The Californian company OpenAI just announced a significant change to its content policy for ChatGPT. Its CEO, Sam Altman, stated this week that in its next version of ChatGPT will open the door to messages and erotic content exclusively for verified adult usersexplaining that the company already has “new tools” that allow them to relax the restrictions that were imposed months ago on the system.
What exactly changes?
Altman explained in a post on the X social network that OpenAI, Until now, it had adopted a very restrictive policy in order to avoid risksespecially related to the mental health of vulnerable users. “We made ChatGPT quite restrictive… to be careful about mental health issues”he acknowledged. But he added: “We know that made it less useful or enjoyable for many users who didn’t have those problems.”
The novelty will bring two key elements:
- Age verification (age-gating): Altman indicated that in December the system will be deployed that will allow those over 18 years of age to be clearly distinguished, without which erotic content will be blocked.
- erotic content opt-in: Only users who expressly request it will be able to access the erotic mode. “We will allow it… but only if you request it, not because we are maximizing use”Altman stressed.
Additionally, the company announces that the chatbot will offer customization options for your “personality”, such as using emoji, friendly tone or “acting like a friend”, whenever the user wants
What are the risks and criticisms?
This turn has not been well received by everyone. Businessman and investor Mark Cuban warned that age verification may be weak and that the measure “is going to backfire.” Experts in legislation and child protection ask themselves: how do they ensure that minors do not access erotic content? Boies Schiller Flexner partner Jenny Kim put the question clearly: “How are you going to make sure that children can’t access the adult-only sections?”
At the regulatory level, states like California already have passed laws requiring AI companies to implement stricter controls about chatbots “buddies” and adult content.
What happens since December?
Although the changes are announced for December 2025, there are still unanswered questions: What level will the age verification be? What definitions will be used for “erotic”? Will there be additional human control? What happens with free versions versus subscription versions? OpenAI has not provided exhaustive details at the moment.
What is clear is that the company is moving towards a model in which adults have more freedom under verified conditionswhile maintaining the rhetoric of protecting the most vulnerable.
For adult users, this opens up new possibilities: more personalized conversations, less institutional tone, access to content that until now was blocked. But it also comes with responsibility: Platforms must ensure that age limits work and that safety protocols remain robust.
The measure highlights a key point in the evolution of conversational AI: the balance between freedom of expression, user personalization and protection of the weakest. In this area, OpenAI seems to have decided to bet on freedom under verification, while opening the door to a new chapter of the chatbot: that of adult content. We will have to see if this bet works, and how society, regulators and the users themselves respond.