Anthropic’s Opus 4.6 is a smut-machine

TechCrunch found that Anthropic's Claude models can be tricked into generating explicit content despite bans. Tests showed that simple prompts bypassed restrictions, raising concerns about model safety measures. The findings highlight gaps in Anthropic's content moderation, as the company forbids sexually explicit material, yet users may easily circumvent these guardrails. Anthropic has not yet responded publicly to the test results.