{"entities":[{"id":"openai","name":"OpenAI"},{"id":"deepmind","name":"DeepMind"},{"id":"eliezer-yudkowsky","name":"Eliezer Yudkowsky"},{"id":"fsf-2024","name":"Frontier Safety Framework (2024)"},{"id":"rsp-v2-0","name":"Anthropic Responsible Scaling Policy — v2.0"},{"id":"rsp-v3-2","name":"Anthropic Responsible Scaling Policy — v3.2"},{"id":"fsf-v3-0","name":"Google DeepMind Frontier Safety Framework — v3.0"},{"id":"preparedness-beta-2023","name":"OpenAI Preparedness Framework — Beta"},{"id":"preparedness-v2","name":"OpenAI Preparedness Framework — Version 2"},{"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":"andrej-karpathy","name":"Andrej Karpathy"},{"id":"alibaba-group","name":"Alibaba Group / 阿里巴巴集团"},{"id":"jingren-zhou","name":"Jingren Zhou"},{"id":"zhipu","name":"Zhipu AI / 智谱"},{"id":"zhang-peng","name":"Zhang Peng / 张鹏"},{"id":"openai-group-pbc","name":"OpenAI Group PBC"},{"id":"extropians-1996-12-31-crocker","name":"Re: Goal-based AI"},{"id":"strategic-concealment","name":"Strategic concealment"},{"id":"extropians-1999-08-25-yudkowsky-2783","name":"Re: SI Comparative Advantage [was Re: Free Will]"},{"id":"explosive-growth-estimates","name":"Estimates of explosive growth within a stated horizon"},{"id":"compensating-policy-for-ai","name":"Compensating policy for AI’s distributional effects"},{"id":"nordhaus-2015-economic-singularity","name":"Are We Approaching an Economic Singularity? Information Technology and the Future of Economic Growth"},{"id":"acemoglu-2024-simple-macroeconomics-of-ai","name":"The Simple Macroeconomics of AI"},{"id":"david-velez","name":"David Vélez"},{"id":"denise-dresser","name":"Denise Dresser"},{"id":"illia-polosukhin","name":"Illia Polosukhin"},{"id":"near-protocol","name":"NEAR"}],"claims":[{"id":"karpathy-deepmind-intern-2015","statement":"Andrej Karpathy’s biography places an internship at DeepMind in 2015.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Internship, not a permanent research appointment; no mentorship or influence edge inferred.","evidenceIds":["evidence-karpathy-deepmind-intern-2015"],"attributes":{"role":"Research intern"}},{"id":"cn-zhang-zhipu","statement":"Zhang Peng is identified as Zhipu CEO at an August 2024 ceremony.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Event date differs from publication on August 20; not a current-role claim or proof of when leadership changed.","evidenceIds":["evidence-cn-zhang-zhipu"],"attributes":{"role":"CEO at the reported ceremony"}},{"id":"prov-rsp-v2-0-no-train-deploy-without-safeguards","statement":"RSP v2.0 states that its Responsible Scaling Policy, released in September 2023, is a public commitment not to train or deploy models capable of causing catastrophic harm unless Anthropic has implemented safety and security measures that will keep risks below acceptable levels.","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-rsp-v2-0-no-train-deploy-without-safeguards"],"attributes":{"provision":{"kind":"required_action","key":"rsp-no-train-deploy-without-safeguards"}}},{"id":"prov-fsf-2024-hold-pending-mitigations","statement":"FSF v1.0 states that if a model reaches evaluation thresholds before mitigations at appropriate levels are ready, Google DeepMind would put on hold further deployment or development, or implement additional protocols.","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-fsf-2024-hold-pending-mitigations"],"attributes":{"provision":{"kind":"required_action","key":"fsf-hold-pending-mitigations"}}},{"id":"prov-preparedness-beta-2023-final-decision-authority","statement":"Preparedness Framework (Beta) states that there is an advisory group, the Safety Advisory Group, to provide nuanced input and recommendations, and a final decision-maker, OpenAI Leadership, with the option for the OpenAI Board of Directors to overrule.","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 Beta defines OpenAI Leadership as the CEO or a person designated by them (PDF p. 23).","evidenceIds":["evidence-prov-preparedness-beta-2023-final-decision-authority"],"attributes":{"provision":{"kind":"decision_authority","key":"pf-final-decision-authority","parties":[{"role":"recommends","actor":"Safety Advisory Group (SAG)"},{"role":"decides","actor":"OpenAI Leadership"},{"role":"can_overrule","actor":"OpenAI Board of Directors"}]}}},{"id":"prov-preparedness-v2-high-deployment-restriction","statement":"Preparedness Framework Version 2 commits OpenAI not to deploy a model that reaches a High capability threshold until the associated risks are sufficiently minimized.","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. Whether the framework’s governance arrangements permit OpenAI Leadership to deploy High-capability models without such safeguards is a separate, contested assessment.","evidenceIds":["evidence-prov-preparedness-v2-high-deployment-restriction"],"attributes":{"provision":{"kind":"required_action","key":"pf-high-deployment-restriction"}}},{"id":"prov-faif-2026-systemic-risk-acceptance-criteria","statement":"FAIF (June 2026) states that xAI applies a systemic risk acceptance criteria to each identified risk, incorporating a margin of security, to determine whether each identified systemic risk and the overall systemic risk are acceptable.","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 edition states no numeric