OpenAI today announced a suite of safety measures for teenage users of ChatGPT, including a technology to better predict a user’s age and forthcoming parental controls, as the company faces scrutiny over the chatbot’s impact on minors.

The company says it is developing a technology that estimates whether a user is under 18. When OpenAI identifies that a user is a minor, they will automatically be directed to an age-appropriate ChatGPT experience that blocks graphic and sexual content and can involve law enforcement in rare cases of acute distress. OpenAI also plans to offer parental controls by the end of the month, allowing parents to link their account to a teen’s via email, set blackout hours, manage which features to disable, and receive notifications if the teen is in acute distress.

Sam Altman, OpenAI’s CEO, wrote in a blog post that “we prioritize safety ahead of privacy and freedom for teens; this is a new and powerful technology, and we believe minors need significant protection.” The announcement comes after the Federal Trade Commission recently launched an inquiry into several tech companies, including OpenAI, over how AI chatbots like ChatGPT potentially negatively affect children and teenagers, and after a lawsuit from a family that blamed the chatbot for their teenage son’s death by suicide. In the same post, Altman said, “These are difficult decisions, but after talking with experts, this is what we think is best and want to be transparent in our intentions.”

It is a striking reordering of priorities for a company that has long sold itself on user freedom — and, on teen safety at least, a bet that caution is the position to be caught defending.