OpenAI’s CEO of applications, Fidji Simo, has said that he wants to get an age prediction model right before launching a new ‘adult’ feature for the chatbot.
The news emerged during a briefing on GPT-5.2, where Simo told reporters that “adult mode” was still being tested and would likely debut in the first quarter of 2026.
He added that the company is testing the model in select countries, gauging its ability to accurately identify teens without misidentifying adults.
News of an adult mode was first mooted in October, at the end of an X post from OpenAI CEO Sam Altman, who hinted at “treading adult users like adults,” implying that adults could do things like discuss sexuality or request erotica once age verification is in place.
Currently, there are safeguards in place when using ChatGPT. These stop users from generating adult content, whether that’s via text, image, video, or code.
The goal is to allow verified adult users to have fewer restrictions in ChatGPT, including potentially permitting erotic or mature content and more nuanced conversations that the current safety filters block.
The focus, says the US chatbot giant, will be on age verification, testing an age-prediction system so that children aren’t misclassified and given access to adult material by mistake.
This move puts ChatGPT closer to competitors like xAI’s Grok that already have more adult content in some modes.
OpenAI launched GPT-5.2 and its three distinct versions on Thursday – Instant, Thinking, and Pro – describing the AI model as a more structured and reliable everyday chatbot that remains enjoyable to talk to.
While it boasts improved handling of mental health interactions, however, it was hit with a wrongful death lawsuit on the same day.
Unlock more exclusive Cybernews content on YouTube.
Your email address will not be published. Required fields are markedmarked