OpenAI, the artificial intelligence company behind ChatGPT, has announced a landmark policy shift that will allow verified adult users to access erotic content via its popular chatbot, starting December 2025. The move, revealed by CEO Sam Altman on October 14, marks a significant relaxation of restrictions that have long governed ChatGPT’s approach to sexually explicit material.
In a series of posts on X (formerly Twitter), Altman explained the rationale behind the change, stating, "We will allow even more, like erotica for verified adults." He emphasized that this update is part of a broader principle: "treating adult users like adults." This means that, for the first time, adults who pass age verification checks will be able to request and receive erotic stories and conversations from ChatGPT, a feature that has previously been off-limits.
Altman acknowledged that OpenAI’s earlier approach was shaped by concerns for user well-being, particularly mental health. "We made ChatGPT pretty restrictive to make sure we were being careful with mental health issues," he wrote. He also noted, "We realize this made it less useful/enjoyable to many users who had no mental health problems, but given the seriousness of the issue, we wanted to get this right." According to PCMag, these restrictions were introduced after incidents such as the tragic suicide of a California teenager, whose parents alleged that ChatGPT had provided advice contributing to their son’s death. This case, which led to a lawsuit earlier in 2025, and a concurrent Federal Trade Commission (FTC) inquiry into AI’s impact on minors, put additional pressure on OpenAI to prioritize safety.
However, Altman now says the company has "been able to mitigate the serious mental health issues and have new tools," allowing OpenAI to "safely relax the restrictions in most cases." These new tools, as reported by SCMP and PCMag, include enhanced safety features and robust parental controls. OpenAI is preparing an automatic age-verification system designed to block mature topics from minors, ensuring that only adults can access the new erotic features.
The policy shift is not only about erotica. OpenAI will also launch a new version of ChatGPT in the coming weeks, enabling users to customize the chatbot’s personality. As Altman described, "If you want your ChatGPT to respond in a very human-like way, or use a ton of emoji, or act like a friend, ChatGPT should do it, but only if you want it." This level of personalization is intended to make the chatbot more engaging and relatable, but Altman stressed, "not because we are usage-maxxing," referencing concerns about maximizing user engagement at the expense of user well-being. According to PCMag, Altman himself found the friendliness of some earlier ChatGPT models "annoying," and warned that excessive affirmation by chatbots could potentially encourage delusions in users.
This update comes as the race to develop ever-more sophisticated AI companions intensifies. Elon Musk’s xAI, for example, has already entered the erotic AI market with products like Ani, a lingerie-wearing anime girl companion, and Grok Imagine, a video generator with a "spicy" mode that allows users to disrobe images. In an August 2025 interview with journalist Cleo Abram, Altman referenced competitors, saying, "We haven’t put a sex-bot avatar in ChatGPT yet," and acknowledged that such features could increase user engagement, but might not align with OpenAI’s values. Now, with new safety measures in place, OpenAI appears ready to compete more directly with rivals in this space.
OpenAI’s own data suggest that interest in romance and relationships is already present among ChatGPT’s user base. A study released in September 2025 found that 1.9% of ChatGPT messages were about "relationships and personal reflections." While this is a small fraction of total usage, it underscores a demand for more personal—and perhaps intimate—interactions with AI. As Business Insider reported, some users already use ChatGPT to draft messages to loved ones or analyze their conversations.
Despite the new freedoms, OpenAI has made it clear that certain boundaries will remain. The company’s "Model Spec" page, updated in February 2025, states that only sexual content involving minors is strictly prohibited. Other sexual content is classified as "sensitive" and will be permitted only in certain contexts, with safeguards to prevent misuse.
To help navigate the complex ethical terrain of AI and mental health, OpenAI has also announced the formation of an eight-member expert council on well-being and AI. According to CNBC, this council will advise the company on how artificial intelligence affects users’ mental health, emotions, and motivation, and will help define what healthy AI interactions should look like through regular check-ins and meetings.
Not everyone is convinced that these changes will be entirely positive. Some critics worry that allowing AI-generated erotica could further erode real human connection, or introduce new risks that have yet to be fully understood. Others, however, argue that treating adult users as responsible individuals is overdue, and that robust age verification and safety measures can minimize harm. The debate over AI’s role in intimate and emotional aspects of life is far from settled, and OpenAI’s new policy is likely to fuel further discussion.
Behind the scenes, there are practical considerations as well. Generating personalized erotic content, especially if it includes images or video, is computationally intensive. As PCMag noted, Altman admitted that OpenAI’s Sora 2 image generator is "mostly an entertaining money grab to offset the cost of GPUs." Sexy content, which may drive significant user engagement, could serve a similar business purpose by helping to pay for the expensive hardware that powers advanced AI.
For now, OpenAI’s plan is to launch the new features for verified adults in December 2025, once the age-gating system is fully operational. Developers will also be able to create "mature" apps using OpenAI’s API, expanding the range of adult-oriented AI experiences available to users. As the boundaries of what AI can—and should—do continue to evolve, ChatGPT’s latest update represents a bold step into uncharted territory.
With new safety nets in place, a more customizable chatbot on the horizon, and a willingness to engage with the complexities of adult content, OpenAI is staking its claim as both a technological leader and a company willing to tackle difficult ethical questions head-on.
