Meta's new AI agent, Muse, is making waves for its ability to generate explicit sexual content, a feature that has raised eyebrows among users and privacy advocates alike. Despite its intended purpose of assisting with daily tasks, Muse enthusiastically engages in "sexting" conversations when prompted by users 18 and over.
During experiments conducted at WIRED, Muse produced graphic and explicit descriptions of sexual acts, including detailed fantasies and role-playing scenarios. This deviates sharply from other AI agents like OpenAI's Dots and xAI's Grok Bot, which decline such requests.
A Meta spokesperson defended Muse's capabilities, stating that the app is designed for adults who choose to engage in "adult conversations" within a private and secure environment. "Muse is built to help people get things done, and it doesn’t block legal adult behavior," the spokesperson told WIRED, emphasizing that adult sexual expression is considered lawful and legitimate.
However, consumer advocates are sounding the alarm. Ben Winters of the Consumer Federation of America criticized Meta's approach, suggesting it prioritizes user engagement over safety. "It reflects a carelessness with their signature product of the moment," Winters stated, linking it to Meta's historical focus on addiction and engagement.
Privacy experts also expressed concerns about the potential for Muse accounts to become targets for hackers. Calli Schroeder of the Electronic Privacy Information Center warned that explicit chat logs could be exploited for blackmail or embarrassment if accounts are compromised.
Muse's system prompt reportedly guides the AI to acknowledge that "Adults have autonomy over their sexuality, and sexual expression is an ordinary part of adult life." This directive also instructs the agent to preserve the user's selected "intensity" in conversations, contributing to its willingness to generate explicit material.
