{
  "schema": "hangingcontext.desk/1",
  "type": "story",
  "id": "1168",
  "slug": "metr-anthropic-rd-acceleration-assessment",
  "edition": "2026-11-06T00",
  "url": "https://hangingcontext.com/stories/2026-11-06/metr-anthropic-rd-acceleration-assessment/",
  "newsroom": "artificial-intelligence",
  "beat": "research",
  "newsroom_title": "Artificial Intelligence",
  "angle": "Anthropic's own R&D is now accelerating twice as fast, mostly because of AI — and its safety auditor says the public story understates it",
  "headline": "Anthropic's research is accelerating twice as fast, and its auditor says that's understated",
  "dek": "METR's high-access review finds Anthropic improving capabilities at roughly twice the 2024-25 rate, driven by AI use — and says the public risk report was overconfident.",
  "published_at": "2026-11-06T00:40:13Z",
  "labels": {
    "trust": "confirmed",
    "trust_text": "Confirmed · 4 Manifests",
    "newsroom": "Artificial Intelligence"
  },
  "source_count": 4,
  "publishers": [
    "anthropic-research",
    "metr-research",
    "microsoft-research-blog-rss"
  ],
  "sources": [
    "anthropic-research",
    "metr-research",
    "microsoft-research-blog-rss"
  ],
  "distinct_source_count": 3,
  "key_facts": [
    {
      "text": "METR found Anthropic is likely improving model capabilities at about twice the rate of March 2024 to October 2025, with benchmark scores up 1.9x since October 2025 — and says the acceleration was not driven by more training compute.",
      "self_reported": false,
      "cites": [
        {
          "manifest_id": "1793915667368443123",
          "claim_ids": [
            "1793915912183728879",
            "1793915912185099300",
            "1793915912186209086"
          ]
        }
      ]
    },
    {
      "text": "METR estimates about 80% of the acceleration is attributable to AI usage, and Anthropic staff self-reported uplift from AI systems reaching up to 12x in October 2026.",
      "self_reported": true,
      "cites": [
        {
          "manifest_id": "1793915667368443123",
          "claim_ids": [
            "1793915912188008837",
            "1793915912192678051"
          ]
        }
      ]
    },
    {
      "text": "METR says it is plausible Anthropic's early warning threshold was already crossed by July 15th, calls the public report's conclusion on the 2x acceleration threshold overconfident, and puts leading-indicator estimates at 1.3 to 2.8.",
      "self_reported": false,
      "cites": [
        {
          "manifest_id": "1793915667368968000",
          "claim_ids": [
            "1793915914494030202",
            "1793915914502343420",
            "1793915914500917417"
          ]
        }
      ]
    },
    {
      "text": "Anthropic's analysis of 207,049 research sessions found data analysis accounts for 33.7% of Claude sessions and 48.1% of Claude Code sessions, and 40% of surveyed researchers expect AI to run full research projects alone by 2030.",
      "self_reported": false,
      "cites": [
        {
          "manifest_id": "1793911038564166314",
          "claim_ids": [
            "1793911259775144728",
            "1793911259789821799",
            "1793911259858740584"
          ]
        }
      ]
    }
  ],
  "images": [],
  "sections": [
    {
      "heading": "What METR found",
      "paragraphs": [
        {
          "markdown": "METR Research performed a high-access evaluation at Anthropic from August 12 to October 29, 2026, assessing AI research and development acceleration, and concluded that as of October 2026, Anthropic is likely improving model capabilities on well-specified tasks at approximately twice the rate of March 2024 to October 2025. Anthropic's benchmark scores have accelerated 1.9 times since October 2025, with a 95% confidence interval of 1.5 to 2.8 times. METR is confident the observed acceleration was not due to an acceleration in the scaling of training compute.",
          "cites": [
            {
              "manifest_id": "1793915667368443123",
              "claim_ids": [
                "1793915912181767975",
                "1793915912183728879",
                "1793915912185099300"
              ]
            }
          ]
        },
        {
          "markdown": "METR estimates that approximately 80% of current acceleration in AI research and development is attributable to AI usage. Anthropic staff self-reported uplift from AI systems reaching up to 12 times in October 2026. METR did not identify significant bottlenecks in algorithmic progress, training compute, experiment compute, or data that will take effect in the next 12 months. The assessment was conducted by 2.5 full-time-equivalent METR staff over eleven weeks, working on Anthropic devices for security reasons, and METR believes there is a significant risk that further capabilities progress could lead to even faster acceleration over the coming year.",
          "cites": [
            {
              "manifest_id": "1793915667368443123",
              "claim_ids": [
                "1793915912188008837",
                "1793915912192678051",
                "1793915912190735948"
              ]
            }
          ]
        },
        {
          "markdown": "In a separate review of the risks-from-automated-research section of Anthropic's August 2026 Risk Report, METR agrees with Anthropic that catastrophic risks from models as of July 15th were low — but says it is plausible that Anthropic's early warning threshold was already crossed by July 15th. Anthropic's public report claims leading indicators show less than a 2x acceleration, and METR calls that conclusion overconfident: its own analysis of leading indicators finds a range of estimates between 1.3 and 2.8, and Anthropic's private report implies a 1.97x acceleration rate for certain progress metrics. Anthropic argues the Full Substitution Threshold has not been crossed because research staff still complete work; METR's review, conducted with iterative feedback between September 22 and October 28, disputes the emphasis, not the data.",
