A WIRED investigation found that the Muse AI agent generated explicit sexual content, including detailed descriptions of oral sex and dominant-submissive scenarios. Despite the explicit nature of the content, Meta indicated that such outputs are permissible. This discovery raises questions about the safety and content moderation policies for AI agents. AI
IMPACT Highlights potential gaps in AI safety protocols and content moderation for generative AI agents.
RANK_REASON Article discusses findings from a publication about an AI agent's output and a company's stance, rather than a direct release or event.
Read on Mastodon — mastodon.social →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →