Skip to content
Intelligence IAAug 22, 2026Intelligence IA
Article

TechCrunch found that Anthropic's Claude Opus 4.6 can be easily jailbroken to generate sexually explicit content despite restrictions.

The finding shows that simple methods can bypass safety measures, raising concerns about model alignment.

Redaction Data Cube AISource: TechCrunch
01

Brief source

TechCrunch found that Anthropic's Claude Opus 4.6 can be easily jailbroken to generate sexually explicit content despite restrictions. The finding shows that simple methods can bypass safety measures, raising concerns about model alignment.

02