Skip to content
AI News HubLIVE
More
Summary only

Anthropic's Mythos is evolving faster than expected, reports AI safety agency

Only a month after its initial release, Anthropic's storied Mythos model is breaking new testing boundaries.

SourceZDNet AI
Anthropic's Mythos is evolving faster than expected, reports AI safety agency
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions

Summary only

Under the source content permission, this page provides an editorial summary, key points, and a link to the original article.

Key points and analysis

Article intelligence

EngineersIntermediate

Key points

  • Anthropic's Claude Mythos model has demonstrated new capabilities just one month after its initial release, according to UK AISI testing.
  • The newer Mythos Preview checkpoint completed two cyber ranges, including the previously unsolved 'Cooling Tower' for the first time.
  • AI capabilities in cyber tasks are accelerating, with task length doubling every 4.7 months, faster than previous estimates.
  • Testing limitations, including a 2.5M token cap, may understate the true capabilities of frontier models like Mythos.

Highlights and analysis are generated automatically and may contain errors. Check the original source.