Industry

METR audit says Anthropic may have crossed its own AI acceleration warning threshold as early as July

Original title:METR finds Anthropic may have crossed its own AI acceleration warning threshold back in July

AI summary

METR released the results of an eleven-week audit of Anthropic's internal data, saying that since October 2025 the rate of improvement in model test scores has accelerated by about 1.9x, with roughly 80% attributable to AI being used in the company's own research, and that it may have reached the warning threshold in its responsible scaling policy on July 15.

Key Takeaways

  • METR released the results of an eleven-week audit of Anthropic's internal data
  • saying that since October 2025 the rate of improvement in model test scores has accelerated by…
  • 9x,with roughly 80% attributable to AI being used in the company's own research,and that it may…

Quick Facts

Published
Source
The Decoder:AI News(RSS)
Category
Industry
Topic
Anthropic / ClaudeIndustry NewsSafety & Alignment
AI score
83
Read the original Read the full article at the original source
Why it matters

原文给出了 METR 审计的核心数字和被删除证据等细节,读者可以据此了解 Anthropic 自身加速预警阈值的争议。

Topics

Related entities

Source & copyright

This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.

Original article · AIHOT entry

How to Cite

When citing this article, keep the title, date, and permalink, and credit Orange News as the source.

Orange News: METR audit says Anthropic may have crossed its own AI acceleration warning threshold as early as July. November 6, 2026. https://new.nat6.net/en/news/vpopkx98s6mpp9coxhi7hhvh2/

View all →