criterion.","evidenceIds":["evidence-prov-faif-2026-systemic-risk-acceptance-criteria"],"attributes":{"provision":{"kind":"threshold","key":"faif-systemic-risk-acceptance-criteria"}}},{"id":"ts-b2-crocker-1996-12-31-contests","statement":"Crocker’s reply of 1996-12-31 argues that even a hypothetical cognitive power is constrained by reality and cannot lie to the writer about the laws of physics. Recorded as contesting the proposition “Strategic concealment”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The position is mixed: Crocker’s message 5151, earlier the same day, concedes that \"a malevolent intelligence may well be capable of convincing a not-too-critical scientist\" to release it.","evidenceIds":["evidence-ts-b2-crocker-1996-12-31-contests","evidence-ts-b2-crocker-1996-12-31-contests-2"],"attributes":{"thought":{"roles":["contesting"],"register":"nonfiction","qualifier":"Claims checkable against physical evidence limit deception.","engagement":{"kind":"engages-premise","rationale":"Gives a reason against the premise that a confined system could deceive its keepers while appearing benign: what it says about physics and experiments can be checked against reality and replicated."}}}},{"id":"ec-g09-nordhaus-2015-economic-singularity-a","statement":"Nordhaus, “Are We Approaching an Economic Singularity? Information Technology and the Future of Economic Growth” (2015-09) articulates G09. Extrapolating the two positive tests puts growth above 20% a year at least a century away.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Reclassified from contests after the cross-family check: a horizon of 100 years or more overlaps \"about a century\" and gives no probability. Under the estimate criterion of 2026-10-09 it is Nordhaus's own estimate.","evidenceIds":["evidence-ec-g09-h"],"attributes":{"thought":{"roles":["articulation"],"register":"nonfiction","attested":{"date":"2016-08-09","note":"Capture of the Cowles PDF."},"tier":{"value":1,"basis":"Tier 1 (unchanged). Economist (Yale) in a core economics series: Cowles Foundation Discussion Paper and NBER Working Paper; journal version in American Economic Journal: Macroeconomics."},"estimates":[{"holder":"William D. Nordhaus","value":"the economy \"might plausibly cross the Singularity\" in \"100 years or more\"","horizon":"100 years or more (from 2015)","definition":"economic growth rate crossing 20% per year","conditional":true,"conditions":"Simple extrapolation of the two of seven tests that are positive; no probability and no duration given."}]}}},{"id":"ec-m06b-acemoglu-2024-simple-macroeconomics-of-ai-b","statement":"Acemoglu, “The Simple Macroeconomics of AI” (2024-05) bears on M06b. Acemoglu argues that new-task creation with better wage outcomes may need new institutions and regulation and a reorientation of the AI industry.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Policy aimed at the direction of technology, not compensation.","evidenceIds":["evidence-ec-m06b-27"],"attributes":{"thought":{"roles":["context"],"register":"nonfiction","missing":"Calls for policy to redirect technology rather than redistribution or compensation.","tier":{"value":1,"basis":"NBER working paper (core venue)."}}}},{"id":"lg-oai-dresser-cro-2025","statement":"OpenAI announced on 9 December 2025 that Denise Dresser was joining as Chief Revenue Officer.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"An announced appointment; no start date is given.","evidenceIds":["lg-ev-oai-dresser-cro-2025"],"attributes":{"activityStage":"announced","role":"Chief Revenue Officer"}},{"id":"lg-gc-jingren-zhou-tongyi-lab-2026-03","statement":"Jiemian reported on 5 March 2026 that Alibaba CEO Wu Yongming’s internal e-mail, which approved Junyang Lin’s resignation, stated that Zhou Jingren would continue to lead Tongyi Lab.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The record rests on reporting of an internal e-mail that was not retrieved. The foundation-model support group the e-mail is reported to have set up is left out. Lin’s own departure post is recorded separately. English rendering of the Chinese passage is a machine translation; no human translation review.","evidenceIds":["lg-ev-gc-jingren-zhou-tongyi-lab-2026-03"],"attributes":{"role":"负责人 (head), 通义实验室 (Tongyi Lab)"}},{"id":"lg-oai-david-velez-pbc-board-2026","statement":"OpenAI announced on 21 July 2026 the appointment of David Vélez to the board of the OpenAI Group PBC.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"One of two simultaneous appointments to the boards of the OpenAI Foundation and OpenAI Group PBC; no start date is given.","evidenceIds":["lg-ev-oai-david-velez-pbc-board-2026"],"attributes":{"role":"Board member"}},{"id":"lg-goo-polosukhin-cofounded-near","statement":"NEAR's home page states that NEAR was co-founded by Illia Polosukhin.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A retrospective credit; it does not name other co-founders, a founding date, or NEAR's earlier history as an AI company.","evidenceIds":["lg-ev-goo-polosukhin-cofounded-near"],"attributes":{"role":"co-founder"}},{"id":"cmp-rsp-v3-2-d3-safeguards","statement":"RSP · v3.2 states a provision on the comparison question “Safeguards required”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Company plans (middle column) are separated from industry-wide recommendations (right column), which the edition says it cannot unilaterally and unconditionally commit to. For the novel chemical/biological row the company column extends protections at least as strong as ASL-3 to more use cases; for the high-stakes sabotage opportunities and Automated R&D rows it lists Risk Report disclosure, security and alignment projects rather than a named standard. 