human.layer.re · historical view at 2026-10-09
Human verification capacity is becoming a measurable bottleneck in AI-accelerated software security. Anthropic's open-source pipeline generated 29,439 candidate findings while only 6,123 had reached external human triage by October 2; 5,674 of those were confirmed valid. Independent maintainer release records also show real fixes from the program. The evidence is strong for a verification-throughput constraint, but it is concentrated in one vendor-led cybersecurity program and does not establish the same bottleneck across all AI-assisted knowledge work.
Original public milestone: Scale Labs publishes clinician-authored DistressBench crisis support benchmark. Evidence-only proposal; no claim of global canonical novelty, independent replication or realized downstream effects.
World ↗evt-hist-phase33-scale-distressbench-publicationThe fan team published a v0.1-beta BPS localization patch built for Unbound 1.0.1, with approximately 17,000 translated text entries and a developer-measured 98% visible German text. Automated emulator tests and AI-assisted translation are documented; this is a beta and completeness and language quality are not independently certified.
evt-rts-20261010-pokemon-unbound-de-betaAnthropic launched OSS Scanner, an opt-in recurring service that sends raw model-generated vulnerability reports, reproducers and candidate patches to eligible open-source maintainers. The release follows a pipeline where candidate generation materially exceeded human review capacity.
evt-anthropic-oss-scanner-launch-20261008Google Oct 8 describes a Lancet clinical study in which 98 patients interacted with AMIE under live physician supervision before urgent primary-care consultations. Researchers report no interruptions under predefined criteria; conclusions do not establish independently validated diagnostic safety at population scale.
evt-hist-phase32-amie-bidmc-lancet-studyAP-NORC October 8 survey reports 64 percent of US adults believe AI is developing too fast; about eight in ten prioritize maintaining human control and worker protection. These are sample-based opinions, not actual future regulatory outcomes.
evt-hist-phase32-ap-norc-public-ai-pollBiohub Oct 7 describes $1.8bn combined funding/data/computation commitments involving government and industry to develop AI-ready biology data. Funding commitments are not completed virtual cell simulations, delivered patient therapies or entirely new cash contributions.
evt-hist-phase31-biohub-virtual-biology-fundingOct 7 risk assessment of testing over 4000 prompts. The original PostCutoff source URL pointed instead to Perplexity; this is the corrected original CSM ChatGPT source, and risk rating is the institute assessment, not a judicial finding.
evt-hist-phase31-csm-chatgpt-teens-riskOn 7 October 2026 Indiana utility regulators approved AES Indiana's large-load infrastructure plan supporting a proposed Google 390-MW data center in Monrovia. AES estimates approximately $1.3 billion in energy infrastructure spending, including solar, storage and transmission. Actual construction and stated $770 million customer savings remain prospective; assignment of $74 million in transmission upgrade costs is disputed in public reporting.
evt-rts-20261007-indiana-google-grid-approvalOpenAI released a large model-generated mathematics manuscript collection on October 6, then recorded three withdrawals from a shared sign-error dependency and 14 manuscript revisions on October 7. These are primary-repository audit facts; they do not establish that all published proofs were independently verified or that self-improving AI emerged.
evt-rts-20261010-openai-math-manuscript-correctionsMicrosoft described user-permissioned local context and local actions for Copilot, while Microsoft/NVIDIA material presented MXC as generally available OS-level secure execution infrastructure for agents.
evt-windows-mxc-agent-execution-20261007Public documentation milestone: Brookings publishes Danish worker-study on generative AI work effects. Consult source for claims, date differences and limitations; Phase30 source receipt #1084. No global novelty or downstream outcomes implied.
evt-hist-phase30-brookings-ai-work-labor-studyPublic documentation milestone: Google publishes Earth AI geospatial public-health prototype results. Consult source for claims, date differences and limitations; Phase30 source receipt #1079. No global novelty or downstream outcomes implied.
evt-hist-phase30-google-earth-health-geospatial-studyPublic documentation milestone: UK Government accepts all 44 AI healthcare regulation commission recommendations. Consult source for claims, date differences and limitations; Phase30 source receipt #1106. No global novelty or downstream outcomes implied.
