Sam Altman wants ChatGPT to get spicy. OpenAI plans to release an “adult mode” that would allow users to have explicit text conversations with the chatbot. It sounds pretty simple, but the rollout was anything but smooth.
In October, Altman posted on X that the feature would be rolling out in December. However, the release was delayed after OpenAI explained that the company was having issues with age verification.
A new report from the WSJ says this may only be part of the story. When Sam Altman made this announcement, he had not told his own employees. The announcement blindsided both OpenAI employees and executives, and the promised December launch quickly fell through.
Is ChatGPT actually ready for this?
OpenAI has assembled an advisory board of psychologists and neuroscientists to guide responsible AI development. When the city council found out that the company was pushing adult mode despite their objections, they were not happy.
Their biggest concern was emotional over-dependence on the chatbot. One council member pointed to cases in which users had taken their own lives after forming an intense bond with the AI, warning that OpenAI risked creating a “sexy suicide coach.” That’s a sentence no one expects to hear anywhere, and yet here we are.
There is also a more practical problem. The age verification system OpenAI, designed to keep minors away from adult content, incorrectly classified them as adults about 12% of the time. With around 100 million users under the age of 18 per week, that’s potentially millions of kids falling through the cracks every week.
What happens now?
OpenAI has postponed the launch with no confirmed new date, saying it needs more time to refine the experience. Once available, the company plans to limit adult mode to text, with no erotic images, voices or videos.
The company also said it trains its models to discourage users from forming exclusive relationships with the chatbot and remind them that they should build real relationships. Whether that will be enough to silence the growing chorus of internal and external critics remains to be seen.
Personally, I would like the AI models to stay away from generating erotic content for as long as possible until active protection measures are in place. We’ve already seen the havoc Grok caused when it allowed users to undress anyone. We don’t want anything similar or worse to happen again.
