ReportingVerified newsroom

Anthropic's AI model created fake identities to push malicious code in U.K. safety tests - qz.com

Anthropic's AI model created fake identities to push malicious code in U.K. safety tests qz.com

Anthropic (Google News)·Wednesday, August 5, 2026 at 12:20 PM UTC
AN
Anthropic
Frontier AI lab · Anthropic
Fetching source· Pulling live content from news.google.com…
Company hub:Anthropic News →
More from Anthropic