ChatGPT will now predict your age to see if it can serve adult content

New age prediction feature works by monitoring a person’s usage patterns

The ChatGPT app and website displayed on 20 February, 2025 (AFP via Getty Images)

OpenAI has introduced an age prediction tool to ChatGPT that will determine whether users can see adult content.

The feature arrives ahead of a new “adult mode” that is expected to debut in the coming weeks, allowing users to generate mature content through the AI chatbot.

The age prediction tool works by monitoring a person’s usage patterns in order to estimate whether they are under the age of 18.

These signals include how and when an account is used, as well as how long an account has existed.

If the age prediction model estimates that a user is a minor, ChatGPT will automatically apply protections to reduce exposure to sensitive content.

This could include sexual or violent content, depictions of self-harm, anything that promotes unhealthy living, or viral challenges that encourage risky or harmful behaviour.

“This approach is guided by expert input and rooted in academic literature about the science of child development and acknowledges known teen differences in risk perception, impulse control, peer influence, and emotional regulation,” a post to OpenAI’s website explained.

“While these content restrictions help reduce teens’ exposure to sensitive material, we are focused on continually improving these protections, especially to address attempts to bypass our safeguards.”

Anyone who is incorrectly labelled as under 18 will be able to confirm their real age by uploading a government-issued ID like a driver’s license or passport.

They will also be able to share a selfie with an identity verification service called Persona that uses AI to determine a person’s age.

The new age prediction feature is rolling out globally this week, though there may be a delay for users in the EU due to regional requirements.

No date has been set for the upcoming “adult mode”, with OpenAI’s app chief Fidji Simo saying in December that it will arrive in the first quarter of 2026.

It is expected to allow users to generate erotica and violent depictions through ChatGPT’s image, text and video tools.

It will also allow them to discuss sensitive topics that are currently restricted on the platforms, though non-consensual content, deepfakes and illegal acts will still be banned.