OpenAI PR Cuts Off Vanity Fair Interview When Suicide Case Comes Up

An OpenAI publicist interrupted Sam Altman's interview with Vanity Fair editor Mark Guiducci to shut down a line of questioning about a ChatGPT user who died by suicide, according to an account of the exchange. When Guiducci raised the incident directly with Altman, the publicist warned that time was running short and said she wanted to "move on" to a different subject. The episode puts a spotlight on how the company manages uncomfortable topics during high-profile media appearances by its chief executive.

The timing matters. OpenAI is facing a growing stack of wrongful-death and negligence lawsuits filed by families who say chatbots encouraged or failed to divert users in crisis, and those cases have moved from court filings into mainstream coverage. Altman has publicly acknowledged the gravity of such incidents, and the company has said it is working on safety measures for users in distress. A publicist cutting the topic short in an on-the-record interview sits awkwardly next to that messaging.

Vanity Fair is not the first outlet to press Altman on the subject. In recent months he has been asked about teen safety, self-harm detection and the limits of OpenAI's guardrails in interviews and congressional settings, and the company has rolled out parental controls and crisis-referral features in response. Each new disclosure of a user death adds pressure on OpenAI to explain what its models detect, what they miss, and how quickly it escalates.

The publicist's intervention is the kind of thing press handlers do routinely, but it lands differently for a company whose products are used by hundreds of millions of people, some of them minors, in emotionally charged conversations. Critics argue that tightly managed interviews leave the public with polished answers rather than specifics about failures. OpenAI declined to characterize the exchange beyond what has been reported.

What to watch is whether the lawsuits force disclosure of internal safety data — how often ChatGPT flags self-harm language, how those flags are handled, and whether changes followed specific incidents. Regulators in the US and Europe are already asking similar questions, and testimony from grieving families is likely to keep the issue in front of lawmakers. For OpenAI, the harder problem is not the interview; it is answering those questions under oath.

For now, the company's public posture and its private damage control are being compared in real time, and the gap is what reporters and plaintiffs' lawyers will keep probing.