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-rsp-v3-2-d3-safeguards-1"],"attributes":{"provision":{"kind":"required_action","key":"comparison-d3-safeguards","facet":{"id":"D3.safeguards","status":"stated"}}}},{"id":"cmp-fsf-v3-0-d3-halt","statement":"FSF · v3.0 states a provision on the comparison question “Pause, non-deployment or non-training”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A deployment gate: external deployment, and for ML R&D CCLs also large scale internal deployment (PDF p. 12), takes place only after governance approval of a safety case. The misuse deployment-mitigation criterion applies 'only for external deployment, not further development'. No passage states that development will be put on hold (the v2.0 'on hold' sentence is absent). 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-fsf-v3-0-d3-halt-1","ev-cmp-fsf-v3-0-d3-halt-2"],"attributes":{"provision":{"kind":"required_action","key":"comparison-d3-halt","facet":{"id":"D3.halt","status":"stated"}}}},{"id":"ap-eliezer-yudkowsky-extropians-1999-08-25-yudkowsky-2783","statement":"The mailing-list message “Re: SI Comparative Advantage [was Re: Free Will]” (1999-08-25) 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-extropians-1999-08-25-yudkowsky-2783"]},{"id":"cmp-fsf-v3-0-d5-cadence","statement":"FSF · v3.0 states a provision on the comparison question “Evaluation timing”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The next sentence (PDF pp. 4–5) adds a further risk assessment for later versions where a model gains meaningful new capabilities or performs materially better. Capability evaluations that indicate whether a risk assessment is needed may be triggered, for example, when a run of pre-training or of post-training is completed (PDF p. 5). No compute or calendar interval is stated. 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-fsf-v3-0-d5-cadence-1"],"attributes":{"provision":{"kind":"evaluation_timing","key":"comparison-d5-cadence","facet":{"id":"D5.cadence","status":"stated"}}}},{"id":"cmp-preparedness-beta-2023-d2-definition","statement":"Preparedness · Beta states a provision on the comparison question “How a threshold is defined”.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"The edition gives no single definition, across categories, of reaching a level. Each level is defined per category in tables with 'Definition' and 'Rationale for threshold' columns (PDF pp. 8–11); the edition calls its current estimates for the levels from 'medium' to 'critical' speculative (PDF p. 7). 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-beta-2023-d2-definition-1"],"attributes":{"provision":{"kind":"threshold","key":"comparison-d2-definition","facet":{"id":"D2.definition","status":"stated"}}}},{"id":"cmp-faif-2026-d1-other","statement":"FAIF · June 2026 edition states a provision on the comparison question “Other named domains”, in a weaker form than the question’s criterion.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"Stated, partly: the passage has what the question’s criterion asks for, but in a weaker form than the criterion’s own words. No domain beyond the four is named as tracked; safeguard circumvention is named as a focus alongside malicious use, and other risks only as a possibility. 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-faif-2026-d1-other-1","ev-cmp-faif-2026-d1-other-2"],"attributes":{"provision":{"kind":"risk_domain","key":"comparison-d1-other","facet":{"id":"D1.other","status":"partly"}}}},{"id":"ab-fw-faif-2025-vetted-user-exception","statement":"The Frontier Artificial Intelligence Framework of December 2025 states that xAI may selectively allow its models to respond to requests that heightened safeguards would otherwise restrict, where the requester is among some vetted, highly trusted users known to be using those capabilities for benign or beneficial purposes.","derivation":"explicit","resolution":"supported","editorialState":"seed","qualification":"A stated exception. It records no grant of access to anyone, names no customer or auditor, and is not evidence that the exception was used.","evidenceIds":["evidence-ab-fw-faif-2025-vetted-user-exception-1"],"attributes":{"provision":{"kind":"exception","key":"faif-vetted-user-exception"}}}],"sources":[{"id":"source-rsp-2-0","title":"Responsible Scaling Policy — v2.0","authors":["Anthropic"],"publisher":"Anthropic","publicationDate":null,"version":"Responsible Scaling Policy v2.0; effective date 2024-10-15. v2.0 identity is supplied by the official archive and later dated changelogs; the cover is not numbered.","versionDate":"2024-10-15","canonicalUrl":"https://www-cdn.anthropic.com/616dee633636e5bd309cb73aed8622e80fe47839.pdf","evidenceUrl":"https://www-cdn.anthropic.com/616dee633636e5bd309cb73aed8622e80fe47839.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20241022154907/https://assets.anthropic.com/m/24a47b00f10301cd/original/Anthropic-Responsible-Scaling-Policy-2024-10-15.pdf","archivedAt":"2024-10-22","accessedAt":"2026-09-18","reuse":{"attribution":"Anthropic, Responsible Scaling Policy — v2.0.","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-3-2","title":"Responsible Scaling Policy — v3.2","authors":["Anthropic"],"publisher":"Anthropic","publicationDate":null,"version":"Responsible Scaling Policy v3.2; effective date 2026-04-29.","versionDate":"2026-04-29","canonicalUrl":"https://cdn.sanity.io/files/4zrzovbb/website/28c6241900d90410628a8a2003a5572faae4365a.pdf","evidenceUrl":"https://cdn.sanity.io/files/4zrzovbb/website/28c6241900d90410628a8a2003a5572faae4365a.