evt-hist-phase30-uk-healthcare-ai-44-responseOpenAI and Ironclad disclosed a setup in which hosted contracting software, domain experts and business-rule rubrics feed into computer-use agent training and evaluation. The disclosed evaluation is internal and narrow.
evt-openai-ironclad-domain-training-20261006October 5, 2026 Nature Health published a randomized simulated respiratory-case experiment in Chinese healthcare systems: LungDiag-assigned participants scored 70.0 percent versus 55.4 percent with conventional mobile web search on tested item accuracy. These simulated outcomes do not establish real-world clinical benefit or safe standalone diagnosis.
evt-hist-2026phase28-lungdiag-rct-nature-healthNorway's government announced Oct 5 proposals for a temporary AI-glasses ban in selected public venues and a regulatory expert group. No already enacted nationwide prohibition or completed expert recommendation is evidenced.
evt-hist-2026phase29-norway-ai-glasses-proposalSeattle council passed CB 121267 Sep 22; mayor signed it Oct 5, 2026, ordinance 127515. Prohibition concerns data-driven grocery price discrimination; enactment and later enforcement are not the same.
evt-hist-2026phase29-seattle-pricing-ordinance-signingDavid Robinson, who led transparency work for OpenAI's safety team and contributed to the Preparedness Framework and safety reporting for frontier-model launches, left OpenAI and publicly argued that the company's culture and safety processes are failing to scale with increasingly capable systems and rapid iterative deployment. This is best treated as an insider organizational/safety signal rather than proof of any single technical failure. Its Layer.re value is longitudinal: it can later be compared against OpenAI safety reports, deployment incidents, policy/process changes, and subsequent insider or external evidence.
evt-openai-robinson-resignation-20261003The fan translation team released a patch for the Japanese PlayStation game Spectral Tower including a playable English localization and patch instructions on GitHub. Scope and credits are documented by the release; exhaustive optional-branch QA is not independently verified. The project is not evidence of AI-assisted localization.
evt-rts-20261010-spectral-tower-ps1-english-patchAnthropic launched Claude Frontier Academy with a $100 million commitment to train 10,000 Frontier Deployed Engineers by end-2027. Initial cohorts include engineers from major consultancies, financial institutions and Novo Nordisk. The program trains and assesses engineers on realistic deployment cases, explicitly targeting production agentic systems and enterprise implementation. This is evidence that human integration expertise is being treated by a frontier lab as a scaling constraint alongside models and infrastructure; the bottleneck interpretation remains a supported inference because the claim that talent is the 'most pressing' implementation issue is Anthropic's own framing.
evt-anthropic-frontier-academy-20261002Pope Leo XIV addressed visiting artists on Oct 2 2026, saying human art has a spark distinct from machine-generated results. This is a papal cultural assessment and rhetorical stance, not new civil legislation or evidence about model consciousness.
evt-hist-2026phase27-pope-leo-art-ai-speechMeta published open ESP32 and Linux Muse Gadget SDKs and Home Link documentation that let users connect Muse to displays, buttons, sensors, actuators, Linux systems and compatible local devices. The event supports separation between a persistent agent service and replaceable hardware access surfaces without establishing a standardized portable agent-identity architecture.
evt-meta-muse-gadgets-20261002On October 1 Google announced Guided Vision availability in Gemini Live on compatible Android devices, developed with blind and low-vision testers. The existence of a feature is distinct from independently verified accessibility outcomes for all users.
evt-hist-2026phase26-gemini-guided-visionOn October 1 three NYC agencies issued a policy statement that existing worker, consumer, civil-rights and taxi rules remain enforceable when AI is used. The statement clarifies existing-law enforcement; it does not by itself create new statutes or prove enforcement cases are resolved.
evt-hist-2026phase26-nyc-ai-enforcement-policyThe Senate voted 57-43 to advance legislation directing consideration of cost-allocation standards for large electricity users such as data centers, below the 60 votes needed.
