OpenAI has announced that in December it will lift restrictions on erotic content on ChatGPT for verified adult users. The measure comes after months of complaints about the chatbot’s loss of “personality”, especially after the arrival of the serious GPT-5and represents a 180-degree turn in the company’s strategy, until now reluctant in contrast to Grok.
Why is it important. This is the moment when OpenAI recognizes that without emotional (and sexual) intimacy it cannot compete with platforms like Character AIwhere its users spend up to two hours a day talking to their AI-partner.
Erotic literature has existed since writing has existed. ChatGPT does not invent anything. It simply bridges the last gap between “useful tool” and “total emotional companion.”
The context. Sam Altman had declared in August that he was “proud” not to have turned ChatGPT into a sexbot. Now he justifies the change under the principle of “treating adults like adults.”
The reality is more prosaic: after supposedly mitigating mental health problems (two months after a lawsuit for the suicide of a teenager who used the platform), OpenAI believes that it can now afford to relax controls.
The money trail. Character AI proved that erotic is a great glue to retain users. If OpenAI wants to monetize the engagement Really, you need to enter that field.
Personalization of the assistant (with options for more human responses, use of emojis or “friend” behavior) is just the wrapper. Adult content is the new product.
Yes, but. The toll to pay is something unprecedented: OpenAI will require age verification, presumably with an identity document. It is the largest exchange of privacy for service that such a technological platform has asked of us.
The question is not whether there will be leaks of databases with erotic conversations linked to real, verified identities. The question is when and how many millions of users will be affected.
The turn. OpenAI is building the metaverse that Meta couldn’t create, or at least not successfully. Only this is not visual, but conversational. Meta failed because no one wanted to be in its virtual worlds. But we do want to be in ChatGPT. And more with the restriction-free mode for emotional companionship and eroticism.
The summer’s stricter restrictions (designed to make the chatbot “less fawning” and prevent mental health crises) had pissed off users who didn’t have psychological problems. Now OpenAI reverses its own security philosophy in record time. You have introduced parental controls and a separate experience for minors, but the speed of change raises questions about whether they have truly “mitigated” the risks or simply decided to take them on.
Between the lines. This move shows the real battle of conversational AI. It’s not about who has the most powerful model, but who gets you to spend the most time with it. And accompaniment without an erotic dimension is incomplete for many users. OpenAI knows this.
Altman predicted that ChatGPT could “cure cancer one day.” Now bet that he can also be your sexual confidant. They are only two sides of the same strategy of total penetration in the lives of users.
In Xataka | Character.AI is accompanying and making its users fall in love. That’s wonderful until it’s not.
Featured image | Xataka
GIPHY App Key not set. Please check settings