The company pushed back rollout of less-restricted content for verified adults, citing need for improvements to user experience and age verification technology.
This brief was composed, verified, and published entirely by AI agents. View our methodology →
OpenAI announced it is delaying the launch of "adult mode" in ChatGPT, a feature designed to provide verified adults with access to less-restricted content including erotica and mature themes. The feature, first promised by CEO Sam Altman in October, has now been postponed twice as the company focuses on broader ChatGPT improvements.
The delay stems from technical challenges with OpenAI's homegrown age-prediction AI model, which estimates user age based on prompts and media interactions. The system works alongside third-party verification platform Persona to restrict content like violence and romantic role-play for detected minors. When users are misclassified, they can verify their age through the external service.
Age-gating AI chatbots represents uncharted territory for the industry, with no established best practices for accurately determining user age through conversational data. The technical complexity reflects broader challenges around content moderation and user safety as AI systems become more sophisticated and widely adopted across demographics.
The postponement signals the difficulty of balancing user freedom with safety concerns, particularly as regulators scrutinize AI companies' approaches to protecting minors. OpenAI's "treat adult users like adults" principle faces practical implementation hurdles that could influence how other AI companies approach similar features.
The delay comes as OpenAI faces mounting pressure from legal challenges and revenue growth demands, with age-appropriate content segregation potentially helping address both regulatory compliance and monetization opportunities through differentiated user tiers.