OpenAI rolled out a quiet however vital security improve on Tuesday, Jan 20, within the US. This comes within the type of an age-prediction system designed to determine when ChatGPT customers are probably below 18.
The system displays the rising concern amongst policymakers, mother and father, and educators over the misuse of generative AI by minors and the shortage of adequate safeguards.
Teen security is on the forefront of conversations at present, particularly as generative AI is getting used to create sexualized photographs and as a way for physique shaming; some youth are additionally utilizing chatbots as “friends,” main some mother and father and guardians to allege OpenAI is dangerous for younger individuals.
Notably, Australia has already banned social media use for kids below 16, and several other US states are curbing cellphone use in faculties.
In August 2025, Matt and Maria Raine, mother and father of a 16-year-old boy who died by suicide, filed a lawsuit (warning: the lawsuit incorporates descriptions of self-harm) in opposition to OpenAI, claiming that ChatGPT helped their son Adam discover strategies of suicide and even helped him write a suicide observe.
OpenAI has refuted the claims and alleges misuse of AI’s capabilities, whereas saying improved security measures.
ChatGPT has over 700 million weekly energetic customers.
Shutterstock
Extra not too long ago, Elon Musk’s Grok chatbot was within the information after it generated 4.4 million photographs. Of which it publicly shared roughly 1.8 million sexualized photographs of girls, the New York Occasions reported.
In a sequence of enhancements to make ChatGPT safer for minor customers, it is a relatively sensible one.
In response to OpenAI’s current weblog submit, ChatGPT, relatively than relying solely on self-reported ages, which has typically been some extent of rivalry for social media firms, will use behavioral and account-level indicators, comparable to utilization patterns and interplay sorts, to estimate whether or not a consumer is a minor.
Moreover, it’s going to monitor the time of day when the consumer is energetic on-line, together with the consumer’s reported age, to determine the consumer’s age.
What modifications for the under-18 customers?
If ChatGPT finds {that a} consumer is below 18, it’s going to robotically apply protections to stop publicity to dangerous content material, in accordance with its security pointers for teenagers.
This contains censoring customers from publicity to the next content material:
- Sexual, romantic, or violent function play
- Depictions of self-harm
- Content material selling excessive magnificence requirements, unhealthy weight-reduction plan, or physique shaming
- Viral challenges that encourage dangerous behaviors in minors
OpenAI has been vocal about introducing security protocols to guard minors. In September 2025, Sam Altman revealed “Teen safety, freedom, and privacy,” the place Altman talked concerning the age prediction system.
Extra Tech Shares:
- Morgan Stanley units jaw-dropping Micron value goal after occasion
- Nvidia’s China chip downside isn’t what most buyers suppose
- Quantum Computing makes $110 million transfer no one noticed coming
- Morgan Stanley drops eye-popping Broadcom value goal
- Apple analyst units daring inventory goal for 2026
As a part of the brand new security pointers, OpenAI additionally launched parental controls to boost safety and guarantee cautious monitoring by adults. Mother and father can set quiet hours when minors can not use ChatGPT. They will additionally personalize how mannequin coaching works and obtain alerts “if signs of acute distress are detected” by the system.
For customers who had been mistakenly positioned within the under-18 expertise, there’s a easy choice to reverse it. ChatGPT makes use of Persona, a safe third-party identification verification service, which permits customers to replace a selfie and restore their full entry.
Within the European Union, this characteristic will roll out within the upcoming weeks resulting from regulatory necessities.
The security replace arrives amid renewed criticism from Elon Musk, who has accused OpenAI of abandoning its authentic mission to stay solely a non-profit. Altman has brazenly dismissed the arguments as “cherry-picked” by Musk.
OpenAI argues that the present construction and ongoing modifications to its enterprise mannequin are essential to stability entry, security, and scalability.
Associated: Trump’s Davos message hints at a pricey shift for shoppers
