ANTHROPIC’S BENCHMARK PROGRESS SPED UP

METR assessment · 5 November 2026 · As of 7 November 2026

1.9× THE EARLIER PACE
AS OF 7 NOVEMBER 2026 · DATED SOURCES
Transcript

Full transcript

The narration below matches the explainer. Its original reporting dates and qualifications still apply.

Read the companion article

ANTHROPIC’S BENCHMARK PROGRESS SPED UP

Bar chart: Anthropic’s benchmark progress sped up — 1.9× the earlier pace. METR estimates that Anthropic’s AI benchmark scores improved one point nine times as fast after October twenty twenty-five, compared with the earlier trend.
Anthropic’s benchmark progress sped up — 1.9× the earlier pace METR assessment · 2026-11-05

METR estimates that Anthropic’s AI benchmark scores improved one point nine times as fast after October twenty twenty-five, compared with the earlier trend.

METR assessment · 2026-11-05

THE SPEEDUP ESTIMATE HAS A WIDE RANGE

Bar chart: The speedup estimate has a wide range — 1.5× to 2.8×. The confidence interval runs from one point five to two point eight times the earlier pace.
The speedup estimate has a wide range — 1.5× to 2.8× METR assessment · 2026-11-05

The confidence interval runs from one point five to two point eight times the earlier pace. That range describes uncertainty in the benchmark estimate.

METR assessment · 2026-11-05

METR POINTS TO AI ASSISTANCE

Flow diagram: METR points to AI assistance — AI helped researchers. Staff interviews and surveys suggest that AI assistance sped up research.
METR points to AI assistance — AI helped researchers METR assessment · 2026-11-05

Staff interviews and surveys suggest that AI assistance sped up research. METR thinks it likely explains most of the recent acceleration, with substantial uncertainty about its share.

METR assessment · 2026-11-05

THE PUBLIC SAW ONLY PART OF THE EVIDENCE

Flow diagram: The public saw only part of the evidence — Limited public evidence. METR had access to internal benchmarks and researchers.
The public saw only part of the evidence — Limited public evidence METR assessment · 2026-11-05

METR had access to internal benchmarks and researchers. Anthropic redacted most of the evidence METR considered most informative, limiting what outside readers can assess.

METR assessment · 2026-11-05

A FASTER BENCHMARK TREND LEAVES OPEN QUESTIONS

Scenario diagram: A faster benchmark trend leaves open questions — Faster gains · Uncertain cause. The benchmark results point to faster improvement on measured tasks.
A faster benchmark trend leaves open questions — Faster gains · Uncertain cause METR assessment · 2026-11-05

The benchmark results point to faster improvement on measured tasks. How much AI contributed and how far the finding generalizes remain open questions.

METR assessment · 2026-11-05

SOURCES AND QUALIFICATIONS

METR published the assessment and its statistical range. The Associated Press reported on the findings and Anthropic’s response.

METR assessment · 2026-11-05

Source links

All Explainers
[00:00]
Read the full article

METR Finds Anthropic’s AI Benchmark Progress Sped Up 1.9×

As of 7 November 2026. The 1.9× estimate compares rates of benchmark-score improvement on Anthropic’s version of the Epoch Capabilities Index (AECI). METR’s judgment about AI’s contribution is more uncertain, and Anthropic redacted much of the most informative underlying evidence.

METR assessment. 2026-11-05. Primary report with the benchmark estimate, confidence interval, evidence on AI assistance and details of what Anthropic redacted.

Associated Press. 2026-11-05. Independent coverage of METR’s findings, how it obtained internal access and Anthropic’s response.