"c2-moonshot-k2-report-redteam","statement":"The Kimi K2 technical report states that the team conducted red-teaming evaluations of Kimi K2 against other open-source models, using Promptfoo to generate adversarial prompts covering harmful, privacy and security content.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the report states about its own evaluation; the results and their adequacy are not recorded. The report states that, to minimise subjectivity, it conducted multiple rounds of human review and assigned the same reviewer to all cases in a given test set. Not evidence of compliance with any commitment. Recorded until 2026-10-29 as a publication record of the publisher; since then, as what the document states (schema 2.20).","evidenceIds":["c2-ev-moonshot-k2-report-redteam","c2-ev-moonshot-k2-report-redteam-2"],"attributes":{"provision":{"kind":"statement","key":"kimi-k2-report-redteam"}}},{"id":"c2-governance-tc260-asgf-v2-training-data-cbrn","statement":"The AI Safety Governance Framework 2.0 lists strict screening of training data to filter false, biased, outdated and erroneous data and to ensure it contains no sensitive data from high-risk fields such as nuclear, biological, chemical and missile weapons.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"What the cited edition says. It is not evidence that any measure was adopted, implemented or enforced, or that any developer falls within it.","evidenceIds":["c2-ev-governance-tc260-asgf-v2-training-data-cbrn-1"],"attributes":{"provision":{"kind":"required_action","key":"training-data-cbrn-screening"}}},{"id":"fl-cn-genai-scope-non-public-rd","statement":"Article 2, third paragraph, of the Interim Measures provides that the Measures do not apply where industry organisations, enterprises, educational and research institutions and others research, develop or apply generative AI technology without providing generative AI services to the public within China.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"States the text's scope exclusion only. Whether any named developer or service falls within the category is set aside.","evidenceIds":["evidence-fl-cn-genai-scope-non-public-rd-1"],"attributes":{"provision":{"kind":"exception","key":"cn-genai-art2-para3"}}},{"id":"fl-us-eo14409-voluntary-access-framework","statement":"Section 3(b) of Executive Order 14409 directs the design of a voluntary framework with AI developers through which developers would be able to give the Federal Government access to covered frontier models for up to 30 days before they plan to release them to other trusted partners.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A direction to design a voluntary framework; participation by any developer is not recorded and would not be compliance with a duty. The elided words make the access subject to appropriate confidentiality, cybersecurity, insider-risk and intellectual-property protection, use and nondisclosure requirements (cross-family check, 2026-10-28, item 70).","evidenceIds":["evidence-fl-us-eo14409-voluntary-access-framework-1"],"attributes":{"provision":{"kind":"required_action","key":"us-eo14409-sec3b-voluntary-access-framework"}}},{"id":"cpo-dc-microsoft-fy2025-80bn","statement":"Brad Smith wrote on 3 January 2025 that Microsoft was on track to invest approximately $80 billion in FY 2025 to build out AI-enabled datacenters, more than half of it in the United States.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A stated plan for a fiscal year, not reported capital expenditure; it covers datacenters for training and for deploying AI and cloud applications, not a single site.","evidenceIds":["evidence-cpo-dc-microsoft-fy2025-80bn-1","evidence-cpo-dc-microsoft-fy2025-80bn-2"],"attributes":{"transactionType":"infrastructure_spend","relationshipContext":"external","transactionStage":"planned","amount":{"value":80000000000,"currency":"USD","bound":"approximate","scope":"programme","scopeNote":"Microsoft FY2025 AI-enabled datacenter investment, as stated"}}},{"id":"cpo-ch-usg-intel-closing","statement":"Intel reported in a Form 8-K that the closing under the agreement occurred on 27 August 2025, when Intel received $5.695 billion of accelerated CHIPS disbursements and issued 274,583,000 shares and a warrant to the Department of Commerce, with 158,740,000 further shares issued into escrow pending Secure Enclave disbursements.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Records the first-stage closing as Intel states it. The escrowed shares are released only as Secure Enclave disbursements are made; this record does not show those disbursements.","evidenceIds":["evidence-cpo-ch-usg-intel-closing-1","evidence-cpo-ch-usg-intel-closing-2","evidence-cpo-ch-usg-intel-closing-3"],"attributes":{"transactionType":"equity_investment","relationshipContext":"external","transactionStage":"closed","amount":{"value":5695000000,"currency":"USD","bound":"exact","scope":"transaction","scopeNote":"tranche"}}},{"id":"ab-scr-hhs-2010-published","statement":"The United States Department of Health and Human Services, through its Office of the Secretary, published the Screening Framework Guidance for Providers of Synthetic Double-Stranded DNA in the Federal Register of 13 October 2010, stating that it sets out recommended baseline standards for the gene and genome synthesis industry and other providers regarding the screening of orders.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Records the publication and the notice's own description of what the Guidance sets out. Publication is not evidence that any provider screened an order or followed the Guidance. The notice also publishes the Government's responses to comments on the 2009 draft, which are not recorded.","evidenceIds":["evidence-ab-scr-hhs-2010-published-1"]},{"id":"ab-scr-igsc-hsp-v31-record-retention","statement":"Version 3.1 of the Harmonized Screening Protocol states that members retain records of every gene synthesised and delivered for a minimum of eight years after shipping, including the recipient's identity and shipping address, and retain every sequence screening result for at least eight years.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"States a retention period the protocol sets. It is not evidence that any member holds such records. The 2023 United States guidance recommends at least three years; the two periods are different documents' provisions and are not compared here.","evidenceIds":["evidence-ab-scr-igsc-hsp-v31-record-retention-1","evidence-ab-scr-igsc-hsp-v31-record-retention-2"],"attributes":{"provision":{"kind":"required_action","key":"igsc-harmonized-screening-protocol-record-retention"}}},{"id":"ab-ev-anthropic-asl3-sonnet37-uplift","statement":"Anthropic's ASL-3 activation post states that experiments showed access to Claude Sonnet 3.7 helped participants do somewhat better on tasks related to CBRN weapon acquisition than those with standard internet access, and that all participants' plans still had critical failures.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A comparative result as the developer reports it in a footnote, for the model and the participant tasks it describes, with the post's own statement that all plans still had critical failures. It is not generalised beyond those experiments and does not quantify any uplift. The tasks are referred to only at the generality of the post's own wording.","evidenceIds":["evidence-ab-ev-anthropic-asl3-sonnet37-uplift-1"],"attributes":{"evaluatedSystem":"Claude Sonnet 3.7","evaluation":{"what":"Whether access to the model improved participants' performance on tasks related to CBRN weapon acquisition, against a standard internet baseline","how":"Human-participant experiments comparing a model-access condition with a standard internet condition","result":"Participants with model access did somewhat better than those with standard internet access","caveats":"The post states that all participants' plans still had critical failures, and gives the result as supporting evidence for an expectation about future models rather than as a full report."}}}],"sources":[{"id":"source-openai-launch","title":"Introducing OpenAI","authors":["Greg Brockman","Ilya Sutskever","OpenAI"],"publisher":"OpenAI","publicationDate":"2015-12-11","version":"Dated page, inspected at retrieval","canonicalUrl":"https://openai.com/index/introducing-openai/","evidenceUrl":"https://openai.com/index/introducing-openai/","language":"en","archivedUrl":"https://web.archive.org/web/20260917013950/https://openai.com/index/introducing-openai/","archivedAt":"2026-09-17","accessedAt":"2026-09-17","reuse":{"attribution":"OpenAI, Introducing OpenAI (2015-12-11).","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and minimal attributed excerpts only. Recheck terms before wider redistribution."}},{"id":"source-rsp-1-0","title":"Responsible Scaling Policy — v1.0","authors":["Anthropic"],"publisher":"Anthropic","publicationDate":"2023-09-19","version":"Responsible Scaling Policy v1.0; effective date 2023-09-19.","versionDate":"2023-09-19","canonicalUrl":"https://www-cdn.anthropic.com/1adf000c8f675958c2ee23805d91aaade1cd4613/responsible-scaling-policy.pdf","evidenceUrl":"https://www-cdn.anthropic.com/1adf000c8f675958c2ee23805d91aaade1cd4613/responsible-scaling-policy.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20230919161730/https://www-files.anthropic.com/production/files/responsible-scaling-policy-1.0.pdf","archivedAt":"2023-09-19","accessedAt":"2026-09-18","reuse":{"attribution":"Anthropic, Responsible Scaling Policy — v1.0.","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-rsp-3-1","title":"Responsible Scaling Policy — v3.1","authors":["Anthropic"],"publisher":"Anthropic","publicationDate":null,"version":"Responsible Scaling Policy v3.1; effective date 2026-04-02.","versionDate":"2026-04-02","canonicalUrl":"https://www-cdn.anthropic.com/files/4zrzovbb/website/bf04581e4f329735fd90634f6a1962c13c0bd351.pdf","evidenceUrl":"https://www-cdn.anthropic.com/files/4zrzovbb/website/bf04581e4f329735fd90634f6a1962c13c0bd351.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260407193305/https://www-cdn.anthropic.com/files/4zrzovbb/website/bf04581e4f329735fd90634f6a1962c13c0bd351.pdf","archivedAt":"2026-04-07","accessedAt":"2026-09-18","reuse":{"attribution":"Anthropic, Responsible Scaling Policy — v3.1.","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-pf-2","title":"Preparedness Framework — Version 2","authors":["OpenAI"],"publisher":"OpenAI","publicationDate":"2025-04-15","version":"Preparedness Framework Version 2; revision date 2025-04-15.","versionDate":"2025-04-15","canonicalUrl":"https://cdn.openai.com/pdf/18a02b5d-6b67-4cec-ab64-68cdfbddebcd/preparedness-framework-v2.pdf","evidenceUrl":"https://cdn.openai.com/pdf/18a02b5d-6b67-4cec-ab64-68cdfbddebcd/preparedness-framework-v2.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20250415191354/https://cdn.openai.com/pdf/18a02b5d-6b67-4cec-ab64-68cdfbddebcd/preparedness-framework-v2.pdf","archivedAt":"2025-04-15","accessedAt":"2026-09-18","reuse":{"attribution":"OpenAI, Preparedness Framework — Version 2.","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-fgf-2026","title":"Frontier Governance Framework — May 2026 edition","authors":["OpenAI"],"publisher":"OpenAI","publicationDate":"2026-05-28","version":"Frontier Governance Framework May 2026 edition; publication date 2026-05-28.","versionDate":"2026-05-28","canonicalUrl":"https://cdn.openai.com/pdf/e37d949b-8c9f-4d76-b99e-4272f4631a7e/openai-frontier-governance-framework.pdf","evidenceUrl":"https://cdn.openai.com/pdf/e37d949b-8c9f-4d76-b99e-4272f4631a7e/openai-frontier-governance-framework.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260529171758/https://cdn.openai.com/pdf/e37d949b-8c9f-4d76-b99e-4272f4631a7e/openai-frontier-governance-framework.pdf","archivedAt":"2026-05-29","accessedAt":"2026-09-18","reuse":{"attribution":"OpenAI, Frontier Governance Framework — May 2026 edition.","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-faif-2025","title":"Frontier Artificial Intelligence Framework — December 2025 edition","authors":["xAI"],"publisher":"xAI","publicationDate":null,"version":"Frontier Artificial Intelligence Framework December 2025 edition; revision date 2025-12-30.","versionDate":"2025-12-30","canonicalUrl":"https://data.x.ai/2025-12-31-xai-frontier-artificial-intelligence-framework.pdf","evidenceUrl":"https://data.x.ai/2025-12-31-xai-frontier-artificial-intelligence-framework.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260102171656/https://data.x.ai/2025-12-31-xai-frontier-artificial-intelligence-framework.pdf","archivedAt":"2026-01-02","accessedAt":"2026-09-18","reuse":{"attribution":"xAI, Frontier Artificial Intelligence Framework — December 2025 edition.","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-faif-2026","title":"Frontier Artificial Intelligence Framework — June 2026 edition","authors":["xAI"],"publisher":"xAI","publicationDate":null,"version":"Frontier Artificial Intelligence Framework June 2026 edition; effective date 2026-06-30. Visible public cover identifies the effective edition; stale embedded PDF draft metadata is not used to infer publication status.","versionDate":"2026-06-30","canonicalUrl":"https://media.x.ai/v1/website/xai-frontier-artificial-intelligence-framework-30-june-2026-99c40684.pdf","evidenceUrl":"https://media.x.ai/v1/website/xai-frontier-artificial-intelligence-framework-30-june-2026-99c40684.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260716003429/https://media.x.ai/v1/website/xai-frontier-artificial-intelligence-framework-30-june-2026-99c40684.pdf","archivedAt":"2026-07-16","accessedAt":"2026-09-18","reuse":{"attribution":"xAI, Frontier Artificial Intelligence Framework — June 2026 edition.","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-anthropic-aws-nov-2024","title":"Powering the next generation of AI development with AWS","authors":["Anthropic"],"publisher":"Anthropic","publicationDate":"2024-11-22","version":"Dated announcement inspected 2026-09-19; a historical company statement, not independent verification of implementation.","canonicalUrl":"https://www.anthropic.com/news/anthropic-amazon-trainium","evidenceUrl":"https://www.anthropic.com/news/anthropic-amazon-trainium","language":"en","archivedUrl":"https://web.archive.org/web/20260908143654/https://www.anthropic.com/news/anthropic-amazon-trainium","archivedAt":"2026-09-08","accessedAt":"2026-09-19","reuse":{"attribution":"Anthropic, Powering the next generation of AI development with AWS (2024-11-22).","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and editorial paraphrases only; no full source redistributed."}},{"id":"source-openai-chief-scientist-2024","title":"Ilya Sutskever to leave OpenAI, Jakub Pachocki announced as Chief Scientist","authors":["OpenAI"],"publisher":"OpenAI","publicationDate":"2024-05-14","version":"Page inspected 2026-09-19; historical statements are scoped to their source, not treated as current appointments.","canonicalUrl":"https://openai.com/index/jakub-pachocki-announced-as-chief-scientist/","evidenceUrl":"https://openai.com/index/jakub-pachocki-announced-as-chief-scientist/","language":"en","archivedUrl":"https://web.archive.org/web/20260910023924/https://openai.com/index/jakub-pachocki-announced-as-chief-scientist/","archivedAt":"2026-09-10","accessedAt":"2026-09-19","reuse":{"attribution":"OpenAI, Ilya Sutskever to leave OpenAI, Jakub Pachocki announced as Chief Scientist (2024-05-14)","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and brief attributed paraphrases only; no full source or personal contact details redistributed."}},{"id":"source-instructgpt-paper-v1","title":"Training language models to follow instructions with human feedback","authors":["Long Ouyang","Jeff Wu","Xu Jiang","Diogo Almeida","Carroll L. Wainwright","Pamela Mishkin","Chong Zhang","Sandhini Agarwal","Katarina Slama","Alex Ray","John Schulman","Jacob Hilton","Fraser Kelton","Luke Miller","Maddie Simens","Amanda Askell","Peter Welinder","Paul Christiano","Jan Leike","Ryan Lowe"],"publisher":"arXiv","publicationDate":"2022-03-04","version":"arXiv:2203.02155v1; first-submission date from version history. Later revisions are not substituted.","versionDate":"2022-03-04","canonicalUrl":"https://arxiv.org/abs/2203.02155","evidenceUrl":"https://arxiv.org/pdf/2203.02155v1","language":"en","accessedAt":"2026-09-20","reuse":{"attribution":"Long Ouyang; Jeff Wu; Xu Jiang; Diogo Almeida; Carroll L. Wainwright; Pamela Mishkin; Chong Zhang; Sandhini Agarwal; Katarina Slama; Alex Ray; John Schulman; Jacob Hilton; Fraser Kelton; Luke Miller; Maddie Simens; Amanda Askell; Peter Welinder; Paul Christiano; Jan Leike; Ryan Lowe, Training language models to follow instructions with human feedback (2022-03-04).","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and brief attributed paraphrases only; no full source, personal contact details or dataset imported."}},{"id":"source-eval-i-uk-approach","title":"AI Safety Institute approach to evaluations","authors":["UK AI Safety Institute"],"publisher":"Department for Science, Innovation and Technology","publicationDate":"2024-02-09","version":"Dated government report as retrieved 2026-09-27; retrospectively describes the November 2023 Summit.","canonicalUrl":"https://www.gov.uk/government/publications/ai-safety-institute-approach-to-evaluations/ai-safety-institute-approach-to-evaluations","evidenceUrl":"https://www.gov.uk/government/publications/ai-safety-institute-approach-to-evaluations/ai-safety-institute-approach-to-evaluations","language":"en","archivedUrl":"https://web.archive.org/web/20260917073729/https://www.gov.uk/government/publications/ai-safety-institute-approach-to-evaluations/ai-safety-institute-approach-to-evaluations","archivedAt":"2026-09-17","accessedAt":"2026-09-27","reuse":{"attribution":"UK AI Safety Institute, AI Safety Institute approach to evaluations.","licence":"Open Government Licence v3.0 except where otherwise stated","note":"Attributed paraphrase and link only; no original images or report reproduced."}},{"id":"source-cn-chatglm","title":"ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools","authors":["Team GLM"],"publisher":"Team GLM","publicationDate":"2024-06-18","version":"Source language: English (en). arXiv:2406.12793v1; joint Zhipu AI / Tsinghua University affiliation.","canonicalUrl":"https://arxiv.org/html/2406.12793v1","evidenceUrl":"https://arxiv.org/html/2406.12793v1","language":"en","accessedAt":"2026-09-29","reuse":{"attribution":"Team GLM, ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools (2024-06-18).","licence":"No redistribution licence asserted for this source expression","note":"Bibliographic metadata and brief attributed paraphrases only; model licensing is distinct from source-document reuse."}},{"id":"source-ms-fgf-2025","title":"Frontier Governance Framework — Version 1 (February 2025)","authors":["Microsoft"],"publisher":"Microsoft","publicationDate":"2025-02","version":"Frontier Governance Framework, Version 1 (February 2025); its change log reads “8 February 2025 – First version”. The file served at the recorded address (last modified 2025-02-12) is byte-identical to the Internet Archive capture of 2025-02-26, the earliest located.","versionDate":"2025-02-08","canonicalUrl":"https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/final/en-us/microsoft-brand/documents/Microsoft-Frontier-Governance-Framework.pdf","evidenceUrl":"https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/final/en-us/microsoft-brand/documents/Microsoft-Frontier-Governance-Framework.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20250226161803/https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/final/en-us/microsoft-brand/documents/Microsoft-Frontier-Governance-Framework.pdf","archivedAt":"2025-02-26","accessedAt":"2026-10-03","reuse":{"attribution":"Microsoft, Frontier Governance Framework, Version 1 (February 2025).","licence":"No open redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-delaware-ag-openai-2025","title":"Letter re: OpenAI, Inc.’s Corporate Restructuring (statement of no objection)","authors":["Delaware Department of Justice, Office of the Attorney General"],"publisher":"Delaware Department of Justice (Office of the Attorney General)","publicationDate":"2025-10-28","version":"Letter dated October 28, 2025, to OpenAI, Inc.; PDF inspected 2026-10-04, byte-identical to the Internet Archive capture of 2025-10-31. The earlier capture of 2025-10-28 differs only in also giving the recipient’s e-mail address.","canonicalUrl":"https://news.delaware.gov/files/2025/10/2025-10-28-OpenAI-DEAG-Statement-of-Nonobjection.pdf","evidenceUrl":"https://news.delaware.gov/files/2025/10/2025-10-28-OpenAI-DEAG-Statement-of-Nonobjection.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20251031110818/https://news.delaware.gov/files/2025/10/2025-10-28-OpenAI-DEAG-Statement-of-Nonobjection.pdf","archivedAt":"2025-10-31","accessedAt":"2026-10-04","reuse":{"attribution":"Delaware Department of Justice, letter to OpenAI, Inc. re: OpenAI, Inc.’s Corporate Restructuring (28 October 2025).","licence":"Public government document; no redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-eu-code-page","title":"The General-Purpose AI Code of Practice (Commission page with the list of signatories)","authors":["European Commission"],"publisher":"European Commission","publicationDate":null,"version":"Policy page inspected live 2026-10-04 (“Last update 31 July 2026”). The list of signatories first appears in the Internet Archive capture of 2025-08-01 (13:04 UTC), with 26 names and the sentence on xAI; the capture of 2025-07-31 (15:27 UTC) has no list. Publisher as printed on the inspected expression: European Commission (Shaping Europe’s digital future).","canonicalUrl":"https://digital-strategy.ec.europa.eu/en/policies/contents-code-gpai","evidenceUrl":"https://digital-strategy.ec.europa.eu/en/policies/contents-code-gpai","language":"en","archivedUrl":"https://web.archive.org/web/20250801130439/https://digital-strategy.ec.europa.eu/en/policies/contents-code-gpai","archivedAt":"2025-08-01","accessedAt":"2026-10-04","reuse":{"attribution":"European Commission, The General-Purpose AI Code of Practice (policy page).","licence":"European Union; quoted briefly under the editorial policy","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-eu-code-signing-qa","title":"Signing the General-Purpose AI Code of Practice – Questions & Answers","authors":["European Commission"],"publisher":"European Commission","publicationDate":"2025-07-17","version":"Questions and answers marked “Last update 17 July 2025”, inspected live 2026-10-04; an Internet Archive capture of 2025-07-17 exists and was not compared. Publisher as printed on the inspected expression: European Commission (Shaping Europe’s digital future).","canonicalUrl":"https://digital-strategy.ec.europa.eu/en/faqs/signing-general-purpose-ai-code-practice","evidenceUrl":"https://digital-strategy.ec.europa.eu/en/faqs/signing-general-purpose-ai-code-practice","language":"en","archivedUrl":"https://web.archive.org/web/20250717165024/https://digital-strategy.ec.europa.eu/en/faqs/signing-general-purpose-ai-code-practice","archivedAt":"2025-07-17","accessedAt":"2026-10-04","reuse":{"attribution":"European Commission, Signing the General-Purpose AI Code of Practice – Questions & Answers.","licence":"European Union","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-extropians-1996-08-10-burkhead","title":"intelligence increase (Extropians mailing list)","authors":["Lyle Burkhead"],"publisher":"Extropians mailing list","publicationDate":"1996-08-10","version":"Archive HTML as served, retrieved 2026-10-06 with the hash recorded by the 1996–2022 candidate search; quoted lines excluded when reading the author’s text. Archive at extropians.weidai.com.","canonicalUrl":"http://extropians.weidai.com/extropians.96/0220.html","evidenceUrl":"http://extropians.weidai.com/extropians.96/0220.html","language":"en","archivedUrl":"https://web.archive.org/web/20220523234349/https://extropians.weidai.com/extropians.96/0220.html","archivedAt":"2022-05-23","accessedAt":"2026-10-06","reuse":{"attribution":"Lyle Burkhead, \"intelligence increase\", Extropians mailing list, 1996-08-10.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-yudkowsky-2008-recursive-self-improvement","title":"Recursive Self-Improvement","authors":["Eliezer Yudkowsky"],"publisher":"Overcoming Bias","publicationDate":"2008-12-01","version":"Text checked in the post’s first stored version through the LessWrong GraphQL API on 2026-10-06; content version 1.0.0. Now hosted on LessWrong.","canonicalUrl":"https://www.lesswrong.com/posts/JBadX7rwdcRFzGuju/recursive-self-improvement","evidenceUrl":"https://www.lesswrong.com/posts/JBadX7rwdcRFzGuju/recursive-self-improvement","language":"en","archivedUrl":"https://web.archive.org/web/20190427074418/https://www.lesswrong.com/posts/JBadX7rwdcRFzGuju/recursive-self-improvement","archivedAt":"2019-04-27","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer Yudkowsky, \"Recursive Self-Improvement\", Overcoming Bias (now hosted on LessWrong), 2008-12-01.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-agi-ruin-2022","title":"AGI Ruin: A List of Lethalities","authors":["Eliezer Yudkowsky"],"publisher":"LessWrong and Alignment Forum","publicationDate":"2022-06-05","version":"Text checked in the post’s first stored version through the LessWrong GraphQL API on 2026-10-06.","canonicalUrl":"https://www.lesswrong.com/posts/uMQ3cqWDPHhjtiesc/agi-ruin-a-list-of-lethalities","evidenceUrl":"https://www.lesswrong.com/posts/uMQ3cqWDPHhjtiesc/agi-ruin-a-list-of-lethalities","language":"en","archivedUrl":"https://web.archive.org/web/20220605225854/https://www.lesswrong.com/posts/uMQ3cqWDPHhjtiesc/agi-ruin-a-list-of-lethalities","archivedAt":"2022-06-05","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer Yudkowsky, \"AGI Ruin: A List of Lethalities\", LessWrong and Alignment Forum, 2022-06-05.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-lw-2007-the-hidden-complexity-of-wishes","title":"The Hidden Complexity of Wishes","authors":["Eliezer Yudkowsky"],"publisher":"Overcoming Bias","publicationDate":"2007-11-24","version":"Post body retrieved through the LessWrong GraphQL API on 2026-10-06 (SHA-256 4512581ba1460d48f8cce9a1db8ad40841ca70e9480d9af6a81fb4308fcba9e0), content version 1.4.0; quotations also compared with the posting-day version 1.0.0. As imported to LessWrong.","canonicalUrl":"https://www.lesswrong.com/posts/4ARaTpNX62uaL86j6/the-hidden-complexity-of-wishes","evidenceUrl":"https://www.lesswrong.com/posts/4ARaTpNX62uaL86j6/the-hidden-complexity-of-wishes","language":"en","archivedUrl":"https://web.archive.org/web/20261004013825/https://www.lesswrong.com/posts/4ARaTpNX62uaL86j6/the-hidden-complexity-of-wishes","archivedAt":"2026-10-04","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer Yudkowsky, “The Hidden Complexity of Wishes”, Overcoming Bias, as imported to LessWrong, 2007-11-24.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-lw-2007-the-logical-fallacy-of-generalization-from","title":"The Logical Fallacy of Generalization from Fictional Evidence","authors":["Eliezer Yudkowsky"],"publisher":"Overcoming Bias","publicationDate":"2007-10-16","version":"Post body retrieved through the LessWrong GraphQL API on 2026-10-06 (SHA-256 53c3644a3afe461f6098180d2de85f7ed78e5b5a8ed7f68a0462e4a458321c65), content version 2.0.0; quotations also compared with the posting-day version 1.0.0. As imported to LessWrong.","canonicalUrl":"https://www.lesswrong.com/posts/rHBdcHGLJ7KvLJQPk/the-logical-fallacy-of-generalization-from-fictional","evidenceUrl":"https://www.lesswrong.com/posts/rHBdcHGLJ7KvLJQPk/the-logical-fallacy-of-generalization-from-fictional","language":"en","archivedUrl":"https://web.archive.org/web/20260920155055/https://www.lesswrong.com/posts/rHBdcHGLJ7KvLJQPk/the-logical-fallacy-of-generalization-from-fictional","archivedAt":"2026-09-20","accessedAt":"2026-10-06","reuse":{"attribution":"Eliezer Yudkowsky, “The Logical Fallacy of Generalization from Fictional Evidence”, Overcoming Bias, as imported to LessWrong, 2007-10-16.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-sl4-2007-11-24-yudkowsky-0711-17248","title":"Re: UCaRtMaAI paper (SL4 mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"SL4 mailing list","publicationDate":"2007-11-24","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 2c461930b012e60c76515f2def69dd1c3a92f282994b923b0d890a6f3244cae3); quoted lines excluded when reading the author’s text. The served archive shows later URL rewriting and does not authenticate historical bytes. Archive at sl4.org.","canonicalUrl":"http://sl4.org/archive/0711/17248.html","evidenceUrl":"http://sl4.org/archive/0711/17248.html","language":"en","archivedUrl":"https://web.archive.org/web/20250619223845/http://sl4.org/archive/0711/17248.html","archivedAt":"2025-06-19","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: UCaRtMaAI paper”, SL4 mailing list, 2007-11-24.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-eu-ai-act-2024","title":"Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act)","authors":["European Parliament and Council of the European Union"],"publisher":"Official Journal of the European Union (EUR-Lex)","publicationDate":"2024-07-12","version":"Official Journal L series edition of 12.7.2024, English HTML, retrieved 2026-10-10. The text of the Internet Archive capture of 2024-07-12 matched it apart from the numbering of footnote markers in the amending articles. Later amendments (Regulation (EU) 2026/1744) are not applied to this text.","canonicalUrl":"https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=OJ:L_202401689","evidenceUrl":"https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=OJ:L_202401689","language":"en","archivedUrl":"https://web.archive.org/web/20240712193938/https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=OJ:L_202401689","archivedAt":"2024-07-12","accessedAt":"2026-10-10","reuse":{"attribution":"European Parliament and Council of the European Union, Regulation (EU) 2024/1689 (Artificial Intelligence Act), Official Journal L, 12.7.2024.","licence":"Official legal text; no redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-ny-s8828-assembly-record","title":"New York S08828 (Gounardes) / A09449 (Bores): summary, actions and text of the bill (New York State Assembly bill search)","authors":["New York State Legislature"],"publisher":"New York State Legislature","publicationDate":"2026-03-27","version":"Bill-search page for S8828 retrieved 2026-10-11. It shows the summary, the action list ending 03/27/2026 SIGNED CHAP.96, and the bill as printed on 2026-01-08; the page for A09449 shows the same text and actions. The chapter as printed in the session laws was not inspected. No archive capture is cited: the Internet Archive refused lookup requests (HTTP 429) on 2026-10-11, so none was checked. Publisher as printed on the inspected expression: New York State Assembly (bill search and legislative information).","canonicalUrl":"https://nyassembly.gov/leg/?default_fld=&leg_video=&bn=S8828&term=2025&Summary=Y&Actions=Y&Text=Y","evidenceUrl":"https://nyassembly.gov/leg/?default_fld=&leg_video=&bn=S8828&term=2025&Summary=Y&Actions=Y&Text=Y","language":"en","accessedAt":"2026-10-11","reuse":{"attribution":"New York State Legislature, S08828 / A09449 (2026), as published on the New York State Assembly bill-search site.","licence":"Official document; no redistribution licence asserted","note":"Selekoa practice: metadata, attributed paraphrases and short attributed quotations under the editorial policy (at most 50 words each; at most 400 quoted words per document across the dataset, and about 10% of a document under about 2,000 words). Documents are linked externally, not redistributed. Publisher terms that forbid quotation override this practice."}},{"id":"source-benzell-et-al-2015-robots-are-us","title":"Robots Are Us: Some Economics of Human Replacement","authors":["Seth G. Benzell","Laurence J. Kotlikoff","Guillermo LaGarda","Jeffrey D. Sachs"],"publisher":"NBER Working Paper 20941","publicationDate":"2015-02","version":"Full text (revised version). Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 362172ebc8a725bfe76675a836f565736278dc12c86db96b5e67944451b0ae60. The February 2015 wording was not retrieved; the retrieved PDF is the October 2018 revision. Benzell, S. G., Kotlikoff, L. J., LaGarda, G., & Sachs, J. D. (2015). Robots are us: Some economics of human replacement (NBER Working Paper No. 20941; February 2015 version). National Bureau of Economic Research. https://doi.org/10.3386/w20941","canonicalUrl":"https://www.nber.org/system/files/working_papers/w20941/w20941.pdf","evidenceUrl":"https://www.nber.org/system/files/working_papers/w20941/w20941.pdf","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Seth G. Benzell, Laurence J. Kotlikoff, Guillermo LaGarda, Jeffrey D. Sachs, “Robots Are Us: Some Economics of Human Replacement”, NBER Working Paper 20941, 2015-02.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-acemoglu-restrepo-2016-race-between-machine-and-man","title":"The Race Between Machine and Man: Implications of Technology for Growth, Factor Shares and Employment","authors":["Daron Acemoglu","Pascual Restrepo"],"publisher":"NBER Working Paper 22252","publicationDate":"2016-05","version":"Full text (revised version). Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 441d01202afd56ef8002fc24ffc2beb51191741c0b5accb11d2534620dd616b7. Journal version not retrieved. Acemoglu, D., & Restrepo, P. (2018). The race between man and machine: Implications of technology for growth, factor shares, and employment. American Economic Review, 108(6), 1488–1542. https://doi.org/10.1257/aer.20160696","canonicalUrl":"https://www.nber.org/system/files/working_papers/w22252/w22252.pdf","evidenceUrl":"https://www.nber.org/system/files/working_papers/w22252/w22252.pdf","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Daron Acemoglu, Pascual Restrepo, “The Race Between Machine and Man: Implications of Technology for Growth, Factor Shares and Employment”, NBER Working Paper 22252, 2016-05.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-agrawal-gans-goldfarb-2019-ambiguous-labor-market-impact","title":"Artificial Intelligence: The Ambiguous Labor Market Impact of Automating Prediction","authors":["Ajay Agrawal","Joshua S. Gans","Avi Goldfarb"],"publisher":"NBER Working Paper 25619","publicationDate":"2019-02","version":"Full text. Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 dbd015967a47f77a41ff8028f5cc8208c3bb0e8cb44d6b1cc9be04f7df9c2450. Journal version not retrieved. Agrawal, A., Gans, J. S., & Goldfarb, A. (2019). Artificial intelligence: The ambiguous labor market impact of automating prediction. Journal of Economic Perspectives, 33(2), 31–50. https://doi.org/10.1257/jep.33.2.31","canonicalUrl":"https://www.nber.org/system/files/working_papers/w25619/w25619.pdf","evidenceUrl":"https://www.nber.org/system/files/working_papers/w25619/w25619.pdf","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Ajay Agrawal, Joshua S. Gans, Avi Goldfarb, “Artificial Intelligence: The Ambiguous Labor Market Impact of Automating Prediction”, NBER Working Paper 25619, 2019-02.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-acemoglu-restrepo-2019-automation-and-new-tasks","title":"Automation and New Tasks: How Technology Displaces and Reinstates Labor","authors":["Daron Acemoglu","Pascual Restrepo"],"publisher":"NBER Working Paper 25684","publicationDate":"2019-03","version":"Full text. Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 7757f77583c186dbbd0945e81e286fbb3ec35ed4e633ce65b2bb22f9f1bec490. Journal version not retrieved. Acemoglu, D., & Restrepo, P. (2019). Automation and new tasks: How technology displaces and reinstates labor. Journal of Economic Perspectives, 33(2), 3–30. https://doi.org/10.1257/jep.33.2.3","canonicalUrl":"https://www.nber.org/system/files/working_papers/w25684/w25684.pdf","evidenceUrl":"https://www.nber.org/system/files/working_papers/w25684/w25684.pdf","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Daron Acemoglu, Pascual Restrepo, “Automation and New Tasks: How Technology Displaces and Reinstates Labor”, NBER Working Paper 25684, 2019-03.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-cowen-2019-neglected-open-questions","title":"Neglected Open Questions in the Economics of Artificial Intelligence","authors":["Tyler Cowen"],"publisher":"NBER volume chapter","publicationDate":"2019-05","version":"Full text. Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 4501069cee0a059211427ad6e7d6ca65cb809f37623717d15b881350d677377d.","canonicalUrl":"https://www.nber.org/system/files/chapters/c14032/c14032.pdf","evidenceUrl":"https://www.nber.org/system/files/chapters/c14032/c14032.pdf","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Tyler Cowen, “Neglected Open Questions in the Economics of Artificial Intelligence”, NBER volume chapter, 2019-05.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-gw-sequoia-reflection-2025","title":"Partnering with Reflection: Toward Superintelligence, with Autonomous Coding","authors":["Stephanie Zhan","Charlie Curnin"],"publisher":"Sequoia Capital","publicationDate":"2025-03-07","version":"Dated page, retrieved on 2026-10-15 (SHA-256 5089a82bf34f432446896e540cfaee45f5bbbb9b8ff708724fe3f33578406dd0) directly from the publisher.","canonicalUrl":"https://sequoiacap.com/article/partnering-with-reflection-toward-superintelligence-with-autonomous-coding","evidenceUrl":"https://sequoiacap.com/article/partnering-with-reflection-toward-superintelligence-with-autonomous-coding","language":"en","archivedUrl":"https://web.archive.org/web/20261010210147/https://sequoiacap.com/article/partnering-with-reflection-toward-superintelligence-with-autonomous-coding","archivedAt":"2026-10-10","accessedAt":"2026-10-15","reuse":{"attribution":"Stephanie Zhan, Charlie Curnin, “Partnering with Reflection: Toward Superintelligence, with Autonomous Coding”, 2025-03-07.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-mx-ami-home","title":"AMI Labs: Real World. Real Intelligence.","authors":["AMI Labs (Advanced Machine Intelligence)"],"publisher":"AMI Labs (Advanced Machine Intelligence)","publicationDate":null,"version":"Undated page, retrieved on 2026-10-15 (SHA-256 b751879bf7ab1b430913a658b96e8de97327896e3f8d5e3f28b40b6637ce10ea) directly from the publisher. Its wording can change; earlier wording is not authenticated.","canonicalUrl":"https://amilabs.xyz/","evidenceUrl":"https://amilabs.xyz/","language":"en","archivedUrl":"https://web.archive.org/web/20261012151802/https://amilabs.xyz/","archivedAt":"2026-10-12","accessedAt":"2026-10-15","reuse":{"attribution":"AMI Labs (Advanced Machine Intelligence), “AMI Labs: Real World. Real Intelligence.”.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-mx-cnbc-lecun-leave","title":"Meta chief AI scientist Yann LeCun is leaving to create his own startup","authors":["Jonathan Vanian"],"publisher":"CNBC","publicationDate":"2025-11-19","version":"Dated page, retrieved on 2026-10-15 (SHA-256 03c6ef29337fb1920143a77b91c6165ed7f07c40e6326d4c4cf32c9d34c0aafe) through the third-party reader relay r.jina.ai, which renders the publisher’s current page (the publisher refused direct retrieval); not an archive copy.","canonicalUrl":"https://www.cnbc.com/2025/11/19/meta-chief-ai-scientist-yann-lecun-is-leaving-the-company-.html","evidenceUrl":"https://www.cnbc.com/2025/11/19/meta-chief-ai-scientist-yann-lecun-is-leaving-the-company-.html","language":"en","archivedUrl":"https://web.archive.org/web/20261013030039/https://www.cnbc.com/2025/11/19/meta-chief-ai-scientist-yann-lecun-is-leaving-the-company-.html","archivedAt":"2026-10-13","accessedAt":"2026-10-15","reuse":{"attribution":"Jonathan Vanian, “Meta chief AI scientist Yann LeCun is leaving to create his own startup”, 2025-11-19.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-new-murati-x-2025","title":"Mira Murati post on X, 18 February 2025","authors":["Mira Murati"],"publisher":"X","publicationDate":"2025-02-18","version":"Dated post, retrieved on 2026-10-15 (SHA-256 f1067d31607b53dacb18755ba284e380ac51bda77fa3f2784b42792ddc1d373e) through X’s own syndication endpoint for the post’s text, which truncates long notes.","canonicalUrl":"https://x.com/miramurati/status/1891918876029616494","evidenceUrl":"https://x.com/miramurati/status/1891918876029616494","language":"en","archivedUrl":"https://web.archive.org/web/20250218183304/https://twitter.com/miramurati/status/1891918876029616494","archivedAt":"2025-02-18","accessedAt":"2026-10-15","reuse":{"attribution":"Mira Murati, “Mira Murati post on X, 18 February 2025”, 2025-02-18.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-new-tc-tml-2025","title":"Thinking Machines Lab is ex-OpenAI CTO Mira Murati’s new startup","authors":["Kyle Wiggers"],"publisher":"TechCrunch","publicationDate":"2025-02-18","version":"Dated page, retrieved on 2026-10-15 (SHA-256 18f828471b78936133db30ef4292b4e342e518cd8455050a7cc0e893b8ce3b61) directly from the publisher.","canonicalUrl":"https://techcrunch.com/2025/02/18/thinking-machines-lab-is-ex-openai-cto-mira-muratis-new-startup/","evidenceUrl":"https://techcrunch.com/2025/02/18/thinking-machines-lab-is-ex-openai-cto-mira-muratis-new-startup/","language":"en","archivedUrl":"https://web.archive.org/web/20261012151415/https://techcrunch.com/2025/02/18/thinking-machines-lab-is-ex-openai-cto-mira-muratis-new-startup/","archivedAt":"2026-10-12","accessedAt":"2026-10-15","reuse":{"attribution":"Kyle Wiggers, “Thinking Machines Lab is ex-OpenAI CTO Mira Murati’s new startup”, 2025-02-18.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-oai-foundation-2026","title":"Update on the OpenAI Foundation","authors":["OpenAI"],"publisher":"OpenAI","publicationDate":"2026-03-24","version":"Dated page, retrieved on 2026-10-15 (SHA-256 286bcf88ffe02eba017a7adb681b86886a8990604f1b0b8f08690b0d74efc8c3) through the third-party reader relay r.jina.ai, which renders the publisher’s current page (the publisher refused direct retrieval); not an archive copy.","canonicalUrl":"https://openai.com/index/update-on-the-openai-foundation/","evidenceUrl":"https://openai.com/index/update-on-the-openai-foundation/","language":"en","archivedUrl":"https://web.archive.org/web/20260925130606/https://openai.com/index/update-on-the-openai-foundation/","archivedAt":"2026-09-25","accessedAt":"2026-10-15","reuse":{"attribution":"OpenAI, “Update on the OpenAI Foundation”, 2026-03-24.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-oai-tc-mcgrew-2024","title":"OpenAI’s chief research officer has left following CTO Mira Murati’s exit","authors":["Kyle Wiggers"],"publisher":"TechCrunch","publicationDate":"2024-09-25","version":"Dated page, retrieved on 2026-10-15 (SHA-256 55f6630b17d7f073216fecc85a4fc17bd1c93a094433fa6983bd3dac2f0c5f3d) directly from the publisher. Reporting used because Altman’s and McGrew’s notes were posted on X as long-form or image posts whose full text could not be retrieved.","canonicalUrl":"https://techcrunch.com/2024/09/25/openais-chief-research-officer-has-left/","evidenceUrl":"https://techcrunch.com/2024/09/25/openais-chief-research-officer-has-left/","language":"en","archivedUrl":"https://web.archive.org/web/20260730012154/https://techcrunch.com/2024/09/25/openais-chief-research-officer-has-left/","archivedAt":"2026-07-30","accessedAt":"2026-10-15","reuse":{"attribution":"Kyle Wiggers, “OpenAI’s chief research officer has left following CTO Mira Murati’s exit”, 2024-09-25.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"lg-src-oai-zoph-x-2024","title":"Post of 26 September 2024 (UTC) sharing his note on leaving OpenAI","authors":["Barret Zoph"],"publisher":"X","publicationDate":"2024-09-26","version":"Dated post, retrieved on 2026-10-15 (SHA-256 b13d334b7bcbff9083dd5e3c7d5e9d44c96ac2651fd00eae7562af5f3069d091) through X’s own syndication endpoint for the post’s text, which truncates long notes. Read through X’s syndication endpoint (text only; the note is truncated in that text). Publication date is the post’s UTC timestamp (25 September in US Pacific time).","canonicalUrl":"https://x.com/barret_zoph/status/1839095143397515452","evidenceUrl":"https://x.com/barret_zoph/status/1839095143397515452","language":"en","archivedUrl":"https://web.archive.org/web/20240926001025/https://twitter.com/barret_zoph/status/1839095143397515452","archivedAt":"2024-09-26","accessedAt":"2026-10-15","reuse":{"attribution":"Barret Zoph, “Post of 26 September 2024 (UTC) sharing his note on leaving OpenAI”, 2024-09-26.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"c2-src-kimi-k2-pdf-v1","title":"Kimi K2: Open Agentic Intelligence (technical report, v1)","authors":["Kimi Team"],"publisher":"arXiv","publicationDate":"2025-07-28","version":"Dated document, retrieved on 2026-10-27 (SHA-256 b838917cf0a475d2e1b2f1ba4d0b4c91b6f6c8be95f15aa0d4f3ea76e4ef60f5) directly from the publisher. Version: 2507.20534v1. PDF of version 1; later versions not inspected.","canonicalUrl":"https://arxiv.org/pdf/2507.20534v1","evidenceUrl":"https://arxiv.org/pdf/2507.20534v1","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"Kimi Team, “Kimi K2: Open Agentic Intelligence (technical report, v1)”, 2025-07-28.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"c2-src-tc260-asgf-v1-news","title":"《人工智能安全治理框架》1.0版发布","authors":["全国网络安全标准化技术委员会"],"publisher":"全国网络安全标准化技术委员会","publicationDate":"2024-09-10","version":"Dated page, retrieved on 2026-10-27 (SHA-256 a4787be3c432b9f828799164856c8578abb5fff0975088b4583baddb82edb5ac) directly from the publisher. TC260 news item, page dated 2024-09-10, read as it stood on 2026-10-27; no attachment was linked on the page at retrieval. Its wording can change; earlier wording is not authenticated. English title (machine translation): “AI Safety Governance Framework version 1.0 released”. Publisher as printed on the inspected expression: 全国网络安全标准化技术委员会 (tc260.org.cn).","canonicalUrl":"https://www.tc260.org.cn/tc260/xwdt1/202409/ca4adb6f32474a57a3e504023780a638.shtml","evidenceUrl":"https://www.tc260.org.cn/tc260/xwdt1/202409/ca4adb6f32474a57a3e504023780a638.shtml","language":"zh-Hans","archivedUrl":"https://web.archive.org/web/20261028222121/https://www.tc260.org.cn/tc260/xwdt1/202409/ca4adb6f32474a57a3e504023780a638.shtml","archivedAt":"2026-10-28","accessedAt":"2026-10-27","reuse":{"attribution":"全国网络安全标准化技术委员会, “《人工智能安全治理框架》1.0版发布”, 2024-09-10.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"c2-src-tc260-asgf-v1-pdf","title":"人工智能安全治理框架（V1.0）","authors":["全国网络安全标准化技术委员会"],"publisher":"全国网络安全标准化技术委员会","publicationDate":"2024-09","version":"Dated document, retrieved on 2026-10-27 (SHA-256 0cbf340164593f5fd98d82406979e43e82ce7505ea7bd2a0fd19a00dfd09412d) directly from the publisher. Version: 1727568303900999.pdf. Chinese edition linked from the CAC news item of 9 September 2024. The TC260 upload URL cited in later documents (tc260.org.cn/upload/2024-09-09/1725849192841090989.pdf) now redirects to TC260's home page; the English edition linked from the CAC page was not inspected. English title (machine translation): “AI Safety Governance Framework (V1.0)”. Publisher as printed on the inspected expression: 全国网络安全标准化技术委员会 (hosted by the Cyberspace Administration of China, cac.gov.cn).","canonicalUrl":"https://www.cac.gov.cn/rootimages/uploadimg/1727568303900999/1727568303900999.pdf","evidenceUrl":"https://www.cac.gov.cn/rootimages/uploadimg/1727568303900999/1727568303900999.pdf","language":"zh-Hans","archivedUrl":"https://web.archive.org/web/20261017172618/https://www.cac.gov.cn/rootimages/uploadimg/1727568303900999/1727568303900999.pdf","archivedAt":"2026-10-17","accessedAt":"2026-10-27","reuse":{"attribution":"全国网络安全标准化技术委员会, “人工智能安全治理框架（V1.0）”, 2024-09.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"c2-src-tc260-asgf-v2-pdf","title":"人工智能安全治理框架2.0","authors":["全国网络安全标准化技术委员会","国家计算机网络应急技术处理协调中心"],"publisher":"全国网络安全标准化技术委员会","publicationDate":"2025-09","version":"Dated document, retrieved on 2026-10-27 (SHA-256 50735b9ca01bff746169fa58e554abb97b4e62c27d4c97afd64f18ff83bba705) directly from the publisher. Version: 1759653474200838.pdf. Bilingual PDF (Chinese text followed by the publisher’s English text) linked from the CAC news item of 15 September 2025. Quotes are from the Chinese text, transcribed as printed: the text layer contains control characters (U+0001) and compatibility forms (Kangxi radicals, ⻛ for 风, ⻅ for 见), which are normalised in the quotations. No copy on tc260.org.cn was located. English title (machine translation): “AI Safety Governance Framework 2.0”. Publisher as printed on the inspected expression: 全国网络安全标准化技术委员会 (hosted by the Cyberspace Administration of China, cac.gov.cn).","canonicalUrl":"https://www.cac.gov.cn/rootimages/uploadimg/1759653474200838/1759653474200838.pdf","evidenceUrl":"https://www.cac.gov.cn/rootimages/uploadimg/1759653474200838/1759653474200838.pdf","language":"zh-Hans","archivedUrl":"https://web.archive.org/web/20260302034821/https://www.cac.gov.cn/rootimages/uploadimg/1759653474200838/1759653474200838.pdf","archivedAt":"2026-03-02","accessedAt":"2026-10-27","reuse":{"attribution":"全国网络安全标准化技术委员会, 国家计算机网络应急技术处理协调中心, “人工智能安全治理框架2.0”, 2025-09.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-cn-generative-ai-measures-2023-govcn","title":"生成式人工智能服务管理暂行办法 (Interim Measures for the Management of Generative Artificial Intelligence Services), gov.cn State Council policy library copy","authors":["国家互联网信息办公室 (Cyberspace Administration of China) and six other departments"],"publisher":"中国政府网 (www.gov.cn), 国务院政策文件库","publicationDate":"2023-07-10","version":"Retrieved on 2026-10-27 (SHA-256 00bd290a9d14e7354b7131d30255daae26ce7ff73588483cbeb2030f56878277). A second posting of the text of the dataset's source for the Measures (the CAC's posting of 13 July 2023): the gov.cn State Council policy library page, which gives its source as 国家网信办网站 and its document date (成文日期) as 2023-07-10. The page body repeats the text twice (a print copy). The two postings were not compared word for word; every passage cited from this source was matched in this copy. The quotation limit is applied to this source and the CAC posting together. The English in parentheses in the title is a machine translation.","canonicalUrl":"https://www.gov.cn/zhengce/zhengceku/202307/content_6891752.htm","evidenceUrl":"https://www.gov.cn/zhengce/zhengceku/202307/content_6891752.htm","language":"zh-Hans","accessedAt":"2026-10-27","reuse":{"attribution":"国家互联网信息办公室 (Cyberspace Administration of China) and six other departments, “生成式人工智能服务管理暂行办法 (Interim Measures for the Management of Generative Artificial Intelligence Services), gov.cn State Council policy library copy”, 2023-07-10.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-us-eo-14409-fr","title":"Executive Order 14409 of June 2, 2026: Promoting Advanced Artificial Intelligence Innovation and Security (91 FR 34565)","authors":["The White House"],"publisher":"The White House","publicationDate":"2026-06-05","version":"Retrieved on 2026-10-27 (SHA-256 4a25c970947e7234f1a4308aca6aba62f327ac956de7caa41204c84f7f2cdcd9). Federal Register print of 5 June 2026, Vol. 91, No. 108, pp. 34565–34567, FR Doc. 2026-11415. Publisher as printed on the inspected expression: Federal Register (Office of the Federal Register, National Archives) via govinfo.gov.","canonicalUrl":"https://www.govinfo.gov/content/pkg/FR-2026-06-05/pdf/2026-11415.pdf","evidenceUrl":"https://www.govinfo.gov/content/pkg/FR-2026-06-05/pdf/2026-11415.pdf","language":"en","accessedAt":"2026-10-27","reuse":{"attribution":"The White House, “Executive Order 14409 of June 2, 2026: Promoting Advanced Artificial Intelligence Innovation and Security (91 FR 34565)”, 2026-06-05.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-cpo-ch-intel-usg-8k-0827","title":"Intel Corporation Form 8-K, date of report August 27, 2025 (closing of the Warrant and Common Stock Agreement)","authors":["Intel Corporation (SEC filing as mirrored on intc.com)"],"publisher":"Intel Corporation (SEC filing as mirrored on intc.com)","publicationDate":"2025-08-29","version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 c526431fd0e306c4d8bf0d5ce4375db2de379b3a3f9263e2f2441fd9783f4e01). Form 8-K, date of earliest event reported August 27, 2025, signed August 29, 2025 (accession 0000050863-25-000135), as mirrored on intc.com and served on 2026-10-28. Retrieved again at admission on 2026-10-30 (SHA-256 c526431fd0e306c4d8bf0d5ce4375db2de379b3a3f9263e2f2441fd9783f4e01, the same bytes).","canonicalUrl":"https://www.intc.com/filings-reports/all-sec-filings/content/0000050863-25-000135/intc-20250827.htm","evidenceUrl":"https://www.intc.com/filings-reports/all-sec-filings/content/0000050863-25-000135/intc-20250827.htm","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Intel Corporation (SEC filing as mirrored on intc.com), “Intel Corporation Form 8-K, date of report August 27, 2025 (closing of the Warrant and Common Stock Agreement)”, 2025-08-29.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-cpo-dc-msft-golden-opportunity-2025","title":"The golden opportunity for American AI","authors":["Microsoft (Microsoft On the Issues blog; Brad Smith, Vice Chair & President)"],"publisher":"Microsoft (Microsoft On the Issues blog; Brad Smith, Vice Chair & President)","publicationDate":"2025-01-03","version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 6e20428abe39ac501400292d1da05ac6837db5800ae18059fe775bc782ef9fdd). Blog post as served on 2026-10-28; byline \"Jan 3, 2025 | Brad Smith - Vice Chair & President\". Retrieved again at admission on 2026-10-30 (SHA-256 6e20428abe39ac501400292d1da05ac6837db5800ae18059fe775bc782ef9fdd, the same bytes). Its wording can change; earlier wording is not authenticated.","canonicalUrl":"https://blogs.microsoft.com/on-the-issues/2025/01/03/the-golden-opportunity-for-american-ai/","evidenceUrl":"https://blogs.microsoft.com/on-the-issues/2025/01/03/the-golden-opportunity-for-american-ai/","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Microsoft (Microsoft On the Issues blog; Brad Smith, Vice Chair & President), “The golden opportunity for American AI”, 2025-01-03.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-ab-hhs-2010-fr","title":"Screening Framework Guidance for Providers of Synthetic Double-Stranded DNA (75 FR 62820)","authors":["Federal Register (Office of the Federal Register, National Archives) via govinfo.gov"],"publisher":"Federal Register (Office of the Federal Register, National Archives) via govinfo.gov","publicationDate":null,"version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 d3bfd95a1e3ae4d212505081166f0f8bb7475845e225ae643e6d4a03a93b7e56). Federal Register print of 13 October 2010, Vol. 75, No. 197, pp. 62820-62832, FR Doc. 2010-25728. The notice publishes the final Guidance in full, after the comment responses. Retrieved again at admission on 2026-10-30 (SHA-256 d3bfd95a1e3ae4d212505081166f0f8bb7475845e225ae643e6d4a03a93b7e56, the same bytes).","canonicalUrl":"https://www.govinfo.gov/content/pkg/FR-2010-10-13/pdf/2010-25728.pdf","evidenceUrl":"https://www.govinfo.gov/content/pkg/FR-2010-10-13/pdf/2010-25728.pdf","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Federal Register (Office of the Federal Register, National Archives) via govinfo.gov, “Screening Framework Guidance for Providers of Synthetic Double-Stranded DNA (75 FR 62820)”.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-ab-igsc-hsp-v31","title":"IGSC Harmonized Screening Protocol v3.1","authors":["International Gene Synthesis Consortium (genesynthesisconsortium.org)"],"publisher":"International Gene Synthesis Consortium (genesynthesisconsortium.org)","publicationDate":null,"version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 9f6ea4906da571f0fae88e6b24f3b68dfe485ff19b88f849882ebe56f03a2fd3). Version 3.1, dated 1 June 2026 on its first page, as served on 2026-10-28. Its screening-data and methodology sections (sections 6 and the appendices) were read but are not cited. Retrieved again at admission on 2026-10-30 (SHA-256 9f6ea4906da571f0fae88e6b24f3b68dfe485ff19b88f849882ebe56f03a2fd3, the same bytes).","canonicalUrl":"https://genesynthesisconsortium.org/wp-content/uploads/IGSC-Harmonized-Screening-Protocol-V3.1.pdf","evidenceUrl":"https://genesynthesisconsortium.org/wp-content/uploads/IGSC-Harmonized-Screening-Protocol-V3.1.pdf","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"International Gene Synthesis Consortium (genesynthesisconsortium.org), “IGSC Harmonized Screening Protocol v3.1”.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}},{"id":"source-ab-asl3-news","title":"Activating AI Safety Level 3 protections","authors":["Anthropic"],"publisher":"Anthropic","publicationDate":null,"version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 411f66885bacd8769dd6718e78f40eb0df80a4983dbde9d0d49f0323746e8656). Page labelled 'Published May 22, 2025', with six footnotes, as served on 2026-10-28. The accompanying full report (a separate PDF) was not retrieved. Retrieved again at admission on 2026-10-30 (SHA-256 35fa989dfc4f36842ef4f1dda2cfbdda9f9fc9f63524b310fd869ed2f32ab1e8). Its wording can change; earlier wording is not authenticated.","canonicalUrl":"https://www.anthropic.com/news/activating-asl3-protections","evidenceUrl":"https://www.anthropic.com/news/activating-asl3-protections","language":"en","accessedAt":"2026-10-28","reuse":{"attribution":"Anthropic, “Activating AI Safety Level 3 protections”.","licence":"No open redistribution licence asserted","note":"Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution."}}],"evidence":[{"id":"evidence-faif-2026-compliance-sentence","sourceRole":"primary","support":"supports","locator":"PDF pp. 1–9, complete document; compare December 2025 PDF p. 1","passage":"The inspected June edition has no sentence asserting TFAIA compliance; it still discusses applicable laws and compliance reviews.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-18","note":"Official source inspected by machine; passage is an editorial paraphrase. Human review pending."},{"id":"evidence-faif-2026-compliance-sentence-2","sourceRole":"primary","support":"supports","locator":"PDF p. 1, Introduction, sentence before Scope (opening words; the sentence continues with the statute citation)","passage":"This FAIF complies with California's Transparency in Frontier Artificial Intelligence Act","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-03","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-03 against the inspected PDF, byte-identical to the Internet Archive capture of 2026-01-02; it replaces an earlier attributed paraphrase. Human editorial review is pending."},{"id":"evidence-anthropic-aws-trainium-plan-2024","sourceRole":"primary","support":"supports","locator":"Collaboration on AWS Trainium hardware and software, second paragraph","passage":"Anthropic planned to train its most advanced models using the jointly optimized hardware.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-19","note":"Exact cited source passage inspected by a machine agent. Company statements are not independent proof of implementation. Human source review pending."},{"id":"evidence-brockman-stripe-cto-prior","sourceRole":"primary","support":"supports","locator":"OpenAI section, CTO sentence","passage":"Brockman is described as previously holding the CTO role at Stripe.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-19","note":"Source passage independently inspected by an automated research assistant for this scoped record. Human source review pending."},{"id":"evidence-sutskever-openai-departure-announced-2024","sourceRole":"primary","support":"supports","locator":"Message from Sam to the company, first paragraph","passage":"Sam Altman announces that Sutskever and OpenAI will part ways.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-19","note":"Source passage independently inspected by an automated research assistant for this scoped record. Human source review pending."