Summary
- OpenAI now predicts user age from behavior and account signals, auto-restricting suspected under-18 accounts.
- Suspect accounts get prompted to verify via Persona; failing verification keeps features locked.
- Many users worry about privacy handing ID to Persona, though it encrypts data and limits retention.
If you’ve suddenly noticed ChatGPT refusing to cooperate, you’re not alone. OpenAI is paying much closer attention to certain signals coming from your account to predict the age of the user; it could see some accounts accidentally caught up in the mix.
Of course, any account OpenAI suspects is operated by someone younger than its ToC will be suspended, but given that the whole system runs on AI predictions, get set for some age-restriction teething problems while the new system beds in.
OpenAI is predicting your age now
It probably always was
OpenAI is taking a hard stance on underage accounts accessing ChatGPT, specifically taking aim at allegations that its platform doesn’t do enough to protect vulnerable people. OpenAI first mooted changes to its Model Spec in December 2025, explaining that it wanted ChatGPT to “provide a safe, age-appropriate experience for teens aged 13 to 17.”
Now, OpenAI says its updated age model profiling can better spot accounts that claim to be 18+, but in reality are teenagers who want adult access.
The model looks at a combination of behavioral and account-level signals, including how long an account has existed, typical times of day when someone is active, usage patterns over time, and a user’s stated age.
OpenAI’s new model will automatically prompt suspect accounts to verify their age through Persona. This is a secure identity verification service used on other platforms like LinkedIn and Coursera, comparing your data to known government ID checks, biometrics, and so on. Obviously, if you fail the age verification check, your account will remain restricted.
On top of that, OpenAI will automatically apply restrictions to accounts it suspects are operated by an under 18, keeping in line with its “four guiding commitments” to protect teens from dangerous content.
We’ve implemented safeguards to guide the model to take extra care when discussing higher-risk areas, including self-harm and suicide, romantic or sexualized roleplay, graphic or explicit content, dangerous activities and substances, body image and disordered eating, and requests to keep secrets about unsafe behavior.
Caught out by age verification? Just verify your identity
But that feels problematic
I’m far from the only person who feels like handing over your identity to a third party feels like a privacy problem. It’s a hard choice between verifying your ChatGPT account and trusting another service with your private data.
Persona is a well-trusted verification tool that encrypts your data at rest with AES-256, making it extremely well-protected. Furthermore, Persona doesn’t hold your data unless requested by the service provider. So, for example, Persona holds Roblox verifications for up to 30 days, while for Reddit, the number is much lower at around 7 days.
So, while it is frustrating, there is at least comfort in the fact that Persona doesn’t hold your data infinitely. And if you’re inclined to or worried OpenAI will lock down your account, you can opt for ChatGPT’s manual age verification to avoid accidental lockouts.
I regret ignoring this ChatGPT feature for so long
You don’t need prompts to fix ChatGPT. You need this setting.
Ask how old ChatGPT thinks you are
This isn’t actual advice
So, how do you check how old OpenAI thinks you are?
Of course, I turned to ChatGPT and asked, “How old do you think I am?” Not that this is anyway useful; OpenAI will use a different system to identify underage accounts, not just ask ChatGPT for your age. Still, I was actually impressed with how accurate ChatGPT was, guessing my age to the dot.
Unerring, but I appreciate that my “tone feels more ‘seasoned pro’ than ‘early-career hustle’.”
However, that doesn’t actually solve the age-restriction problem (but it is fun), but it could point to whether you come across as a very young person. As said, this isn’t actual advice, and the only way to actually unlock your account is through the verification system.
The real question is what this means for the so-called “ChatGPT Adult Mode” OpenAI made reference to towards the end of 2025—is enhanced age-verification the next step in the process? There are more than a few ChatGPT subscribers hoping so!
