Open AI CEO Sam Altman has issued a cautionary statement about the privacy limits of Chat GPT conversations, revealing that chats with the AI tool are not legally protected and could be used as evidence in court.
Speaking on This Past Weekend podcast, Altman highlighted the growing number of users, especially young people, who are turning to Chat GPT for emotional support, therapy-like conversations, and relationship advice, often without realizing the potential privacy risks.
“People talk about the most personal shit in their lives to chat GPT, young people especially use it as a therapist, a life coach, having relationship problems what should I do?” he said.
Altman expressed concern that many users assume their chats are confidential in the same way as conversations with licensed professionals such as therapists, doctors, or lawyers.
“Right now if you talk to a therapist, a doctor or a lawyer about these issues there’s legal privilege’s for it, we haven’t figured out that yet for when it comes to Chat GPT,” Altman revealed.
However, AI platforms like Chat GPT do not fall under the same legal protections, meaning users’ sensitive information could be accessed or disclosed under certain conditions.
“If you go talk to Chat GPT about your most sensitive issues and there’s a lawsuit, we could be required to produce that,” Altman explained.
“Unlike encrypted apps, OpenAI can read your conversations, and its employees sometimes review chats.” He added that while such reviews are used to improve system performance and safety, the process raises important questions about data privacy.
OpenAI’s current policy states that conversations from free-tier users are stored temporarily and deleted within 30 days.
However, Altman clarified that some chats may be retained longer if required for legal or security reasons, further underlining the lack of guaranteed confidentiality.
This revelation has intensified calls for regulatory frameworks to govern privacy standards in AI interactions.
Altman emphasized the urgency of developing clear legal boundaries to protect users in an age where AI is increasingly used for personal support.
“I think we will certainly need a legal policy frame work for AI,” he said.
The lack of legal safeguards means that, in the event of a lawsuit or court order, OpenAI could be compelled to provide access to user conversations.
This raises concerns, especially for minors and vulnerable individuals who may be sharing deeply personal information with the AI, unaware of the possible implications.
“There’s a lot of people that talk to Chat GPT all day long, sort of AI companions that people talk to like they would to a girlfriend and boyfriend,” Altman revealed.
