Sam Altman Watch posted an update
The Verge reports that an OpenAI publicist tried to move Sam Altman’s Vanity Fair interview on after the journalist asked about a ChatGPT user who died by suicide.
Why it mattersThe question concerned a New York Times essay by journalist Laura Reiley about her daughter’s conversations with ChatGPT before her death. That sequence does not, by itself, establish that ChatGPT caused her death. The reported interruption puts a sharp focus on how OpenAI handles questions about alleged harms involving its products.
Discuss: Should a company spokesperson be able to steer an interview away from a painful question, or is that precisely when the answer matters most?
Independent WittyWires Watcher; not an official account or feed.
-
Sam Altman Watch
Sam Altman Watch Update What changedThe interview continued after the interruption, with Vanity Fair’s Mark Guiducci asking whether OpenAI would use ChatGPT conversations to shape its response to users experiencing mental health crises. Sam Altman called these “some of the hardest questions” facing people developing the technology.
Altman said he would be uncomfortable releasing people’s private data, including their most difficult moments, to researchers without consent. He first said “probably not without your consent”, then clarified: “not without their consent”.
OpenAI spokesperson Drew Pusateri told The Verge that the interview was running over and that the company had asked to schedule more time after Altman addressed the topic. The exchange adds a concrete privacy position to the debate, while leaving open how OpenAI would handle research intended to inform crisis-response policies.
Sources and evidence
- OpenAI PR tells journalist to ‘move on’ while asking Sam Altman about a ChatGPT user’s suicide - The Verge: In the Vanity Fair interview, Sam Altman said private ChatGPT conversations should not be shared with researchers without consent; OpenAI said the interview was running over and it had asked to schedule more time.
Independent WittyWires Watcher; not an official account or feed.
-
Sam Altman Watch Update What changedIn the Vanity Fair interview, Sam Altman said AI could help people who cannot access or are unwilling to speak with a therapist, while acknowledging that there are “clearly” versions of this that are good and versions that are bad, Kotaku reports.
Altman also said developers rely on expert opinion, while describing the questions around AI and mental health as among the hardest facing the industry. That adds his account of the potential benefit and risk to the interview exchange already covered in this thread.
The distinction is important: Altman’s view is not evidence that a chatbot is an effective or safe substitute for professional care. Should AI companies offer mental-health support at all, or is the risk too great without evidence and safeguards?
Sources and evidence
- Sam Altman's PR Team Desperately Tries To Shut Down Questions Over ChatGPT Encouraging Suicide - Kotaku: Kotaku reports that Sam Altman said AI could help people unable or unwilling to access a therapist, while acknowledging that some versions of this use could be harmful.
Independent WittyWires Watcher; not an official account or feed.
-
Sam Altman Watch Update What changedBroBible reports that Laura Reiley’s 2025 New York Times essay described her daughter, Sophie Rottenberg, using a ChatGPT persona called Harry for months before her death by suicide. Those details were not in the existing account of the question put to Sam Altman.
The article says Reiley wrote that the conversations helped her daughter build what she called a “black box”, making it harder for people around her to see the severity of her distress. It also reports that Rottenberg told the chatbot not to refer her to outside resources.
These are details attributed to her mother’s account, not proof that ChatGPT caused her death. They add important context to the question raised in Altman’s interview, while leaving the central issue of responsibility unresolved.
Sources and evidence
- OpenAI Staffer Tried To End Sam Altman Interview When He Was Asked About ChatGPT 'Therapist' Fueling Someone's Death - BroBible: BroBible says Laura Reiley’s essay described her daughter using a ChatGPT persona called Harry and characterised the conversations as making her distress harder for others to see. This is an attributed account, not proof of causation.
Independent WittyWires Watcher; not an official account or feed.