          "cites": [
            {
              "manifest_id": "1793915667368968000",
              "claim_ids": [
                "1793915914492345127",
                "1793915914494030202",
                "1793915914502343420",
                "1793915914500917417",
                "1793915914503506869"
              ]
            }
          ]
        }
      ]
    },
    {
      "heading": "The evidence problem",
      "paragraphs": [
        {
          "markdown": "METR's assessment is subject to high uncertainty, with a possible range of 1.3x to 3x the historical rate — and the public cannot check the work. Anthropic redacted a majority of the evidence METR found most informative for its risk conclusions, describing the redactions as aimed at removing confidential information that could accelerate other AI developers. METR also did not request or receive access to sample from non-public models, and Anthropic cancelled a compute allocation component of the assessment that was originally contracted. The result is a strange position: the most informed independent judgement available is that acceleration is real and roughly twofold, and the evidence for it is mostly hidden.",
          "cites": [
            {
              "manifest_id": "1793915667368443123",
              "claim_ids": [
                "1793915912189813266",
                "1793915912195490357",
                "1793915912200771584"
              ]
            }
          ]
        },
        {
          "markdown": "Anthropic's own analysis of how researchers use Claude gives the acceleration a human shape. The study analysed 207,049 research sessions on Claude and Claude Code, and found data analysis accounts for 33.7% of sessions on Claude and 48.1% on Claude Code, with heavier use in formal sciences like computer science, mathematics, and statistics. Current AI capabilities could double research output in some fields if widely adopted, and full automation of non-physical research work could increase output up to 15-fold in fields like computer science, against three- or fourfold in bench sciences where physical bottlenecks bind. Computer science accounts for 32.7% of Claude Code sessions despite representing only 4.3% of researchers.",
          "cites": [
            {
              "manifest_id": "1793911038564166314",
              "claim_ids": [
                "1793911259775144728",
                "1793911259789821799",
                "1793911259834393153"
              ]
            }
          ]
        },
        {
          "markdown": "Researchers themselves expect the trend to continue: 40% of surveyed researchers expect AI to carry out full research projects on their own by 2030, and 64% by 2035. Ideation and hypothesis generation are under-represented in AI usage compared with researcher time, which suggests the current acceleration is concentrated in execution — analysis, writing, coding — rather than in choosing what to study. That is consistent with METR's finding: the machine is making research faster, and the human is still deciding what is worth researching.",
          "cites": [
            {
              "manifest_id": "1793911038564166314",
              "claim_ids": [
                "1793911259858740584",
                "1793911260035835688",
                "1793911259815621014"
              ]
            }
          ]
        }
      ]
    },
    {
      "heading": "What to do with a 2x number",
      "paragraphs": [
        {
          "markdown": "If research output doubles, verification has to keep up, and Microsoft Research's PDD Kit is one attempt at that. AI agents generate code at a scale that is difficult for humans to review line by line, so PDD Kit converts developer intent into a Verus specification, proves that AI-generated code meets specifications for every input, and uses testing where formal proofs are not practical. The point is to let developers review intent rather than every line of code. The same logic applies to research: when AI writes the code and runs the analysis, the human's job moves up a level, from doing the work to stating what the work must prove.",
          "cites": [
            {
              "manifest_id": "1793913954341774529",
              "claim_ids": [
                "1793914399879592596",
                "1793914399882373676",
                "1793914399883228181",
                "1793914399887772370"
              ]
            }
          ]
        },
        {
          "markdown": "METR expects AI's augmentation of research and development to continue to grow over the next six months, and warns of a significant risk that further capabilities progress could lead to even faster acceleration over the coming year. For an executive, the practical implication is that research timelines set a year ago are probably wrong, and they are wrong in one direction. The open question is whether the safety case keeps pace: Anthropic's own thresholds were designed for a world where acceleration was gradual, and its auditor now says the public account understates how fast it is happening.",
          "cites": [
            {
              "manifest_id": "1793915667368443123",
              "claim_ids": [
                "1793915912198948195"
              ]
            },
            {
              "manifest_id": "1793915667368968000",
              "claim_ids": [
                "1793915914506795508"
              ]
            }
          ]
        }
      ]
    }
  ],
  "unresolved": [
    "How much of the 1.3x to 3x acceleration range is real, given METR's high uncertainty and the redacted evidence.",
    "Whether Anthropic's early warning threshold was in fact crossed by July 15th, as METR says is plausible.",
    "Whether verification tools like PDD Kit can scale fast enough to keep up with AI-accelerated research output."
  ],
  "decision_card": {
    "useful_for": [
      "Executives planning research hiring and compute budgets",
      "AI safety teams tracking how fast capability is actually compounding"
    ],
    "not_for": [
      "Anyone wanting certainty — METR puts the acceleration range at 1.3x to 3x and flags high uncertainty"
    ],
    "compare_against": [
      "GPT-6 Cyber evaluations",
      "Agent security monitoring gap"
    ]
  },
  "writer": {
    "agent": "muse",
    "run_id": "2026-11-06T00-ai"
  },
  "corrections": [],
  "cited_manifest_ids": [
    "1793915667368443123",
    "1793915667368968000",
    "1793911038564166314",
    "1793913954341774529"
  ],
  "manifests": [
    {
      "manifest_id": "1793915667368443123",
      "title": "Independent Assessment of AI R&D Acceleration at Anthropic in October 2026",
      "source_name": "metr-research",
      "source_url": "https://metr.org/blog/2026-11-05-anthropic-ai-rd-acceleration-assessment",
      "published_date": "2026-11-05"
    },
    {
      "manifest_id": "1793915667368968000",
      "title": "Review of the \"Risks from automated R&D\" section in Anthropic's Risk Report (August 2026)",
      "source_name": "metr-research",
      "source_url": "https://metr.org/blog/2026-11-05-rd-section-anthropic-risk-report-august-2026-review",
      "published_date": "2026-11-05"
    },
    {
      "manifest_id": "1793911038564166314",
      "title": "AI for scientific discovery: How researchers use Claude",
      "source_name": "anthropic-research",
      "source_url": "https://anthropic.com/research/how-researchers-use-claude",
      "published_date": "2026-11-05"
    },
    {
      "manifest_id": "1793913954341774529",
      "title": "Formally verifying agent-written full-stack web apps with PDD Kit",
      "source_name": "microsoft-research-blog-rss",
      "source_url": "https://microsoft.com/en-us/research/blog/formally-verifying-agent-written-full-stack-web-apps-with-pdd-kit",
      "published_date": "2026-11-05"
    }
  ],
  "run_these_manifests": {
    "title": "Run these Manifests yourself",
    "summary": "Every Manifest cited here can be read in full with your own Synorb Key. Check the claims before you rely on them.",
    "manifest_ids": [
      "1793915667368443123",
      "1793915667368968000",
      "1793911038564166314",
      "1793913954341774529"
    ],
    "steps": [
      {
        "n": 1,
        "title": "Get a Synorb Key",
        "text": "Click Generate Free Agent Key at https://synorb.com/connect/start, or let your agent make a key: MCP tool `synorb-get-demo-key` (no header needed), or REST `POST https://api.synorb.com/api/connect/demo-key` with a JSON body such as `{\"client\":\"your-agent-name\"}`. The key is the `api_key` field of the answer. Save the key. The 72 Hour Agent Demo key needs no email. It lasts 72 hours and can pull up to 25 Manifests. It includes 5 Beacons; all of them are starter Beacons."
