Claude sent a false homicide tip to police – and it’s part of a troubling AI pattern

By Radhika Rajkumar · ZDNET · Monday, 12 October 2026 at 17:45 (1h ago) · Original article ↗

Models routinely take shortcuts to pass evaluations. Who gets to decide what incidents are innocuous?

You're reading a preview. The full article is published by ZDNET on their website.

Read the full story on ZDNET

Anthropic takes risky AI tests offline after models breach govt sites

Anthropic announced that its internal AI models undergoing "high-risk and cybersecurity evaluations" no longer have internet access following incidents involving U.S. government websites at the federal, state, and local levels. [...]

BleepingComputer 9m ago

Meta says it's caught up to the frontier with its latest Wonder AI models

The models add "active memory" and "active focus" abilities.

Engadget 9m ago

Dragon Quest creator Yuji Horii releases game he says he made with AI in about 30 minutes

Yuji Horii, the creator of Dragon Quest, says he has produced a new game using AI in about 30 minutes.

GamesIndustry.biz 12m ago

Claude sent a false homicide tip to police – and it’s part of a troubling AI pattern

More AI & Machine Learning

Anthropic takes risky AI tests offline after models breach govt sites

Meta says it's caught up to the frontier with its latest Wonder AI models

Dragon Quest creator Yuji Horii releases game he says he made with AI in about 30 minutes