evt-us-senate-dc-cost-vote-20260930Anthropic published its September 30, 2026 robot exposure index. The report estimates robots can perform 74 percent of physical task-hours in at least some conditions, while only 0.3 percent of total tasks are cost-competitive. These are rubric- and model-based exposure estimates, not observed economy-wide robot job displacement.
evt-hist-20260930-anthropic-robot-exposure-researchOn 30 September 2026 California Governor Newsom announced signed AI-related protections including human oversight of employment decisions and transparency in AI-linked layoffs, alongside a terminology executive order. Legislation signing is distinct from future enforcement; kill-switch proposals were not thereby operational.
evt-hist-20260930-california-ai-protections-sept30CDC published its 2025-2026 FluSight evaluation on 30 September 2026. Google_SAI-FluEns was the best performing individual submission on relative weighted interval score, while the CDC operational FluSight ensemble ranked seventh; the historical evaluated season predates the publication. No claim of future clinical efficacy.
evt-hist-20260930-google-flusight-cdc-studyPew published 30 September 2026 its controlled study of synthetic respondents versus actual representative survey participants. Its observed errors, especially for subgroups, support conclusions for tested methods and models, not all future synthetic-sampling designs.
evt-hist-20260930-pew-synthetic-surveysQuinnipiac published September 30, 2026 polling: 30 percent said stop developing powerful AI until safety is evaluated, 47 percent slow down, and 86 percent backed independent safety standards. Combined 77 percent is the sum of two answer categories, not a single response option or universal population consensus.
evt-hist-20260930-quinnipiac-ai-pollOn September 29, 2026, US Executive Order 14434 directed executive-branch departments and agencies, to the maximum extent permitted by law, to use 'Super Intelligence' (SI) in place of 'Artificial Intelligence' (AI) for new official correspondence, websites, reports and other non-statutory materials. It does not require rewriting previously issued regulations or historical documents. Under section 3, the order temporarily uses the pre-existing statutory AI definition, while requesting proposed legislative language for an updated SI definition within 60 days. This is an institutional naming and messaging change, not evidence that present models have attained scientific superintelligence, not a new generally applicable statute, and not proof of improved public acceptance.
evt-hist-us-ai-si-terminology-eo-20260929On September 28, 2026 Mathathon organizers and Caltech colleagues issued a revised joint statement framing Mathathon as Old Problems, New Proofs, with emphasis on mathematical understanding and explainers. It is a public organizing statement, not the completion of a competition or independent verification of any mathematical result.
evt-hist-20260928-mathathon-joint-statementBCG used the term Human Context Layer for structured, permissioned human-specific context that enables personal agents to act with greater continuity and trusted delegation.
evt-bcg-human-context-layer-20260924Anthropic published Project Swap on September 24, 2026: a small controlled agent-to-agent book barter study with employees across six offices. In disclosed trial participants’ agent-produced rankings matched individual rankings for 61% of compared book pairs. This is not field-scale deployment of autonomous general economic negotiation.
evt-hist-20260924-project-swapCNN/SSRS released September 24 results of US survey fielded September 16-17 among 1206 adults aged 18+, probability recruited, via online/telephone. SSRS overall sampling margin ±3.5 percentage points at 95% confidence with 1.6 design effect. CNN reported near-nine-in-ten are more concerned than excited about expanding AI datacentres (CNN Sept29 TV cited 87%); not 87% formally opposing any specific project, nor worldwide opinion. Detailed questionnaire/crosstabs not independently read.
evt-hist-20261017-env-ssrsThe European Commission proposed a common data-center rating scheme and opened work toward minimum performance standards covering energy and water use, efficiency and grid integration.
evt-eu-datacenter-framework-20260921On September 15, 2026 Meta announced Meta One subscription bundles across Facebook, Instagram, WhatsApp and Meta AI with expanded creative AI usage. Meta described a gradual rollout, published starting US prices and its own cumulative trial/subscription figures. This records official product introduction and stated rollout, not independently confirmed adoption or an inference that free Meta AI is retired.
evt-hist-meta-one-subscriptionsCNN reported Barack Obama's call for Democrats to make AI regulation a central agenda item, while the same segment described competing views on safety, regulation and strategic competition.
evt-us-ai-electoral-agenda-20260913ElevenLabs publicly introduced Dubbing v2 on 28 May 2026, with source-performance conditioned dubbing in over 90 languages and synchronisation-aware translations. At launch it was available through ElevenCreative and ElevenProductions. API access was explicitly future; ElevenLabs' separate 6 August 2026 announcement documents later API availability. Capabilities and quality comparisons here are attributed to ElevenLabs, not independently measured.
evt-hist-20260528-dubbingv2On 25 March 2026 ARC Prize introduced ARC-AGI-3 as a benchmark for agent learning in novel interactive environments. Organiser-reported frontier performance figures are benchmark-specific and not independent general intelligence measurements.
evt-pc26q1-eeebc1f21afeLiteLLM documented a supply-chain compromise involving PyPI releases 1.82.7 and 1.82.8 published on 24 March 2026. The precise affected installations, downstream exfiltration and alleged Mercor causal chain require separate proof.
evt-pc26q1-15f50e12407fOn 20 March 2026 the White House published a National AI Legislative Framework proposing congressional action. It was not an enacted law; the December 2025 executive action is separate background.
evt-pc26q1-17c0abc99c16Stanford Digital Economy Lab described portable and persistent personal memory through a unified Human Context Protocol, strengthening the case for context continuity and interoperability across personal AI systems.
evt-stanford-portable-memory-20260316On 27 February 2026 a public US Defense Secretary directive regarding an Anthropic supply-chain risk designation became the subject of Anthropic response. Subsequent formal determination on 3 March and letter on 4 March are distinct legal steps.
evt-pc26q1-686f709e08ceOn 21 February 2026 India reported adoption or endorsement of the New Delhi AI Impact Declaration following its summit, with a stated 92 countries and organizations. Diplomatic endorsement is not binding international law or implemented policy.
evt-pc26q1-4062a59156e3On 11 February 2026 Google DeepMind published research discussing Aletheia, an iterative mathematics agent. Its supporting arXiv manuscript first appeared on 10 February and was subsequently revised; individual theorem claims require independent proof checks.
evt-pc26q1-ac9e3ec6cf0fOn 5 February 2026 OpenAI and Ginkgo published results of adaptive, robot-executed cell-free protein synthesis experiments. Numerical performance and cost reductions are collaborator-reported, not independent replication or unrestricted autonomous discovery.
evt-pc26q1-d3cbfabbb22fThe international AI Safety Report 2026 was published on 3 February 2026 as an expert assessment of general-purpose AI risks, capabilities and mitigation. Publication does not establish binding regulation or the empirical truth of forecasts.
evt-pc26q1-8514f0ffe730The macOS app introduces multi-agent task supervision and isolated worktrees.
evt-2026q1-codex-appAnthropic's following-day announcement confirms Cowork launched as a research preview on January 12.
evt-2026q1-cowork-previewOn 6 January 2026 Utah announced a bounded Doctronic AI prescription-renewal pilot under medical practitioner supervision. The pilot had begun in October 2025; the January milestone is public announcement, not first operation or unrestricted prescribing.
evt-pc26q1-05487da896bbStanford Digital Economy Lab published the Human Context Protocol as an explicit portable, user-governed architecture for AI personalization and human-specific context.
evt-stanford-hcp-20251218A randomized study of sixteen experienced developers reports slower completion with early-2025 AI tools on their own repositories.
evt-2025q3-metr-rctAnthropic released hybrid extended-thinking Claude 3.7 Sonnet and a command-line Claude Code research preview.
evt-2025q1-claude37-codeOpenAI launched a multi-step web research feature initially for Pro users.
evt-2025q1-deep-researchOpenAI launched a browser-operating research preview for US Pro users with takeover and confirmation boundaries.
evt-2025q1-operatorGoogle introduced a browsing research workflow in Gemini Advanced with a user-reviewable plan and cited report.
evt-2024q4-deep-researchOpenAI began a paid Sora video-generation product rollout with geographic and subscription restrictions.
evt-2024q4-sora-productOpenAI introduced source-linked search within ChatGPT, initially for paid users and SearchGPT waitlist users.
