{
 "retrieved": "2026-11-17",
 "note": "Verified milestones, newest first. Claims are reported as claims, with disputes where they exist.",
 "items": [
  {
   "date": "2026-11-13",
   "kind": "Expert timelines",
   "title": "Dwarkesh Patel shortens his timeline for automated AI research",
   "what": "In an essay titled \"The avalanche that hit math is coming for AI development\", Patel writes that AI as good as the best human AI researchers is a couple of years away, possibly months. His median is 2028. Earlier in 2026 he put superintelligence at least most of a decade away.",
   "work": "A prominent long-timelines voice moved sharply after the math results. Plans that assume a decade of slow change are the ones to stress-test.",
   "src": "Dwarkesh Patel, Dwarkesh Podcast, Nov 13, 2026",
   "url": "https://www.dwarkesh.com/p/rsi"
  },
  {
   "date": "2026-11-09",
   "kind": "Capability",
   "title": "New top score on GDPx: 16.4%",
   "what": "Claude Fable 5.2 was listed first on SemiAnalysis's GDPx leaderboard at 16.4%, a week after launch, ahead of Claude Opus 5.5 at 12.0%.",
   "work": "The share of multi-hour professional tasks an AI finishes alone went from 12.0% to 16.4% in the benchmark's first week.",
   "src": "ArtificialWatch, Nov 10, 2026, reading the GDPx leaderboard",
   "url": "https://artificialwatch.com/wire/claude-fable-5-2-tops-semianalysis-gdpx-at-16-4-percent"
  },
  {
   "date": "2026-11-02",
   "kind": "Capability",
   "title": "SemiAnalysis launches GDPx: the best model scores about 12%",
   "what": "GDPx tests AI agents working alone for up to eight hours on tasks in 11 fields, including chip design, hardware, finance, legal and software engineering. SemiAnalysis reported that the best model scores about 12%, and that under strict pass-or-fail grading every model scored zero on finance and legal.",
   "work": "AI already completes about one multi-hour professional task in eight with no human help. The score covers defined tasks; judgment, accountability and trust, the blind spot of AI, sit outside it.",
   "src": "SemiAnalysis, Nov 2, 2026",
   "url": "https://newsletter.semianalysis.com/p/gdpx-measuring-frontier-model-capabilities"
  },
  {
   "date": "2026-10-28",
   "kind": "Usage",
   "title": "Google's model APIs pass 32 billion tokens a minute",
   "what": "Alphabet said its first-party model APIs process nearly 32 billion tokens a minute, more than four times a year earlier.",
   "work": "Businesses are building AI into their own systems at scale. API traffic is work being routed to models, not people trying a chatbot.",
   "src": "Alphabet, Q3 2026 earnings call, Oct 28, 2026",
   "url": "https://blog.google/company-news/inside-google/message-ceo/alphabet-earnings-q3-2026/"
  },
  {
   "date": "2026-10-26",
   "kind": "Math and science",
   "title": "OpenAI claims an existence proof for four-dimensional gauge theories, including QCD",
   "what": "OpenAI said its models produced existence proofs for interacting quantum field theories with gauge fields and fermions in four dimensions, a family that includes QCD, the theory of the strong force. A committee led by the physicist David Gross is examining the work before the proof is released. As of mid-November no independent check or peer review had been published, and the mass-gap part of the related Millennium Prize problem remains open. This is a claim, not a verified result.",
   "work": "If it holds, AI is producing frontier mathematics, not only routine analysis. Until independent experts confirm it, treat it as a claim.",
   "src": "Interesting Engineering, Oct 26, 2026; Tech Insider, Nov 2026",
   "url": "https://interestingengineering.com/ai-robotics/openai-proofs-math-computing-physics"
  },
  {
   "date": "2026-10-06",
   "kind": "Math and science",
   "title": "OpenAI publishes about 720 AI-written math manuscripts",
   "what": "OpenAI released a public repository of manuscripts from an unreleased internal model working on open problems. On Nov 17, 2026 the repository listed 721 manuscripts in 373 families; about 42% of top-line results have machine-checked Lean proofs. Mathematicians including Frank Calegari and Tristan Buckmaster have called many papers poorly written and some wrong, and OpenAI did not release the prompts.",
   "work": "Research output that took careers can now be produced in bulk. Checking it, judging which results matter and taking responsibility for them stays with people.",
   "src": "OpenAI math repository on GitHub (Apache 2.0); Let's Data Science and Tech Insider coverage, Oct to Nov 2026",
   "url": "https://github.com/openai/math"
  },
  {
   "date": "2026-09-08",
   "kind": "Math and science",
   "title": "OpenAI claims a proof of blow-up for the 3D Navier–Stokes equations",
   "what": "OpenAI said an internal system of about 10,000 agents, running for 88 hours, produced a 166-page proof with a Lean formalization that a smooth fluid at rest can develop a singularity under a smooth external force. The Clay Mathematics Institute said on Sep 11 that the problem has apparently been settled, with its review deliberately unhurried. The unforced case remains open, and mathematician Tristan Buckmaster disputes how credit was handled.",
   "work": "One of the seven Millennium Prize problems drew a claimed answer from 88 hours of machine time. Mathematicians say the proof teaches little so far: understanding and explaining it remains human work.",
   "src": "Clay Mathematics Institute, Sep 11, 2026; NPR via GPB, Sep 22, 2026; The Tufts Daily, Sep 24, 2026",
   "url": "https://www.claymath.org/news/navier-stokes-announcement/"
  },
  {
   "date": "2026-09-08",
   "kind": "Math and science",
   "title": "Terence Tao warns good open problems are being used up",
   "what": "Tao wrote that fruitful open problems are being depleted in a non-renewable fashion as AI tools solve them faster than new ones are found.",
   "work": "When answers get cheap, the scarce skill moves to choosing the right questions. That shift is coming to every field that runs on defined problems.",
   "src": "Terence Tao on Mathstodon, Sep 8, 2026",
   "url": "https://mathstodon.xyz/@tao/117237320796901560"
  },
  {
   "date": "2026-08-25",
   "kind": "Adoption",
   "title": "McKinsey: 39% expect AI to cut headcount next year",
   "what": "In McKinsey's 2026 survey of 1,719 respondents, 39% expect AI to reduce their organization's total headcount over the next year, up from 32%. Among large enterprises, 40% are scaling AI agents, up from 27%.",
   "work": "Workforce plans are being written around AI before the hiring data shows it. That is the gap this page exists to close.",
   "src": "McKinsey, The state of AI in 2026, Aug 25, 2026",
   "url": "https://www.mckinsey.com/capabilities/mckinsey-analytics/our-insights/the-state-of-ai"
  },
  {
   "date": "2026-07-23",
   "kind": "Math and science",
   "title": "AI systems report perfect scores at the 2026 Math Olympiad",
   "what": "AFP reported that Huawei and Xiaohongshu said their systems scored 42 of 42 at IMO 2026. Xiaohongshu said its answers were submitted to the organisers for grading. A year earlier, Google DeepMind's gold-medal 35 of 42 was officially certified by IMO coordinators.",
   "work": "Hard, well-posed problems with a checkable answer are now within reach of machines. Most paid work is less well-posed, which is where human judgment still earns its keep.",
   "src": "AFP via The Standard, Jul 23, 2026; Google DeepMind, Jul 21, 2025",
   "url": "https://www.standardmedia.co.ke/sci-tech/article/2001553533/ai-catches-up-with-humans-to-score-100pc-at-top-maths-contest"
  },
  {
   "date": "2026-06-11",
   "kind": "Adoption",
   "title": "One in five Canadian businesses now uses AI",
   "what": "Statistics Canada reported that 19.2% of businesses used AI to produce goods or deliver services in the past 12 months, up from 6.1% two years earlier.",
   "work": "Adoption roughly tripled in two years. The occupations inside those businesses feel it first in what the role requires, before it shows in postings counts.",
   "src": "Statistics Canada, Jun 11, 2026",
   "url": "https://www150.statcan.gc.ca/n1/pub/11-621-m/11-621-m2026010-eng.htm"
  },
  {
   "date": "2026-05-20",
   "kind": "Math and science",
   "title": "OpenAI reports a model disproved Erdős's unit distance conjecture",
   "what": "OpenAI said a model found a counterexample to Erdős's 1946 planar unit distance conjecture, checked by external mathematicians. Mathematician Daniel Litt called it the first AI result he found interesting in itself.",
   "work": "AI moved from answering known questions to producing results experts value on their own merits.",
   "src": "Let's Data Science, May 28, 2026; Quanta Magazine, Aug 3, 2026",
   "url": "https://www.quantamagazine.org/how-ai-tore-through-a-mathematical-community-20260803/"
  },
  {
   "date": "2026-05-19",
   "kind": "Usage",
   "title": "Google processes 3.2 quadrillion tokens a month",
   "what": "Google said it processes over 3.2 quadrillion tokens a month across its products, about seven times the 480 trillion of a year earlier.",
   "work": "Use is compounding far faster than any hiring series moves. That is why use is a leading signal.",
   "src": "Google, Google I/O 2026, May 19, 2026",
   "url": "https://blog.google/innovation-and-ai/sundar-pichai-io-2026/"
  },
  {
   "date": "2026-04-23",
   "kind": "Capability",
   "title": "Professional benchmark reaches 84.9%",
   "what": "OpenAI reported that GPT-5.5 matched or beat an industry professional's deliverable in 84.9% of GDPval comparisons, up from 47.6% for the best model seven months earlier. OpenAI makes both the benchmark and the model.",
   "work": "On defined, well-specified tasks, the best models now match or beat experienced professionals most of the time.",
   "src": "OpenAI, Introducing GPT-5.5, Apr 23, 2026",
   "url": "https://openai.com/index/introducing-gpt-5-5/"
  },
  {
   "date": "2026-04-07",
   "kind": "Capability",
   "title": "Autonomous task length passes 16 hours",
   "what": "METR measured Claude Mythos Preview (early) at a 50% time horizon of about 17 hours of skilled-human work, and noted its task suite cannot measure reliably above 16 hours. A year earlier the best measured model was at about 2 hours.",
   "work": "Agents now finish, half the time, tasks that take a skilled person two working days. Work is being handed over in larger pieces.",
   "src": "METR, Time Horizon 1.1 results, entry added May 8, 2026",
   "url": "https://metr.org/time-horizons/"
  },
  {
   "date": "2026-03-05",
   "kind": "Usage",
   "title": "Anthropic publishes AI use by occupation",
   "what": "Anthropic's \"observed exposure\" estimates how much of each occupation's work time is on tasks where Claude is already used for work. Computer programmers scored highest at 75%. The authors found no systematic rise in unemployment for exposed workers, and suggestive evidence of slower hiring of 22 to 25 year olds into exposed occupations.",
   "work": "This is the first public measure of actual AI use by occupation. JDI shows it on each occupation page next to hiring demand.",
   "src": "Anthropic, Labor market impacts of AI, Mar 5, 2026",
   "url": "https://www.anthropic.com/research/labor-market-impacts"
  },
  {
   "date": "2025-09-25",
   "kind": "Capability",
   "title": "OpenAI launches GDPval",
   "what": "GDPval scores AI against professionals on defined tasks from 44 occupations. The best model at launch, Claude Opus 4.1, matched or beat the professional in 47.6% of comparisons.",
   "work": "The first benchmark built on the deliverables of real occupations, which makes it readable next to job data.",
   "src": "OpenAI, GDPval paper, Sep 25, 2025 (CC BY 4.0)",
   "url": "https://arxiv.org/abs/2510.04374"
  },
  {
   "date": "2025-07-21",
   "kind": "Math and science",
   "title": "AI earns an officially graded Math Olympiad gold",
   "what": "Google DeepMind's Gemini Deep Think scored 35 of 42 at IMO 2025, graded and certified by IMO coordinators.",
   "work": "The start of the run of math results above.",
   "src": "Google DeepMind, Jul 21, 2025",
   "url": "https://deepmind.google/blog/advanced-version-of-gemini-with-deep-think-officially-achieves-gold-medal-standard-at-the-international-mathematical-olympiad/"
  }
 ]
}