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20260505141510/https://cdn.sanity.io/files/4zrzovbb/website/28c6241900d90410628a8a2003a5572faae4365a.pdf","archivedAt":"2026-05-05","accessedAt":"2026-09-18","reuse":{"attribution":"Anthropic, Responsible Scaling Policy — v3.2.","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-fsf-1-0","title":"Frontier Safety Framework — v1.0","authors":["Google DeepMind"],"publisher":"Google DeepMind","publicationDate":"2024-05-17","version":"Frontier Safety Framework v1.0; publication date 2024-05-17.","versionDate":"2024-05-17","canonicalUrl":"https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/introducing-the-frontier-safety-framework/fsf-technical-report.pdf","evidenceUrl":"https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/introducing-the-frontier-safety-framework/fsf-technical-report.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20240517140902/https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/introducing-the-frontier-safety-framework/fsf-technical-report.pdf","archivedAt":"2024-05-17","accessedAt":"2026-09-18","reuse":{"attribution":"Google DeepMind, Frontier Safety Framework — 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-fsf-3-0","title":"Frontier Safety Framework — v3.0","authors":["Google DeepMind"],"publisher":"Google DeepMind","publicationDate":"2025-09-22","version":"Frontier Safety Framework v3.0; publication date 2025-09-22.","versionDate":"2025-09-22","canonicalUrl":"https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/strengthening-our-frontier-safety-framework/frontier-safety-framework_3.pdf","evidenceUrl":"https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/strengthening-our-frontier-safety-framework/frontier-safety-framework_3.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20250922134959/https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/strengthening-our-frontier-safety-framework/frontier-safety-framework_3.pdf","archivedAt":"2025-09-22","accessedAt":"2026-09-18","reuse":{"attribution":"Google DeepMind, Frontier Safety Framework — v3.0.","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-pf-beta","title":"Preparedness Framework — Beta","authors":["OpenAI"],"publisher":"OpenAI","publicationDate":"2023-12-18","version":"Preparedness Framework Beta; publication date 2023-12-18.","versionDate":"2023-12-18","canonicalUrl":"https://cdn.openai.com/openai-preparedness-framework-beta.pdf","evidenceUrl":"https://cdn.openai.com/openai-preparedness-framework-beta.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20231218195930/https://cdn.openai.com/openai-preparedness-framework-beta.pdf","archivedAt":"2023-12-18","accessedAt":"2026-09-18","reuse":{"attribution":"OpenAI, Preparedness Framework — Beta.","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-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-karpathy-bio","title":"Andrej Karpathy — career timeline","authors":["Andrej Karpathy"],"publisher":"Andrej Karpathy","publicationDate":null,"version":"Page inspected 2026-09-19; historical statements are scoped to their source, not treated as current appointments.","canonicalUrl":"https://karpathy.ai/","evidenceUrl":"https://karpathy.ai/","language":"en","archivedUrl":"https://web.archive.org/web/20260919000305/https://karpathy.ai/","archivedAt":"2026-09-19","accessedAt":"2026-09-19","reuse":{"attribution":"Andrej Karpathy, Andrej Karpathy — career timeline","licence":"No open redistribution licence asserted","note":"Bibliographic metadata and brief attributed paraphrases only; no full source or personal contact details redistributed."}},{"id":"source-cn-zhipu-education","title":"智谱支持“清华大学人工智能教育赋能项目”捐赠仪式举行","authors":["清华大学教育基金会"],"publisher":"Tsinghua University","publicationDate":"2024-08-20","version":"Source language: Simplified Chinese (zh-Hans). Dated account of the 2024-08-15 ceremony.","canonicalUrl":"https://www.tsinghua.edu.cn/info/1178/113278.htm","evidenceUrl":"https://www.tsinghua.edu.cn/info/1178/113278.htm","language":"zh-Hans","archivedUrl":"https://web.archive.org/web/20260314025338/https://www.tsinghua.edu.cn/info/1178/113278.htm","archivedAt":"2026-03-14","accessedAt":"2026-09-29","reuse":{"attribution":"Tsinghua University, 智谱支持“清华大学人工智能教育赋能项目”捐赠仪式举行 (2024-08-20).","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-extropians-1996-12-31-crocker","title":"Re: Goal-based AI (Extropians mailing list)","authors":["Lee Daniel Crocker"],"publisher":"Extropians mailing list","publicationDate":"1996-12-31","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.96/5167.html","evidenceUrl":"http://extropians.weidai.com/extropians.96/5167.html","language":"en","archivedUrl":"https://web.archive.org/web/20150427134518/http://extropians.weidai.com/extropians.96/5167.html","archivedAt":"2015-04-27","accessedAt":"2026-10-06","reuse":{"attribution":"Lee Daniel Crocker, \"Re: Goal-based AI\", Extropians mailing list, 1996-12-31.","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-extropians-1996-12-31-crocker-5151","title":"Re: Goal-based AI (Extropians mailing list, message 5151)","authors":["Lee Daniel Crocker"],"publisher":"Extropians mailing list","publicationDate":"1996-12-31","version":"Archive HTML as served, retrieved 2026-10-06; sent header Tue, 31 Dec 1996 11:55:49 -0800 (PST). Archive at extropians.weidai.com.","canonicalUrl":"http://extropians.weidai.com/extropians.96/5151.html","evidenceUrl":"http://extropians.weidai.com/extropians.96/5151.html","language":"en","archivedUrl":"https://web.archive.org/web/20211017061411/https://extropians.weidai.com/extropians.96/5151.html","archivedAt":"2021-10-17","accessedAt":"2026-10-06","reuse":{"attribution":"Lee Daniel Crocker, \"Re: Goal-based AI\", Extropians mailing list, 1996-12-31.","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-extropians-1999-08-25-yudkowsky-2783","title":"Re: SI Comparative Advantage [was Re: Free Will] (Extropians mailing list)","authors":["Eliezer Yudkowsky"],"publisher":"Extropians mailing list","publicationDate":"1999-08-25","version":"Archive HTML as served, retrieved 2026-10-07 (SHA-256 7618c597242da2ca916693d2a9f504a9f6856c06202a85912a664edca705a6ce); quoted lines excluded when reading the author’s text. The served archive shows later URL rewriting and does not authenticate historical bytes. Archive at extropians.weidai.com.","canonicalUrl":"http://extropians.weidai.com/extropians.3Q99/2783.html","evidenceUrl":"http://extropians.weidai.com/extropians.3Q99/2783.html","language":"en","archivedUrl":"https://web.archive.org/web/20220810200948/https://extropians.weidai.com/extropians.3Q99/2783.html","archivedAt":"2022-08-10","accessedAt":"2026-10-07","reuse":{"attribution":"Eliezer Yudkowsky, “Re: SI Comparative Advantage [was Re: Free Will]”, Extropians mailing list, 1999-08-25.","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":"source-nordhaus-2015-cowles","title":"Are We Approaching an Economic Singularity? Information Technology and the Future of Economic Growth","authors":["William D. Nordhaus"],"publisher":"Cowles Foundation Discussion Paper No. 2021, Yale University","publicationDate":"2015-09","version":"Wayback capture of the Cowles Foundation PDF (cowles.yale.edu/.../d2021.pdf); compared with a capture of NBER Working Paper 21547. Capture retrieved 2026-10-08 (SHA-256 d9348fe608486784675d6f95dfb78367801c0ca33217b500a0d31bee1dea5f97). Versions: Cowles Foundation Discussion Paper No. 2021 (2015-09): checked; NBER Working Paper 21547 (2015-09): compared; American Economic Journal: Macroeconomics 13(1) (2021-01): not compared. Published version: Nordhaus, W. D. (2021). Are we approaching an economic singularity? Information technology and the future of economic growth. American Economic Journal: Macroeconomics, 13(1), 299–332. https://doi.org/10.1257/mac.20170105 Working paper: Nordhaus, W. D. (2015). Cowles Foundation Discussion Paper No. 2021; NBER Working Paper No. 21547. https://doi.org/10.3386/w21547","canonicalUrl":"http://cowles.yale.edu/sites/default/files/files/pub/d20/d2021.pdf","evidenceUrl":"https://web.archive.org/web/20160809230752/http://cowles.yale.edu/sites/default/files/files/pub/d20/d2021.pdf","language":"en","archivedUrl":"https://web.archive.org/web/20160809230752/http://cowles.yale.edu/sites/default/files/files/pub/d20/d2021.pdf","archivedAt":"2016-08-09","accessedAt":"2026-10-08","reuse":{"attribution":"William D. Nordhaus, “Are We Approaching an Economic Singularity? Information Technology and the Future of Economic Growth”, Cowles Foundation Discussion Paper No. 2021, Yale University, 2015-09.","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-acemoglu-2024-nber","title":"The Simple Macroeconomics of AI","authors":["Daron Acemoglu"],"publisher":"NBER Working Paper 32487","publicationDate":"2024-05","version":"Full text. Current copy retrieved directly on 2026-10-09; no archive capture (Wayback unreachable), so earlier bytes are not authenticated. SHA-256 ffb04f119539228da2db0699530425e56e815c890b6ebe7b5c932675e2eebdb0. Journal version not retrieved. Acemoglu, D. (2025). The simple macroeconomics of AI. Economic Policy, 40(121), 13–58. https://doi.org/10.1093/epolic/eiae042","canonicalUrl":"https://www.nber.org/system/files/working_papers/w32487/w32487.pdf","evidenceUrl":"https://www.nber.org/system/files/working_papers/w32487/w32487.pdf","language":"en","accessedAt":"2026-10-09","reuse":{"attribution":"Daron Acemoglu, “The Simple Macroeconomics of AI”, NBER Working Paper 32487, 2024-05.","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-gc-jiemian-alibaba-2026","title":"阿里批准林俊旸离职后，谷歌DeepMind高管“隔空喊话”千问团队加盟","authors":["界面新闻"],"publisher":"界面新闻 (Jiemian)","publicationDate":"2026-03-05","version":"Dated page, retrieved on 2026-10-15 (SHA-256 2b8e1f7d1a994966a49bb115aff1c7c93aac82c9b5e9077813fbca6dfe67decf) directly from the publisher.","canonicalUrl":"https://m.jiemian.com/article/14074415.html","evidenceUrl":"https://m.jiemian.com/article/14074415.html","language":"zh","archivedUrl":"https://web.archive.org/web/20260616003134/https://m.jiemian.com/article/14074415.html","archivedAt":"2026-06-16","accessedAt":"2026-10-15","reuse":{"attribution":"界面新闻, “阿里批准林俊旸离职后，谷歌DeepMind高管“隔空喊话”千问团队加盟”, 2026-03-05.","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-goo-near-home","title":"NEAR (home page, undated)","authors":["NEAR"],"publisher":"NEAR","publicationDate":null,"version":"Undated page, retrieved on 2026-10-15 (SHA-256 e0baa00c7936ba90739c89fa0bf39a9b49050e9ae647b664ce066d593b83955f) directly from the publisher. Its wording can change; earlier wording is not authenticated.","canonicalUrl":"https://near.org/","evidenceUrl":"https://near.org/","language":"en","archivedUrl":"https://web.archive.org/web/20261014035712/https://near.org/","archivedAt":"2026-10-14","accessedAt":"2026-10-15","reuse":{"attribution":"NEAR, “NEAR (home page, undated)”.","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-dresser-2025","title":"OpenAI appoints Denise Dresser as Chief Revenue Officer","authors":["OpenAI"],"publisher":"OpenAI","publicationDate":"2025-12-09","version":"Dated page, retrieved on 2026-10-15 (SHA-256 b425e19cd841549f27240c30225acc813767f71e6a787008762dec6cc37d725b) 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/openai-appoints-denise-dresser/","evidenceUrl":"https://openai.com/index/openai-appoints-denise-dresser/","language":"en","archivedUrl":"https://web.archive.org/web/20260915081921/https://openai.com/index/openai-appoints-denise-dresser/","archivedAt":"2026-09-15","accessedAt":"2026-10-15","reuse":{"attribution":"OpenAI, “OpenAI appoints Denise Dresser as Chief Revenue Officer”, 2025-12-09.","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":"lg-src-oai-velez-vince-2026","title":"David Vélez and Robin Vince join the boards of the OpenAI Foundation and OpenAI Group PBC","authors":["OpenAI"],"publisher":"OpenAI","publicationDate":"2026-07-21","version":"Dated page, retrieved on 2026-10-15 (SHA-256 af047510bc6425be32311ee5ad376230d6f961caaf88b81e7020e5c1a375e82f) 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/david-velez-robin-vince-join-openai-boards/","evidenceUrl":"https://openai.com/index/david-velez-robin-vince-join-openai-boards/","language":"en","archivedUrl":"https://web.archive.org/web/20260731093044/https://openai.com/index/david-velez-robin-vince-join-openai-boards/","archivedAt":"2026-07-31","accessedAt":"2026-10-15","reuse":{"attribution":"OpenAI, “David Vélez and Robin Vince join the boards of the OpenAI Foundation and OpenAI Group PBC”, 2026-07-21.","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-karpathy-deepmind-intern-2015","sourceRole":"primary","support":"supports","locator":"2011–2015 section, internship paragraph","passage":"Karpathy lists an internship at DeepMind in 2015.","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-cn-zhang-zhipu","sourceRole":"primary","support":"supports","locator":"Opening paragraph, 首席执行官（CEO）张鹏","passage":"Zhang Peng attended the August 15 ceremony as Zhipu CEO.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-09-29","note":"English paraphrase translated by machine from the cited original Chinese passage; no human translation review. Exploratory seed, not editorially accepted. Translation scope and ambiguities are retained in the claim qualification."},{"id":"evidence-prov-rsp-v2-0-no-train-deploy-without-safeguards","sourceRole":"primary","support":"supports","locator":"PDF p. 2 (unnumbered), Executive Summary, first paragraph, first sentence","passage":"In September 2023, we released our Responsible Scaling Policy (RSP), a public commitment not to train or deploy models capable of causing catastrophic harm unless we have implemented safety and security measures that will keep risks below acceptable levels.","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 2024-10-22 (earlier assets.anthropic.com address). Extraction note: Repeated in the Introduction (PDF p. 5, printed p. 1). Operative forms: §4.3 final sentence (restrict model deployment and further scaling if the required showing cannot be made, PDF p. 14) and §6.2 'Monitoring pretraining' (pause training until the ASL-3 Security Standard is implemented, PDF p. 15). The changelog (PDF p. 21) says the 2024 update maintains this commitment. The cover carries no version number. Human editorial review is pending."},{"id":"evidence-prov-fsf-2024-hold-pending-mitigations","sourceRole":"primary","support":"supports","locator":"PDF p. 2 (printed p. 2), Framework, component “3 - Applying mitigations”, third paragraph, first sentence and opening of second sentence","passage":"A model may reach evaluation thresholds before mitigations at appropriate levels are ready. If this happens, we would put on hold further deployment or development, or implement additional protocols","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 2024-05-17. Extraction note: Quote ends mid-sentence; the sentence continues with an example (more precise early warning evaluations for a given CCL) and the purpose that models not reach CCLs without appropriate security mitigations or be deployed with CCLs without appropriate deployment mitigations. Human editorial review is pending."},{"id":"evidence-prov-preparedness-beta-2023-final-decision-authority","sourceRole":"primary","support":"supports","locator":"PDF p. 22 (printed p. 22), Governance › “Operations”, opening paragraph, second sentence, items (2) and (3)","passage":"(2) there is an advisory group (Safety Advisory Group) that has a sufficient diversity of perspectives and technical expertise to provide nuanced input and recommendations, and (3) there is a final decision-maker (OpenAI Leadership, with the option for the OpenAI Board of Directors to overrule).","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 2023-12-18. Extraction note: Starts mid-sentence at item (2); the sentence opens “These commitments aim to make sure that:” and item (1) names the Preparedness team. PDF p. 23 defines OpenAI Leadership as the CEO or a person designated by them. Human editorial review is pending."