Sam Altman Warns of ChatGPT Privacy Risks

Share


OpenAI CEO Sam Altman has cautioned users about the lack of legal protections surrounding conversations with AI tools like ChatGPT. Speaking on the This Past Weekend podcast hosted by comedian Theo Von, Altman expressed concern that deeply personal exchanges with AI systems are not currently safeguarded under established legal privileges, such as doctor-patient or attorney-client confidentiality.

“People talk about the most personal things in their lives to ChatGPT,” Altman said. “Young people, especially, use it as a therapist or life coach. But unlike real therapists or doctors, there’s no legal privilege in place to protect those conversations.”

Altman warned that in the event of legal proceedings, these conversations could potentially be disclosed. “If you go talk to ChatGPT about your most sensitive stuff and there’s a lawsuit or whatever, we could be required to produce that,” he noted. “And I think that’s very screwed up.”

The remarks come amid growing public reliance on AI tools for emotional support, mental health inquiries, and relationship advice—areas traditionally covered by strong privacy protections. Altman emphasized the urgent need for a new legal and policy framework to address AI-era confidentiality.

While platforms like WhatsApp and Signal use end-to-end encryption to ensure message privacy, OpenAI currently reviews some user conversations for training and safety purposes. According to OpenAI’s policy, chats on the free version of ChatGPT may be retained for up to 30 days and could be stored longer for legal or security reasons.

Altman acknowledged this gap could hinder the long-term adoption of AI for sensitive, human-centered use cases. “We should have the same concept of privacy for your conversations with AI that we do with a therapist or a lawyer,” he said. “No one had to think about that even a year ago—but now we do.”

The comments underscore a pivotal challenge for the AI industry: how to ensure user trust and protect privacy as generative AI becomes increasingly intertwined with daily life.


Recent Random Post: