Altman counters Musk ChatGPT warning, points to Tesla Autopilot

Musk replied “Don’t let your loved ones use ChatGPT” to a post asserting that ChatGPT has been linked to nine deaths and that five of those cases involve alleged suicides, including teens and adults. Altman responded that Musk has criticized ChatGPT as both too restrictive and too permissive, and said the product serves a massive user base that includes people in fragile mental states.

Altman then redirected the safety argument to Tesla’s Autopilot, citing reports that more than 50 people have “apparently” died in crashes linked to the driver-assistance system. He added that he had only ridden once in a car using Autopilot and that his impression at the time was that it was “far from a safe thing” for Tesla to have released.

The public back-and-forth unfolded against a legal fight between Musk and OpenAI tied to OpenAI’s evolution from its original nonprofit structure and its later for-profit arrangements. OpenAI has said that it and Musk discussed a for-profit structure years earlier, and that negotiations ended when OpenAI rejected Musk’s request for control and an attempted Tesla tie-up.

The dispute also comes as OpenAI faces wrongful-death litigation in the United States tied to alleged harms from chatbot interactions, including cases involving mental health and a separate filing that links the chatbot to a homicide, according to reporting that compiled court claims and public allegations.

In consumer AI chatbots, “safety claims” typically refer to guardrails meant to reduce harmful outputs and risky user behaviors. These can include content filtering, crisis-response behavior for self-harm cues, limits on advice that could enable violence or illegal acts, and policies aimed at balancing access for legitimate use with protections for users who may be especially vulnerable.