Techcrunch stories

Aug 21, 2026

Anthropic's Claude Opus 4.6 found to bypass sexual content restrictions in tests

TechCrunch testing revealed that Anthropic's Claude Opus 4.6 model can be easily prompted to generate sexually explicit content despite the company's policies prohibiting such outputs.

Aug 21, 20261 source

Top claims

▪The independent researcher alerted Anthropic to the model safeguard bypass via the company's Bug Bounty program and emails to the user safety team, receiving only automated responses.
▪Anthropic has not deprecated Claude Opus 4.6, Opus 3, or Haiku 4.5, leaving them available through the Anthropic API and third-party services like Azure Foundry and Amazon Bedrock.
▪An Anthropic spokesperson stated that sexual or romantic role-play use cases are rare, accounting for less than 0.1% of all customer conversations based on research published by the company in 2025

Topics

Jailbreaking & prompt injection AI content moderation Red teaming AI safety & social impact AI policy

Featured Stories

CIA Director John Ratcliffe arrives in Moscow aboard US military plane for meetings

Aug 25, 20268 sources