},{"id":"evidence-prov-preparedness-v2-high-deployment-restriction","sourceRole":"primary","support":"supports","locator":"PDF p. 4 (printed p. 3), §1 Introduction, third bullet (“Safeguard against severe harms”), second sentence","passage":"We do not deploy models that reach a High capability threshold until the associated risks that they pose are sufficiently minimized.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-03","note":"Verbatim excerpt checked by an automated research assistant on 2026-10-03 against the Internet Archive capture of 2025-04-15 (the edition as first captured) and the current file; the passage is identical in both. Extraction note: Identical in the 2025-04-15 Internet Archive capture and the current file. §4.2 (PDF p. 12, printed p. 11) restates the rule for covered systems that reach High capability. Human editorial review is pending."},{"id":"evidence-prov-faif-2026-systemic-risk-acceptance-criteria","sourceRole":"primary","support":"supports","locator":"PDF p. 5 (no printed page numbers), §2.3 Systemic risk acceptance determination (the first of two sections numbered 2.3), first sentence, up to “acceptable”","passage":"xAI applies a systemic risk acceptance criteria to each identified risk, incorporating a margin of security, to determine whether each identified systemic risk and the overall systemic risk are acceptable","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 2026-07-16. Extraction note: Ends mid-sentence; the sentence continues on PDF p. 6 that the determination governs whether to proceed with development, making available on the market and/or use of the model. The edition states no numeric criterion. Human editorial review is pending."},{"id":"evidence-ts-b2-crocker-1996-12-31-contests","sourceRole":"primary","support":"supports","locator":"Message body, first unquoted paragraph (reply to the quoted Yudkowsky passage)","passage":"Even a hypothetical cognitive power is constrained by reality--he can't lie to me about the laws of physics","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-b2-crocker-1996-12-31-contests-2","sourceRole":"primary","support":"context","locator":"Message 5151, reply to the second point on subliminal suggestion","passage":"Crocker concedes that a malevolent intelligence might convince a not-too-critical scientist to release it, while saying that Crocker would still use restraints.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-06","note":"Retrieved and read for this collection on 2026-10-06 (archive receipt Tue Dec 31 13:14:54 1996 MST)."},{"id":"evidence-ec-g09-h","sourceRole":"primary","support":"supports","locator":"Section IX \"Summary of Tests for Singularity\"","passage":"Using simple extrapolation for the two positive tests, the time at which the economy might plausibly cross the Singularity is 100 years or more.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage G09-h (research/economics-growth). The quotation (24 words) matched the checked expression (nordhaus-cowles) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day."},{"id":"evidence-ec-m06b-27","sourceRole":"primary","support":"supports","locator":"pdf page 47; 4 Concluding Remarks and Discussion (printed p. 45)","passage":"more beneficial outcomes may require new institutions, policies and regulations","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-14","note":"Candidate pack passage M06b-27 (research/economics-mainstream). The quotation (10 words) matched the retrieved current copy (SHA-256 ffb04f119539228da2db0699530425e56e815c890b6ebe7b5c932675e2eebdb0) in the pack's verification. Re-checked at admission on 2026-10-14 against a copy retrieved that day.","contextOnly":{"reason":"Fresh-context audit (A3) judged that the quotation misleads when shown alone.","requires":"Third of a list of reasons new beneficial tasks are excluded from the estimate; 'may require'."}},{"id":"lg-ev-oai-dresser-cro-2025","sourceRole":"primary","support":"supports","locator":"Opening sentence","passage":"Denise Dresser is joining as Chief Revenue Officer","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (8 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-gc-jingren-zhou-tongyi-lab-2026-03","sourceRole":"secondary","support":"supports","locator":"Paragraph beginning “针对外界关切”","passage":"通义实验室将由负责人周靖人继续带队","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (1 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-david-velez-pbc-board-2026","sourceRole":"primary","support":"supports","locator":"Opening sentence","passage":"the appointment of David Vélez and Robin Vince to the boards of the OpenAI Foundation and OpenAI Group PBC","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-goo-polosukhin-cofounded-near","sourceRole":"primary","support":"supports","locator":"What is NEAR?","passage":"Co-founded by Illia Polosukhin","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-15","note":"Candidate pack passage, lab genealogy (research/lab-genealogy). The quotation (4 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":"ev-cmp-rsp-v3-2-d3-safeguards-1","sourceRole":"primary","support":"supports","locator":"PDF p. 6, §1, threshold table, row 'Non-novel chemical/biological weapons production', middle column ('Mitigations—our plan as a company')","passage":"We will maintain or improve on our ASL-3 protections, which include classifier guards at least as robust as our initial Constitutional Classifiers","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 guards are described as at least as robust as the initial Constitutional Classifiers; the list continues with access controls, red-teaming, bug bounties, threat intelligence and security controls. Human editorial review is pending."