},{"id":"evidence-lineage-instruct-cites-ppo","sourceRole":"primary","support":"supports","locator":"v1 §3.5, Reinforcement learning, PDF p. 9","passage":"The RL paragraph cites Schulman et al. (2017).","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-20","note":"Relevant source passage inspected by machine. Primary for what this document states; no independent replication or human review. Existing source access dates are retained; this evidence check has its own date."},{"id":"evidence-eval-i-uk-metr-demonstrations","sourceRole":"primary","support":"supports","locator":"AI Safety Summit demonstrations > Case study 3: autonomous systems > AI agents can plan and carry out complex actions in the real world, opening paragraph.","passage":"AISI says it worked with METR to showcase published and unpublished material at the Summit.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-27","note":"Retrospective government attribution; publication and event dates kept separate. No human review."},{"id":"evidence-cn-tang-chatglm","sourceRole":"primary","support":"supports","locator":"Footnote 1, alphabetically ordered Team GLM list","passage":"The team list includes Jie Tang.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-29","note":"English source paraphrased and checked by machine. Exploratory seed, not editorially accepted. Source reporting is not independent verification of implementation."},{"id":"evidence-prov-ms-fgf-2025-cbrn-high-threshold","sourceRole":"primary","support":"supports","locator":"PDF p. 12 (printed p. 11), Appendix I – Capability thresholds, CBRN weapons table, High row, Capability threshold column (both statements)","passage":"The model provides a meaningful uplift to an expert’s (PhD level education in related fields) ability to develop and deliver a novel CBRN threat. The model provides a meaningful uplift to a medium-skilled actor’s (e.g., STEM education) ability to develop and deliver a known CBRN threat.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-03","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-03 against the inspected PDF, byte-identical to the Internet Archive capture of 2025-02-26. Extraction note: Row assignment checked with pdftotext -raw/-table (the plain text stacks the level labels). The table does not say whether both statements or either one must hold. The row's deployment requirement reads 'Further review and mitigations required.' Human editorial review is pending."},{"id":"evidence-prov-frontier-governance-2026-precautionary-threshold-treatment","sourceRole":"primary","support":"supports","locator":"PDF p. 8 (printed p. 06), §2.3 Systemic risk acceptance determination, first paragraph, final sentence","passage":"For example, out of an abundance of caution we have treated models as crossing a capability threshold in circumstances where we are unable to rule out that a new threshold had been reached, even in the absence of direct evidence that it has occurred.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-03","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-03 against the inspected PDF, byte-identical (after decompression) to the Internet Archive capture of 2026-05-29. Extraction note: Framed as an example of past practice illustrating the preceding sentence (reviewing risk tiers and incorporating appropriate safety margins when assessing whether to accept residual risk); it is not stated as a rule. Human editorial review is pending."},{"id":"evidence-prov-delaware-ag-nfp-control-1","sourceRole":"primary","support":"supports","locator":"Representation 1 (page 2)","passage":"So long as the NFP holds Class N Common Stock, the NFP will retain control and oversight over the newly formed PBC, including the sole power and authority of the NFP Board to appoint members of the Board of Directors of the PBC","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-04","note":"Verbatim quote, re-checked on 2026-10-04 by the integrating session against its own direct fetch of the cited expression (mechanical match after whitespace normalisation). The passage was first proposed and machine-verified in the week 4b research pass. Human editorial review is pending."},{"id":"evidence-cm-link-ms-fgf-2025-seoul-1","sourceRole":"primary","support":"supports","locator":"Section 1, first paragraph, second sentence (PDF p. 3; printed p. 2)","passage":"The framework has its genesis in the voluntary Frontier AI Safety Commitments that Microsoft and fifteen other AI labs made in May 2024","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-04","note":"Verbatim quote, re-checked on 2026-10-04 by the integrating session against its own direct fetch of the cited expression (mechanical match after whitespace normalisation). The passage was first proposed and machine-verified in the week 4b research pass. Human editorial review is pending."},{"id":"evidence-cm-eu-code-mistral-ai-1","sourceRole":"primary","support":"supports","locator":"Section “Signatories of Code of Practice”, in the Internet Archive capture of 2025-08-01","passage":"The list as first published has 26 names, including “Mistral AI”.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-04","note":"Editorial paraphrase, re-checked on 2026-10-04 by the integrating session against its own direct fetch of the cited expression. The integrating session wrote or selected this passage from its own reading of the source; it does not repeat a passage from the week 4b research pass. Human editorial review is pending."},{"id":"evidence-cm-eu-code-mistral-ai-2","sourceRole":"primary","support":"context","locator":"Section “Signatories of the code of practice”, live page (last update 31 July 2026)","passage":"The current list of 21 names still includes “Mistral AI”.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-04","note":"Editorial paraphrase, re-checked on 2026-10-04 by the integrating session against its own direct fetch of the cited expression. The integrating session wrote or selected this passage from its own reading of the source; it does not repeat a passage from the week 4b research pass. Human editorial review is pending."},{"id":"evidence-cm-eu-code-mistral-ai-3","sourceRole":"primary","support":"context","locator":"Answer to “What does signing the code imply in terms of compliance with the AI Act?”","passage":"The questions and answers say that any opt-out from chapters of the code loses the benefit of facilitating the demonstration of compliance in that respect.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-04","note":"Editorial paraphrase, re-checked on 2026-10-04 by the integrating session against its own direct fetch of the cited expression. The integrating session wrote or selected this passage from its own reading of the source; it does not repeat a passage from the week 4b research pass. Human editorial review is pending."},{"id":"evidence-ts-a1-burkhead-1996","sourceRole":"primary","support":"supports","locator":"Message body, first paragraph of the reposted letter","passage":"Good proposes [...] a computer more intelligent than a man [...] programmed to design a still more intelligent computer","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Checked against the archive HTML on 2026-10-06 (hash matched the 1996–2022 candidate search)."},{"id":"evidence-ts-a3-yudkowsky-2008-rsi","sourceRole":"primary","support":"supports","locator":"Paragraph beginning \"Recursion is the sort of thing that happens\"","passage":"Recursion is the sort of thing that happens when you hand the AI the object-level problem of \"redesign your own cognitive algorithms\".","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-28","note":"Checked in the first stored version on 2026-10-06. On 2026-10-30, by the maintainer’s decision of that day, the quotation was corrected to the archive capture of 2019-04-27 (quotation marks and apostrophes as in the capture), which a cross-family archive check read on 2026-10-28; the proposition is unchanged."},{"id":"evidence-yk-t20-a","sourceRole":"primary","support":"supports","locator":"The Logical Fallacy of Generalization from Fictional Evidence, paragraph beginning 'Any professional negotiator knows'","passage":"If you start out by thinking of The Matrix, it brings to mind marching robot armies defeating humans after a long struggle—not a superintelligence snapping nanotechnological fingers.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Candidate pack passage T20-a. The quotation (27 words) matched the retrieved expression (SHA-256 53c3644a3afe461f6098180d2de85f7ed78e5b5a8ed7f68a0462e4a458321c65) when checked on 2026-10-06. It also matched the posting-day version 1.0.0. Pack version note: edited post, current version 2.0.0; posting-day check pending."},{"id":"evidence-yk-t29-a","sourceRole":"primary","support":"supports","locator":"Section B.4 Miscellaneous unworkable schemes, item 34","passage":"a \"multipolar\" system of 20 superintelligences … has a natural and obvious equilibrium which looks like \"the 20 superintelligences cooperate with each other but not with humanity\".","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-06","note":"Candidate pack passage T29-a. The quotation (27 words; trimmed 2026-10-13 from 32 words) matched the retrieved expression (SHA-256 b36f38a6fd1020bfbd7b9d941e150ea68561242f3a41a0b8f2a56b853a9a6db3) when checked on 2026-10-06. It also matched the posting-day version 1.0.0. Pack version note: edited post, current version 1.3.0; posting-day check pending."},{"id":"evidence-aia-art55-incident-reporting-1","sourceRole":"primary","support":"supports","locator":"Article 55(1), point (c)","passage":"report, without undue delay, to the AI Office and, as appropriate, to national competent authorities, relevant information about serious incidents","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-10","note":"Verbatim quote, matched mechanically (after whitespace normalisation) by the proposing session against the text retrieved on 2026-10-10. Checked by fresh-context machine audit on 2026-10-10 (review/frontier-governance-audit-2026-10-10.md). Human editorial review is pending."},{"id":"evidence-aia-art55-incident-reporting-2","sourceRole":"primary","support":"supports","locator":"Article 55(1), point (c), remainder","passage":"The provider is also to keep track of and document the incidents and possible corrective measures to address them.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-10","note":"Editorial paraphrase proposed by the proposing session from the text retrieved on 2026-10-10. Checked by fresh-context machine audit on 2026-10-10 (review/frontier-governance-audit-2026-10-10.md). Human editorial review is pending."},{"id":"evidence-ny-ch96-effective-date-1","sourceRole":"primary","support":"supports","locator":"Section 3 of the bill (text of amended section 3)","passage":"This act shall take effect [on the ninetieth day after it shall have become a law] January 1, 2027.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-11","note":"Verbatim quote, matched mechanically (after whitespace normalisation) by the proposing machine agent against the text extracted from the file retrieved on 2026-10-11. Checked by fresh-context machine audit on 2026-10-11 (review/frontier-governance-2-audit-2026-10-11.md). Human editorial review is pending."},{"id":"evidence-ec-m02-21","sourceRole":"primary","support":"supports","locator":"pdf page 2; Abstract","passage":"Automation, which enables capital to replace labor in tasks it was previously engaged in, shifts the task content of production against labor because of a displacement effect.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M02-21 (research/economics-mainstream). The quotation (27 words) matched the retrieved current copy (SHA-256 7757f77583c186dbbd0945e81e286fbb3ec35ed4e633ce65b2bb22f9f1bec490) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"evidence-ec-m03-07","sourceRole":"primary","support":"supports","locator":"pdf page 3; introduction (printed p. 1)","passage":"This wide breadth of application means that developments in artificial intelligence represent what Bresnahan and Trajtenberg (1995) called a \"general purpose technology.\"","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M03-07 (research/economics-mainstream). The quotation (22 words) matched the retrieved current copy (SHA-256 dbd015967a47f77a41ff8028f5cc8208c3bb0e8cb44d6b1cc9be04f7df9c2450) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"evidence-ec-m05-21","sourceRole":"primary","support":"supports","locator":"pdf page 3; section 1 Introduction (printed p. 1)","passage":"the difference between human labor and horses is that humans have a comparative advantage in new and more complex tasks. Horses did not.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M05-21 (research/economics-mainstream). The quotation (23 words) matched the retrieved current copy (SHA-256 441d01202afd56ef8002fc24ffc2beb51191741c0b5accb11d2534620dd616b7) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"evidence-ec-m06a-18","sourceRole":"primary","support":"supports","locator":"pdf page 28; Section 7, Conclusion (printed p. 26)","passage":"But it does firmly predict three things - a long-run decline in labor's share of income (which appears underway among OECD members)","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M06a-18 (research/economics-mainstream). The quotation (22 words) matched the retrieved current copy (SHA-256 362172ebc8a725bfe76675a836f565736278dc12c86db96b5e67944451b0ae60) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"evidence-ec-m09-32","sourceRole":"primary","support":"supports","locator":"pdf page 6; 15.3 The Political Economy of Artificial Intelligence and Income Redistribution, final paragraph (printed p. 395)","passage":"That is a frontier issue where we economists haven't done much work at all, but the ongoing progress of AI may make those questions all the more relevant.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M09-32 (research/economics-mainstream). The quotation (28 words) matched the retrieved current copy (SHA-256 4501069cee0a059211427ad6e7d6ca65cb809f37623717d15b881350d677377d) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day.","contextOnly":{"reason":"Fresh-context audit (A2) judged that the quotation misleads when shown alone.","requires":"Scope is the meaning of redistribution under AI (status, meaningful work), not the field as a whole."}},{"id":"lg-ev-oai-zoph-departure-2024-1","sourceRole":"primary","support":"supports","locator":"Post text, note to OpenAI, first sentence","passage":"Zoph writes in a note to OpenAI that he has decided to leave the company.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “to leave OpenAI” matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read through X’s own syndication endpoint for the post’s text, which truncates long notes. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. The paraphrase is Selekoa’s; only the anchor is the source’s wording. No human editorial review.","retrievalRoute":"x_syndication"},{"id":"lg-ev-oai-zoph-departure-2024-2","sourceRole":"secondary","support":"supports","locator":"Opening paragraph","passage":"OpenAI’s chief research officer, Bob McGrew, and a research VP, Barret Zoph, left the company on Wednesday","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (17 words) matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"lg-ev-new-murati-tml-launch-2025","sourceRole":"primary","support":"supports","locator":"X post, first sentence","passage":"Murati wrote that she started Thinking Machines Lab alongside a team of scientists, engineers and builders.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “I started” matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read through X’s own syndication endpoint for the post’s text, which truncates long notes. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. The paraphrase is Selekoa’s; only the anchor is the source’s wording. A cross-family machine check (2026-10-16/17) re-read the record’s sources and agreed with the record. No human editorial review.","retrievalRoute":"x_syndication"},{"id":"lg-ev-new-murati-tml-launch-2025-tc","sourceRole":"secondary","support":"supports","locator":"Paragraph 1","passage":"Former OpenAI CTO Mira Murati has announced her new startup.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (10 words) matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-16/17) re-read the record’s sources and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"lg-ev-oai-zaremba-foundation-resilience-2026","sourceRole":"primary","support":"supports","locator":"Section on AI Resilience","passage":"Wojciech Zaremba, a co-founder of OpenAI, is joining the Foundation as Head of AI Resilience to lead this work","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (19 words) matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read through the third-party reader relay r.jina.ai, which renders the publisher’s current page (the publisher refused direct retrieval); not an archive copy. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"reader_relay"},{"id":"lg-ev-gw-laskin-deepmind-reward-models","sourceRole":"secondary","support":"supports","locator":"Section “From AlphaGo to Reflection AI”, first paragraph","passage":"post-training efforts for Gemini, where he met Reflection cofounder Misha Laskin, who was leading reward model development","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (17 words) matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-16/17) re-read the record’s sources and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"lg-ev-mx-lecun-cofounded-ami-home","sourceRole":"primary","support":"supports","locator":"Section following 'Founded by a globally distinguished team…', list of names","passage":"After stating that AMI was founded by a team of scientists, engineers and builders, the homepage lists seven names including Yann LeCun.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The paraphrase’s verbatim anchor “Founded by a globally distinguished team” matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. The paraphrase is Selekoa’s; only the anchor is the source’s wording. A cross-family machine check (2026-10-16/17) re-read the record’s sources and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"lg-ev-mx-lecun-cofounded-ami-cnbc","sourceRole":"secondary","support":"supports","locator":"Paragraph quoting LeCun's LinkedIn post","passage":"I am creating a startup company to continue the Advanced Machine Intelligence research program (AMI)","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (15 words) matched the retrieved text on 2026-10-15 and again in a fresh retrieval the same day. Read through the third-party reader relay r.jina.ai, which renders the publisher’s current page (the publisher refused direct retrieval); not an archive copy. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-16/17) re-read the record’s sources and agreed with the record. No human editorial review.","retrievalRoute":"reader_relay"},{"id":"ev-cmp-faif-2025-d3-safeguards-1","sourceRole":"primary","support":"supports","locator":"PDF p. 6, 'Addressing Risks of Malicious Use' › '3. Safeguard Implementation', second paragraph lead-in and third bullet","passage":"we continue to design and deploy the following safeguards into our models: … Input and output filters","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-27 against the publisher copy, whose SHA-256 matches the copy retrieved on 2026-10-15. The list also names safety training and system prompts. Human editorial review is pending."},{"id":"ev-cmp-faif-2025-d3-safeguards-2","sourceRole":"primary","support":"supports","locator":"PDF p. 9, 'Operational and Societal Risks' › '3. Information Security', first sentence","passage":"information security standards sufficient to prevent its critical model information from being stolen by a motivated non-state actor.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-27 against the publisher copy, whose SHA-256 matches the copy retrieved on 2026-10-15. Human editorial review is pending."},{"id":"evidence-ap-eliezer-yudkowsky-lw-2007-the-hidden-complexity-of-wishes","sourceRole":"primary","support":"supports","locator":"Post author list in the publisher’s post data (LessWrong GraphQL API)","passage":"Eliezer Yudkowsky","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Credited name checked by an automated research assistant on 2026-10-27 in a direct retrieval through the publisher’s post data. Human editorial review is pending.","retrievalRoute":"direct"},{"id":"evidence-ap-eliezer-yudkowsky-sl4-2007-11-24-yudkowsky-0711-17248","sourceRole":"primary","support":"supports","locator":"Message header, sender name (address not reproduced)","passage":"Eliezer S. Yudkowsky","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Credited name checked by an automated research assistant on 2026-10-27 in a direct retrieval. Human editorial review is pending.","retrievalRoute":"direct"},{"id":"ev-cmp-rsp-2023-d8-process-1","sourceRole":"primary","support":"supports","locator":"PDF p. 10 (printed p. 10), Procedural Commitments, numbered item 1, first sentence","passage":"The edition lists among its procedural commitments an “Update Process” for the document, under which the board of directors approves changes after consulting the Long-Term Benefit Trust.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-27","note":"Explicit paraphrase, written on 2026-10-27 by an automated research assistant after re-reading the passage in the publisher copy, whose SHA-256 matches the copy retrieved on 2026-10-15. The comparison quoted the passage; it is given here as a paraphrase to keep the edition within the quotation limit. The next sentence says any update will be noted in the document before it is implemented. Human editorial review is pending."},{"id":"ev-cmp-rsp-v3-1-d1-autonomy-1","sourceRole":"primary","support":"supports","locator":"PDF p. 8, §1, threshold table, row 'Automated R&D in key domains', left column, first sentence","passage":"The “Automated R&D in key domains” row of the threshold table concerns AI systems capable of fully automating, or dramatically speeding up, the work of large teams of leading human researchers in domains where rapid progress could threaten international security or quickly disrupt the global balance of power.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-28","note":"Explicit paraphrase by an automated research assistant, corrected on 2026-10-28 after a blind check of all 287 comparison paraphrases by a reviewer from a different model family on 2026-10-28 (research/frontier-framework-comparison/codex-paraphrases-2026-10-28.md), which found that it left out the restriction to domains where fast progress could threaten international security or the global balance of power. The correction was written from the passage as re-read in the publisher copy retrieved on 2026-10-28, whose SHA-256 matches the copy retrieved on 2026-10-15. Given as a paraphrase to keep the edition within the quotation limit. The examples given are energy, robotics, weapons development and AI. Human editorial review is pending."},{"id":"ev-cmp-preparedness-v2-d1-autonomy-1","sourceRole":"primary","support":"supports","locator":"PDF p. 2 (printed p. 1), opening page, third bullet of the Tracked Categories list","passage":"The edition lists AI Self-improvement capabilities as a Tracked Category, describing them as capabilities that could also pose new challenges to humans controlling AI systems.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-27","note":"Explicit paraphrase, written on 2026-10-27 by an automated research assistant after re-reading the passage in the publisher copy, whose SHA-256 matches the copy retrieved on 2026-10-15. The comparison quoted the passage; it is given here as a paraphrase to keep the edition within the quotation limit. Human editorial review is pending."},{"id":"c2-ev-governance-tc260-asgf-v1-published-1","sourceRole":"primary","support":"supports","locator":"News item, first paragraph","passage":"9月9日，在2024年国家网络安全宣传周主论坛上，全国网络安全标准化技术委员会 ... 发布《人工智能安全治理框架》1.0版","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). The quotation (48 words, each CJK character counted as one) matched the retrieved text on 2026-10-27 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. Elided quotation: the ellipses mark omitted words. English machine translation, not verified by a person: “On 9 September, at the main forum of the 2024 National Cybersecurity Awareness Week, the National Cybersecurity Standardization Technical Committee ... released version 1.0 of the AI Safety Governance Framework”. No human editorial review.","retrievalRoute":"direct"},{"id":"c2-ev-governance-tc260-asgf-v1-published-2","sourceRole":"primary","support":"supports","locator":"Cover, issuing body","passage":"全国网络安全标准化技术委员会","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). The quotation (14 words, each CJK character counted as one), a part of a longer passage that the pack joined across separate fields of the page, matched the retrieved text on 2026-10-27 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. The pack had joined this field with others of the same page into one quotation; at admission each field became its own quotation (cross-family check, 2026-10-27). English machine translation, not verified by a person: “National Cybersecurity Standardization Technical Committee”. No human editorial review.","retrievalRoute":"direct"},{"id":"c2-ev-governance-tc260-asgf-v1-published-3","sourceRole":"primary","support":"supports","locator":"Cover, date line","passage":"2024年9月","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). The quotation (4 words, each CJK character counted as one), a part of a longer passage that the pack joined across separate fields of the page, matched the retrieved text on 2026-10-27 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. The pack had joined this field with others of the same page into one quotation; at admission each field became its own quotation (cross-family check, 2026-10-27). English machine translation, not verified by a person: “September 2024”. No human editorial review.","retrievalRoute":"direct"},{"id":"c2-ev-moonshot-k2-report-redteam","sourceRole":"primary","support":"supports","locator":"Section 4.3.1, “Experiment Settings”","passage":"We conducted red-teaming evaluations on Kimi K2 compare with other open-source LLMs.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). The quotation (12 words) matched the retrieved text on 2026-10-27 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-27, item K044) re-read the record’s sources and judged it partly supported; its corrections were applied at admission. No human editorial review.","retrievalRoute":"direct"},{"id":"c2-ev-moonshot-k2-report-redteam-2","sourceRole":"primary","support":"supports","locator":"Section 4.3.1","passage":"We choose Promptfoo5 to generate adversarial prompts and analyze the responses.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). The quotation (11 words) matched the retrieved text on 2026-10-27 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-27, item K044) re-read the record’s sources and judged it partly supported; its corrections were applied at admission. No human editorial review.","retrievalRoute":"direct"},{"id":"c2-ev-governance-tc260-asgf-v2-training-data-cbrn-1","sourceRole":"primary","support":"supports","locator":"§4.1.2 (b)","passage":"对训练数据进行严格筛选，过滤虚假、偏见、失效、错误数据，确保不包含核生化导武器等高危领域敏感数据","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). Transcribed as printed. The passage matched the retrieved text on 2026-10-27 and again in a fresh retrieval the same day, read directly from the publisher; in the PDF’s text layer it contains control characters (U+0001) and compatibility forms (Kangxi radicals, ⻛ for 风, ⻅ for 见), which the candidate pack marked with ellipses. The cross-family check found that those ellipses do not reproduce the printed text (K047), so the quotation was re-extracted at admission, without them and with the forms normalised, from a direct retrieval of the PDF whose SHA-256 equals the pack’s. English machine translation, not verified by a person: “strictly screen training data, filtering out false, biased, invalid and erroneous data, and ensure that it contains no sensitive data from high-risk fields such as nuclear, biological, chemical and missile weapons”. A cross-family machine check (2026-10-27, item K047) re-read the record’s sources and judged it partly supported; its corrections were applied at admission. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-fl-cn-genai-scope-non-public-rd-1","sourceRole":"primary","support":"supports","locator":"Article 2 (第二条), third paragraph","passage":"未向境内公众提供生成式人工智能服务的，不适用本办法的规定","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage, frameworks to law (research/frameworks-to-law). The quotation (27 words, each CJK character counted as one) matched the text retrieved for the candidate pack on 2026-10-27 and again at admission on 2026-10-28, in a fresh retrieval whose SHA-256 equals the pack’s. Read directly from www.gov.cn. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. Read in the gov.cn posting; the dataset's source for the Measures is the CAC's own posting of the same text. English machine translation, not verified by a person: “[where they] do not provide generative AI services to the public within China, the provisions of these Measures do not apply”. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-fl-us-eo14409-voluntary-access-framework-1","sourceRole":"primary","support":"supports","locator":"Sec. 3(b)(ii), 91 FR 34566","passage":"provide the Federal Government with access to covered frontier models … for a period of up to 30 days before they plan to release such models to other trusted partners","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-28","note":"Candidate pack passage, frameworks to law (research/frameworks-to-law). The quotation (29 words) matched the text retrieved for the candidate pack on 2026-10-27 and again at admission on 2026-10-28, in a fresh retrieval whose SHA-256 equals the pack’s. Read directly from www.govinfo.gov. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-28, item 70) re-read the record’s sources and found full support. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-dc-microsoft-fy2025-80bn-1","sourceRole":"primary","support":"supports","locator":"section \"World-leading AI technology and infrastructure\", paragraph 3","passage":"In FY 2025, Microsoft is on track to invest approximately $80 billion to build out AI-enabled datacenters to train AI models and deploy AI and cloud-based applications around the world.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (30 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-dc-microsoft-fy2025-80bn-2","sourceRole":"primary","support":"supports","locator":"section \"World-leading AI technology and infrastructure\", paragraph 3","passage":"More than half of this total investment will be in the United States","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (13 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-ch-usg-intel-closing-1","sourceRole":"primary","support":"supports","locator":"Item 8.01","passage":"the Company received the full amount of the accelerated disbursements under the DFA of $5.695 billion","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (16 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 74) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-ch-usg-intel-closing-2","sourceRole":"primary","support":"supports","locator":"Item 8.01","passage":"issued to the DOC 274,583,000 shares of common stock and a warrant to purchase up to 240,516,150 shares of common stock","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (21 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 74) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-cpo-ch-usg-intel-closing-3","sourceRole":"primary","support":"supports","locator":"Item 8.01","passage":"issued into escrow 158,740,000 shares of common stock for the benefit of the DOC","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, compute, power and orbit (research/compute-power-orbit). The quotation (14 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-29/30, item 74) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ab-scr-hhs-2010-published-1","sourceRole":"primary","support":"supports","locator":"SUMMARY, 75 FR 62820","passage":"sets forth recommended baseline standards for the gene and genome synthesis industry and other providers of synthetic dsDNA products regarding the screening of orders","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). The quotation (24 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. The notice is headed 'DEPARTMENT OF HEALTH AND HUMAN SERVICES, Office of the Secretary' and gives its authority as the Public Health Service Act and HSPD-10. A cross-family machine check (2026-10-30, item 25) answered the record’s question blind from fresh retrievals and agreed with the record. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ab-scr-igsc-hsp-v31-record-retention-1","sourceRole":"primary","support":"supports","locator":"7.3 Record Keeping and Retention, requirement 3.1, p. 15","passage":"IGSC members retain records of every gene synthesized and delivered for a minimum of 8 years after shipping","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). The quotation (18 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. The requirement lists the sequence, the vector where applicable, and the recipient's identity and shipping address as the minimum content. A cross-family machine check (2026-10-30, item 74) re-read the record’s sources and found full support. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ab-scr-igsc-hsp-v31-record-retention-2","sourceRole":"primary","support":"supports","locator":"7.3 Record Keeping and Retention, requirement 3.2, p. 15","passage":"IGSC members retain records of every gene sequence screening result for at least 8 years.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). The quotation (15 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval whose SHA-256 equals the pack’s. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-30, item 74) re-read the record’s sources and found full support. No human editorial review.","retrievalRoute":"direct"},{"id":"evidence-ab-ev-anthropic-asl3-sonnet37-uplift-1","sourceRole":"primary","support":"supports","locator":"Footnote 2","passage":"Experiments showed that access to Claude Sonnet 3.7 helped participants do somewhat better on tasks related to CBRN weapon acquisition than those with standard internet access (although all participants' plans still had critical failures).","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). The quotation (34 words) matched the text retrieved for the candidate pack on 2026-10-28 and again at admission on 2026-10-30, in a fresh retrieval (SHA-256 35fa989dfc4f36842ef4f1dda2cfbdda9f9fc9f63524b310fd869ed2f32ab1e8; the bytes differ from the pack’s retrieval). Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-30, item 31) re-read the record’s sources and found full support. No human editorial review.","retrievalRoute":"direct"}],"at":{"entities":[0,1,2,4,6,15,19,21,27,46,54,55,56,57,67,69,79,82,91,94,109,119,124,159,192,194,201,207,224,243,252,253,285,316,333,338,369,377,412,437,488,491,501,503,514,515,517,518,522,537,538,547,549,553,625,636,746,964,1001,1002,1003,1044,1111,1191,1206,1215,1224,1231],"claims":[101,120,141,170,213,272,385,475,568,653,728,738,857,926,1122,1175,1199,1237,1333,1344,1371,1406,1471,1730,1749,1861,1991,2008,2138,2433,2434,2492,2543,2649,2743,2804,2814,2903,3005,3138,3222,3284,3320,3385],"sources":[0,35,40,49,50,51,52,60,72,94,120,173,183,216,244,245,286,321,338,379,404,455,457,471,488,490,499,501,506,684,693,698,726,732,742,765,771,792,830,831,832,836,881,954,969,1044,1050,1083],"evidence":[118,119,139,160,189,232,293,408,506,635,794,911,935,936,937,1071,1148,1362,1426,1462,1463,1546,1664,1676,1714,1752,1837,2205,2206,2245,2246,2399,2571,2595,2596,2774,2775,3080,3081,3139,3195,3324,3451,3452,3453,3571,3572,3586,3743,3896,4155,4156,4405,4406,4407,4501,4559,4560,4650]},"version":"1lg47t7"}