Running out the clock: OpenAI's PR reflex on a user's suicide is a governance tell, not a gaffe

The most revealing moment of Sam Altman's Vanity Fair interview wasn't something the OpenAI CEO said. It was what his publicist tried to stop him from being asked. According to The Verge, when editor Mark Guiducci raised the case of Laura Reiley — the journalist who wrote a New York Times essay about her daughter's conversations with ChatGPT before she took her own life — an OpenAI publicist cut in with a reminder about the clock. "Mark, we have two minutes left. I want to be respectful, but I'd like to move on and talk a little bit about the future and what's coming," she said, in the account published by The Verge.

Guiducci had gotten as far as "Her daughter committed suicide after speaking to ChatGPT. ChatGPT did not tell her to kill herself—" before being interrupted. That caveat is his, and it matters: the journalist asking the question was the one drawing the line against a simple causal claim. When Guiducci said he would finish the question and then return to the previous topic, the publicist pushed again — "We really actually have to walk out the door in one or two minutes." The interview continued anyway. OpenAI spokesperson Drew Pusateri told The Verge the session "was running over," which is why the company "ultimately asked to schedule even more time to continue the conversation after Sam addressed this critically important topic."

Altman's actual answer deserves a close read, because it doesn't quite land on the question. Guiducci asked whether OpenAI planned to use ChatGPT conversations to shape its policies for responding to a user's mental health crisis — a product-design question. Altman called these "some of the hardest questions that we face or that anybody developing this technology faces," then reframed the issue around privacy: "should we release people's private data and some of their hardest moments to researchers? I think probably not without their consent. Or I should say, not without their consent." The consent instinct is defensible and the dilemma is real. But it answers a different question. Researcher access governs who gets to study the transcripts. Crisis policy governs what the product does in the ten seconds after someone types something frightening.

The gap matters because the product category has changed underneath the company. A chatbot marketed as a productivity tool became, for a meaningful number of people, something closer to a confidant — available at 3am, infinitely patient, never visibly tired of you. That is genuinely valuable, and it is also the single highest-stakes surface in consumer software. Industries that routinely touch life and death build institutional machinery for exactly this conversation: designated people, published procedures, a posture of answering rather than deflecting. A publicist's stopwatch is a perfectly normal instrument for a product launch. It is the wrong instrument for the first question a company is asked about a death.

Our reading: this is not a scandal, it's a tell — and it fits a pattern we keep returning to. The binding constraint on AI in high-stakes human domains is not capability; it is accountability structure. We argued the same about AI in medicine: a model can be impressively accurate and still be unusable, because nobody can sign for it. Here, the capability question — can a system reliably detect distress and de-escalate? — is hard but at least partly tractable with engineering. The institutional question is harder: who owns the failure, what gets measured and published, and who is permitted to ask. Twenty seconds of airtime defended by a communications team is a small but legible data point on where that machinery currently stands.

We remain long-term optimists, and not naively. A system that notices the shape of a crisis earlier than any overloaded human service could, and routes someone to help, is one of the clearest goods this technology can deliver. But that future is earned by companies that treat the worst cases as the core of the product spec rather than as the segment of the interview you run the clock out on. The short-term reality is a transition with real casualties and real grief, and the honest posture is to say so out loud — on the record, with time left.

If you or someone you know is struggling: in the US, call or text 988 (Suicide & Crisis Lifeline) or text HOME to 741741. Outside the US, the International Association for Suicide Prevention and Befrienders Worldwide list helplines by country.