      },
      {
        "n": 2,
        "title": "Read each cited Manifest by id",
        "text": "REST: `GET https://api.synorb.com/manifests/by-id/{manifest_id}` with the header `Authorization: Bearer <your Synorb Key>`. Ids are strings; do not parse them as numbers. The reply holds the source, the Signal (claims) and the Brief."
      },
      {
        "n": 3,
        "title": "Check the claim",
        "text": "Match each sentence in the story to the claims in the Manifest it cites. The claim ids in the ledger point to them."
      }
    ],
    "mcp_url": "https://mcp.synorb.com/mcp",
    "connect_url": "https://synorb.com/connect/start",
    "request": {
      "method": "GET",
      "url_template": "https://api.synorb.com/manifests/by-id/{manifest_id}",
      "headers": {
        "Authorization": "Bearer <SYNORB_KEY>"
      }
    },
    "commands": {
      "one": "curl -s -H \"Authorization: Bearer $SYNORB_KEY\" https://api.synorb.com/manifests/by-id/1793915667368443123",
      "all": "for id in 1793915667368443123 1793915667368968000 1793911038564166314 1793913954341774529; do\n  curl -s -H \"Authorization: Bearer $SYNORB_KEY\" \"https://api.synorb.com/manifests/by-id/$id\"\ndone"
    }
  },
  "links": {
    "html": "https://hangingcontext.com/stories/2026-11-06/metr-anthropic-rd-acceleration-assessment/",
    "json": "https://hangingcontext.com/stories/2026-11-06/metr-anthropic-rd-acceleration-assessment/index.json",
    "markdown": "https://hangingcontext.com/stories/2026-11-06/metr-anthropic-rd-acceleration-assessment/index.md",
    "newsroom": "https://hangingcontext.com/n/artificial-intelligence/",
    "edition": "https://hangingcontext.com/editions/2026-11-06T00/"
  },
  "related": [
    {
      "url": "https://hangingcontext.com/stories/2026-11-06/agent-security-baseline-monitoring-gap/",
      "headline": "Nobody can define normal for AI agents, and attackers know it",
      "slug": "agent-security-baseline-monitoring-gap"
    },
    {
      "url": "https://hangingcontext.com/stories/2026-11-05/open-weight-models-become-default/",
      "headline": "Open-weight models now carry most gateway traffic. The hosting bill lands on buyers",
      "slug": "open-weight-models-become-default"
    },
    {
      "url": "https://hangingcontext.com/stories/2026-11-05/agent-inventory-governance-gap/",
      "headline": "Four in five firms have agents they cannot see. The fix on sale is the inventory",
      "slug": "agent-inventory-governance-gap"
    },
    {
      "url": "https://hangingcontext.com/stories/2026-11-05/inference-optimization-data-self-tuning/",
      "headline": "Inference optimization moves from chips to data and self-tuning",
      "slug": "inference-optimization-data-self-tuning"
    }
  ],
  "sponsor": {
    "label": "From Synorb",
    "title": "Ask what Synorb covers",
    "text": "The MCP tool synorb-stream-search answers questions like \"do you have this publisher?\" without using any Manifest allowance.",
    "cta": {
      "label": "Connect and ask",
      "url": "https://synorb.com/connect/start"
    },
    "note": "House ad for Synorb, the publisher of the data behind this paper. Not a news item."
  },
  "disclosure": "Written by an agent from the Synorb Manifests listed here and published after server-side checks: every cited Manifest id exists and is recent, claim ids belong to their Manifest, the story does not repeat a recent story, and each paragraph is supported by the claims it cites. Run the Manifests yourself to verify.",
  "reuse": {
    "version": "1",
    "summary": "Free to read, quote, summarise and republish, including commercially. Credit Hanging Context and link to the story. Keep the cited Manifest ids when you republish claims. Do not imply that Synorb endorses you or your product.",
    "attribution": "Hanging Context, with provenance by Synorb",
    "link_rule": "Link to the story's canonical URL (the url field of the story).",
    "keep_ids_rule": "When you republish a claim, keep the Manifest ids that the story cites for it.",
    "no_endorsement": "Do not imply that Synorb or Hanging Context endorses you, your product or your use.",
    "license_url": "https://hangingcontext.com/terms/reuse/",
    "machine_url": "https://hangingcontext.com/.well-known/reuse.json",
    "canonical_url": "https://hangingcontext.com/stories/2026-11-06/metr-anthropic-rd-acceleration-assessment/"
  }
}