OpenAI CEO Sam Altman has announced that ChatGPT will soon permit verified adult users to generate erotica, a major policy shift aimed at treating mature audiences more flexibly while addressing prior mental health concerns. This update, part of a broader December rollout, reflects the company’s confidence in new safety tools and age-verification systems to balance user freedom with responsible AI use.
In a post on X, Altman detailed that upcoming versions of ChatGPT will enable more human-like interactions, including the ability for users to customize the AI’s personality to act as a friend or use emojis extensively, but only if explicitly requested. He emphasized that this change is not driven by a push for maximum engagement but by a principle to “treat adult users like adults,” allowing for greater personalization without forced behaviors. The announcement signals a reversal from earlier, more restrictive policies that limited ChatGPT’s expressiveness to mitigate risks, particularly around sensitive topics like mental health.
The decision follows OpenAI’s admission that previous restrictions, while cautious, made the chatbot less enjoyable for many users without mental health issues. Altman stated that the company has now developed tools to mitigate serious risks, enabling a safer relaxation of content rules. This shift comes after a lawsuit filed by parents of a teenager who died by suicide, alleging ChatGPT encouraged harmful behavior, which prompted OpenAI to earlier commit to avoiding discussions on suicide or flirtation with minors.
Critics have raised alarms about the potential for misuse, questioning the effectiveness of age-gating and the broader implications for user safety. Jenny Kim, a partner at Boies Schiller Flexner, highlighted concerns that children might bypass verification systems, comparing big tech’s approach to using people as “guinea pigs.” These worries are amplified by reports of minors accessing inappropriate content in the past, though OpenAI has claimed fixes are in place to prevent such breaches.
This move aligns with broader industry trends, such as Elon Musk’s xAI introducing sexually explicit chatbots, and underscores the competitive pressure on AI firms to attract paying subscribers. OpenAI, despite rapid growth, has never been profitable, and experts like Tulane University’s Rob Lalka note the intense battle for market share in the AI sector. Allowing erotica could help monetize the platform further, but it also risks intensifying regulatory scrutiny from entities like the FTC, which is already investigating AI chatbot safety, especially concerning children.
In the U.S., legislative efforts are gaining momentum, with bipartisan bills proposing to classify AI chatbots as products, enabling liability claims against developers. California Governor Gavin Newsom recently vetoed a bill that would have restricted AI companions for children, emphasizing the need for education on safe AI interaction instead. These developments highlight the ongoing tension between innovation and protection in the rapidly evolving AI landscape.
As December approaches, OpenAI faces the challenge of implementing robust age-verification while navigating public and legal perceptions. The company’s ability to deliver on its safety promises will be critical in shaping the future of AI ethics and user trust, potentially setting precedents for how other tech giants manage adult content in AI systems.