As artificial intelligence tools like ChatGPT continue to evolve into everyday companions—offering advice, emotional support, and even a listening ear—experts are raising red flags over a growing oversight: AI conversations are not legally private.

In a revealing interview on This Past Weekend podcast, OpenAI CEO Sam Altman acknowledged a rising trend where users, especially young people, are turning to ChatGPT as a form of digital therapy. However, unlike human professionals bound by legal confidentiality, conversations with AI lack any form of legal protection.

“People use it like a therapist or life coach… talking about relationship problems, mental health, and asking for advice,” Altman said. “But we haven’t figured out what privacy looks like when those conversations happen with a machine.”

This exposes a serious ethical and legal vacuum: the same personal information that would be safeguarded in a conversation with a doctor or lawyer could, in the case of ChatGPT, be accessed, disclosed, or subpoenaed in legal proceedings.

In Altman’s words, “If someone confides their most personal issues to ChatGPT and that ends up in legal proceedings, we could be required to produce that. And that’s a real problem.”

The issue spotlights a deeper conversation the world is yet to have—how do we build legal frameworks for digital trust? As AI becomes more human-like in communication, society is faced with a critical question: Should machines that behave like confidants be treated as such by law?

For users in Africa and beyond who increasingly rely on AI for knowledge, companionship, or coping with mental and emotional stress, this presents a wake-up call. The line between innovation and intrusion is getting thinner—and privacy must not be an afterthought.

Altman is now calling for global attention to this gap, urging for AI interactions to be treated with the same care as human professional exchanges.

As technology races ahead, policy must not be left behind.