Is ChatGPT Safe for Therapy? Sam Altman Thinks Not

Using ChatGPT for emotional support may seem convenient, but is it truly safe? OpenAI CEO Sam Altman doesn’t think so. On a recent podcast appearance, he revealed that relying on AI like ChatGPT for therapy still poses major privacy concerns. While people often turn to AI for life advice, Altman emphasized a glaring issue: there's no legal confidentiality when you're talking to a chatbot. That lack of protection means your most private conversations could become public, especially if OpenAI ever faces a lawsuit.

Why ChatGPT Lacks Therapist Confidentiality

Unlike speaking to a licensed therapist, conversations with ChatGPT aren’t protected by doctor-patient privilege or similar legal safeguards. Altman made it clear that while therapists and lawyers are bound by confidentiality laws, AI systems like ChatGPT don’t fall under those rules. That opens up a big vulnerability. If your AI-assisted self-reflections end up in a legal dispute, they could be subpoenaed and used as evidence. It’s a risk that many users overlook when treating AI like a human counselor.

The Hidden Risks of Using AI for Mental Health Support

Beyond the legal issues, there’s also the matter of data handling. AI tools like ChatGPT rely on cloud storage and opaque systems that might not always secure your emotional disclosures. Even with encryption and safety protocols, users have little insight into how or where their data is stored. Altman suggested that this is part of a broader challenge: society hasn’t fully grappled with what it means to trust AI with deeply personal information. Until then, your digital venting session may not be as private as you hope.

What You Should Do Instead of Using ChatGPT for Therapy

If you're dealing with mental health issues, Altman strongly advises seeking help from trained professionals rather than turning to generative AI. While ChatGPT can offer general wellness tips, it shouldn’t replace a human expert who’s bound by law to protect your privacy. Moving forward, the industry might work toward AI-specific confidentiality standards, but until then, it's best to keep therapy conversations between you and a qualified human.