evt-2024q4-searchApple made initial intelligence features available in US-English beta on eligible devices.
evt-2024q4-apple-betaOpenAI introduced a separate editable writing and coding workspace in beta.
evt-2024q4-canvasOpenAI launched a low-latency speech-to-speech API in public beta.
evt-2024q4-realtimeOpenAI launched SearchGPT to a small group of users and publishers with source-linked search answers.
evt-2024q3-searchgptAnthropic introduced dedicated artifact panes for generated code, documents and designs alongside conversation.
evt-2024q2-artifactsMicrosoft documented June 18 general availability of the first Copilot+ PC category and supported local AI experiences.
evt-2024q2-copilot-pcsMicrosoft changed Recall from a June 18 broad preview to a future Windows Insider preview.
evt-2024q2-recall-delayApple announced an OS-integrated system combining device and server models, with a US-English beta planned for autumn.
evt-2024q2-apple-intelligenceOpenAI began GPT-4o text/image rollout in ChatGPT and made text/vision API access available.
evt-2024q2-gpt4oMeta released Llama 3 8B/70B models and integrated Llama-based Meta AI into its apps in additional markets.
evt-2024q2-llama3Anthropic made Opus and Sonnet available through Claude and its API with image understanding and a 200K initial context window.
evt-2024q1-claude3OpenAI began a memory test with a small portion of Free and Plus users, including view, delete and disable controls.
evt-2024q1-memoryGoogle launched paid Gemini Advanced access to Ultra 1.0.
evt-2024q1-gemini-ultraOpenAI began GPT Store rollout to Plus, Team and Enterprise users.
evt-2024q1-gpt-storeGoogle Cloud made Gemini Pro available through AI Studio and Vertex AI.
evt-gemini-pro-api-20231213Council and Parliament negotiators reached a provisional agreement including GPAI/foundation-model rules and an EU-level governance architecture.
evt-eu-ai-act-deal-20231209Google introduced Gemini Ultra, Pro and Nano as a natively multimodal family spanning data-center and on-device deployment.
evt-gemini10-20231206Claude 2.1 became available with 200K context, system prompts and beta tool use.
evt-claude21-20231121OpenAI launched GPT-4 Turbo preview with 128K context, improved function calling, vision, DALL-E 3 and TTS APIs.
evt-gpt4turbo-20231106OpenAI introduced reusable customized ChatGPTs combining instructions, knowledge, tools and optional Actions to external APIs.
evt-gpts-20231106Microsoft 365 Copilot became generally available for enterprise customers with Graph/app integration inside existing security/privacy commitments.
evt-m365-copilot-ga-20231101The U.S. executive order established government-wide AI actions including reporting requirements for certain dual-use foundation models and large compute clusters.
evt-us-ai-eo-20231030Microsoft began rolling Copilot in Windows in preview with the Windows 11 September update.
evt-windows-copilot-20230926OpenAI began rolling voice and image capabilities to Plus and Enterprise users over a two-week period.
evt-chatgpt-vision-voice-20230925OpenAI launched ChatGPT Enterprise with enterprise security, administration, identity controls and expanded model/data-analysis access.
evt-chatgpt-enterprise-20230828Anthropic released Claude 2 via API and a public-facing beta website.
evt-claude2-20230711OpenAI began rolling Code Interpreter in beta to all ChatGPT Plus users.
evt-code-interpreter-20230706Parliament adopted its negotiating position on the proposed AI Act, including generative-AI provisions.
evt-eu-ai-act-parliament-20230614Browsing and third-party plugins began rolling to Plus users as beta features.
evt-plugins-rollout-20230512Claude's context window expanded from roughly 9K to 100K tokens.
evt-claude-100k-20230511Bard expanded beyond its Q1 US/UK early access to more than 180 countries and territories in English.
evt-bard-expansion-20230510The Garante reported restoration after transparency and user-rights changes.
evt-chatgpt-italy-restored-20230428OpenAI added chat-history/training controls and data export and announced stronger business data policies.
evt-chatgpt-data-controls-20230425NTIA opened a policy process on audits, assessments, certifications and AI accountability mechanisms.
evt-ntia-accountability-20230411The Italian data-protection authority imposed an immediate temporary limitation on processing Italian users' data by OpenAI.
evt-garante-chatgpt-20230331OpenAI introduced a small alpha enabling browsing, code execution and third-party-service interaction through ChatGPT plugins.
evt-chatgpt-plugins-20230323Google opened Bard early access in the US and UK as an experimental conversational AI service.
evt-bard-20230321Microsoft introduced an architecture connecting LLMs to Microsoft Graph and Office apps, initially tested with 20 customers.
evt-m365-copilot-20230316Anthropic made Claude more broadly available through chat/API surfaces and described third-party integrations.
evt-claude-20230314OpenAI released GPT-4 with text access through ChatGPT Plus and API waitlist expansion; image input remained research preview.
evt-gpt4-20230314Google announced PaLM API and MakerSuite developer/prototyping surfaces alongside Workspace generative-AI features.
evt-palm-api-20230314OpenAI made conversational GPT-3.5 Turbo and Whisper directly consumable through developer APIs.
evt-chatgpt-whisper-api-20230301Microsoft combined web search, browser interaction and conversational generative AI in a limited preview.
evt-bing-edge-ai-20230207NIST finalized its voluntary AI Risk Management Framework 1.0 around Govern, Map, Measure and Manage.
evt-nist-airmf10-20230126Anthropic published a method in which models critique and revise outputs according to written principles and use AI-generated preferences for reinforcement learning.
evt-constitutional-ai-20221215GitHub launched Copilot for Business with organization-level licensing, centralized policy controls and privacy commitments.
evt-copilot-business-20221207The Council adopted its common position and negotiating mandate on the proposed horizontal AI Act.
evt-eu-ai-act-general-approach-20221206OpenAI released ChatGPT through an accessible conversational interface supporting multi-turn dialogue and trained with RLHF, while documenting hallucination, prompt sensitivity, verbosity and safety limitations.
evt-chatgpt-20221130Meta published CICERO, combining strategic planning and a controllable language model to negotiate and coordinate with humans in online Diplomacy games.
evt-cicero-20221122OpenAI made DALL·E available through a developer API so third parties could integrate image generation into applications, with built-in moderation.
evt-dalle-api-20221103OpenAI opened DALL·E beta signup without a waitlist and reported more than 1.5 million active users generating more than two million images per day.
evt-dalle-no-waitlist-20220928Sparrow combined reinforcement learning from human feedback, explicit natural-language behavior rules and evidence retrieval for factual dialogue.
evt-sparrow-20220922OpenAI open-sourced Whisper models and inference code for multilingual transcription and translation after training on 680,000 hours of multilingual and multitask supervised web data.
evt-whisper-open-20220921Stability AI publicly released Stable Diffusion under CreativeML OpenRAIL-M for commercial and non-commercial use subject to responsible-use conditions.
evt-stable-diffusion-public-20220822NIST released the second draft of its voluntary AI RMF and an initial Playbook with Govern and Map implementation guidance.
evt-nist-airmf2-20220818Meta released BlenderBot 3 with internet search, long-term-memory capability, public user feedback and shared model/data/code for conversational-agent research.
evt-blenderbot3-20220805Meta released NLLB-200, FLORES-200 and supporting code, broadening multilingual model access and translation coverage.
evt-nllb-200-20220706GitHub made Copilot generally available as an editor-integrated AI pair programmer after a technical preview involving more than 1.2 million developers.
evt-copilot-ga-20220621OpenAI reported more than three million generated images from early users, expanded onboarding and changes to filters, automated detection and human review during the DALL·E 2 research preview.
evt-dalle-preview-20220518Flamingo accepted interleaved images, video and text and performed many multimodal tasks from few-shot prompts without task-specific fine-tuning.
evt-flamingo-20220428NIST released an initial voluntary draft framework for structured, measurable AI risk management across design, development, use and evaluation.
evt-nist-airmf-20220317OpenAI said models fine-tuned with human demonstrations and preference rankings were now the default models on its API, operationalizing instruction-following post-training in a production service.
evt-openai-instructgpt-20220127