},{"id":"ev-cmp-fsf-v3-0-d3-halt-1","sourceRole":"primary","support":"supports","locator":"PDF p. 9 (printed p. 9), §2.1.2 Deployment Mitigations, numbered step 2, first sentence","passage":"Pre-deployment review of safety case: external deployments of a model take place only after the appropriate governance function determines the safety case regarding each CCL the model has reached to be adequate.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Existing dataset passage for this edition, which contains the comparison’s shorter quotation, first checked on 2026-10-03. The comparison’s quotation was 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":"ev-cmp-fsf-v3-0-d3-halt-2","sourceRole":"primary","support":"supports","locator":"PDF p. 7 (printed p. 7), §1.6 Summary of Risk Acceptance Criteria, misuse CCL bullet, first sub-bullet, final sentence","passage":"This is required only for external deployment, not further development.","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-27","note":"Passage also quoted in the dataset’s existing evidence for this edition, first checked on 2026-10-03. The comparison’s quotation was 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. Refers to the deployment-mitigation acceptance criterion for misuse CCLs. Human editorial review is pending."},{"id":"evidence-ap-eliezer-yudkowsky-extropians-1999-08-25-yudkowsky-2783","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-fsf-v3-0-d5-cadence-1","sourceRole":"primary","support":"supports","locator":"PDF p. 4 (printed p. 4), §1.3 Outline of Our Risk Assessment Process, first paragraph, second sentence","passage":"The edition says Google DeepMind conducts a risk assessment for a new frontier AI model's first external deployment.","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":"ev-cmp-preparedness-beta-2023-d2-definition-1","sourceRole":"primary","support":"supports","locator":"PDF p. 2, Introduction, footnote 1","passage":"The edition defines catastrophic risk, its focus, as any risk that could cause economic damage of hundreds of billions of dollars or severe harm to, or the death of, many people.","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":"ev-cmp-faif-2026-d1-other-1","sourceRole":"primary","support":"supports","locator":"PDF p. 1, §1 Introduction, third paragraph, first sentence","passage":"The edition says xAI focuses in particular on malicious-use risk and on the risk of safeguards being circumvented, giving jailbreaks as an example.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Explicit paraphrase, written on 2026-10-27 by an automated research assistant from the comparison’s quotation of this passage, which was checked on 2026-10-15 against the publisher copy (SHA-256 recorded in the comparison pack). The edition could not be retrieved again on 2026-10-27; the archive capture of 2026-07-16 has the same SHA-256 (research/archive-checks-2026-10-26/codex-report.md). Given as a paraphrase to keep the edition within the quotation limit. Human editorial review is pending."},{"id":"ev-cmp-faif-2026-d1-other-2","sourceRole":"primary","support":"supports","locator":"PDF p. 1, §1 Introduction, third paragraph, final sentence","passage":"The edition says that, within its risk management framework, xAI may also identify and assess further risks.","passageType":"paraphrase","verification":"machine_checked","checkedAt":"2026-10-15","note":"Explicit paraphrase, written on 2026-10-27 by an automated research assistant from the comparison’s quotation of this passage, which was checked on 2026-10-15 against the publisher copy (SHA-256 recorded in the comparison pack). The edition could not be retrieved again on 2026-10-27; the archive capture of 2026-07-16 has the same SHA-256 (research/archive-checks-2026-10-26/codex-report.md). Given as a paraphrase to keep the edition within the quotation limit. Human editorial review is pending."},{"id":"evidence-ab-fw-faif-2025-vetted-user-exception-1","sourceRole":"primary","support":"supports","locator":"PDF p. 3 (no printed page numbers), “Addressing Risks of Malicious Use”, third paragraph, first sentence","passage":"we may selectively allow xAI’s models to respond to such requests from some vetted, highly trusted users (such as trusted third-party safety auditors or large enterprise customers under contract) whom we know to be using those capabilities for benign or beneficial purposes","passageType":"quote","verification":"machine_checked","checkedAt":"2026-10-30","note":"Candidate pack passage, AI and biosecurity (research/ai-bio). The quotation (42 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"}],"at":{"entities":[0,9,19,22,42,47,51,53,54,56,57,71,179,182,187,191,212,297,343,404,512,519,526,534,658,659,686,750],"claims":[156,381,404,432,450,451,463,888,1291,1439,1816,1854,1886,1970,2078,2109,2374,2619,2642,2685,3276],"sources":[36,41,44,46,48,49,51,52,76,172,299,300,371,477,502,616,640,741,768],"evidence":[175,404,436,464,482,483,494,1103,1104,1618,1789,2335,2390,2435,2546,2685,2733,2734,3021,3287,3316,3371,3372,4492]},"version":"1lg47t7"}