OpenAI CEO Sam Altman has shared a plan to implement stringent age verification, upgrade parental control features, and offer age-appropriate conversations on ChatGPT. This came a few hours before the Senate Judiciary Committee hearing on the potential risks of AI chatbots to minors. Altman’s plan is a direct response to an ongoing lawsuit alleging that OpenAI helped a teenager commit suicide.
Sam Altman provided insight into OpenAI's decisions through a blog post and statements on social media. He highlighted the intricacy of balancing three key ingredients: freedom, privacy, and safety in the context of AI products, especially for young users.
"We prioritise safety ahead of privacy and freedom for teens; this is a new and powerful technology, and we believe minors need significant protection," the OpenAI CEO stated.
"I don’t expect that everyone will agree with these tradeoffs, but given the conflict, it is important to explain our decision-making," he wrote on X.
The company has proposed an age-prediction system to automatically assign users to separate ChatGPT versions: one for those aged 13 to 17, and another for adults 18 and over.
If there’s any doubt regarding the user's age, OpenAI will take precautions immediately, as Altman explained, "If there is doubt, we’ll play it safe and default to the under-18 experience." This reflects OpenAI's conservative approach to youth safety.
In specific cases or jurisdictions, OpenAI may ask users to provide identification as an additional precaution. Altman explained, "In some cases or countries, we may also ask for an ID; we know this is a privacy compromise for adults, but believe it is a worthy tradeoff."
These OpenAI teen safety upgrades aim to provide age-appropriate content to both adults and minors, even if it causes some loss of privacy on the platform.
The OpenAI CEO acknowledged the challenge of balancing the sensitive nature of interactions on the platform following recent allegations of acting as a ‘suicide coach’ for a teenager.
Sam Altman clarified, saying, "By default, should not provide instructions about how to commit suicide, but if an adult user is asking for help writing a fictional story that depicts a suicide, the model should help with that request." His comments highlight OpenAI's initiative to differentiate harmful guidance from legitimate creative works.
OpenAI has already developed new protocols for responding to users flagged as being at risk of self-harm. In a blog post, OpenAI mentioned it "will attempt to contact the users’ parents and if unable, will contact the authorities in case of imminent harm."
With new parental controls scheduled to be released at the end of the month, guardians can customise ChatGPT's responses for their children. They can also set preferences for memory and blackout hours.
Beyond updated safety measures, what received attention from the analysts is the timing of the announcement. OpenAI’s initiative has been under public scrutiny for a long time. The alleged act of helping teens find suitable locations for suicide compelled the tech giant to implement digital safety features under huge legal pressure.
Since its launch, OpenAI has always insisted that ChatGPT is not intended for children aged below 12. Such direct safeguards were also not implemented earlier by the tech giant. The new safety features will help parents track how their children engage with the AI-powered chatbot.
OpenAI's safety measures reflect the ongoing industry trend of prioritising user safety, especially for vulnerable teenagers who are misusing advanced AI models. By introducing a new set of secure interaction regulations, OpenAI aims to address the ongoing technical and societal concerns.