Sam Altman Addresses ChatGPT Suicide Controversy and AI Risks in Interviews
OpenAI CEO Sam Altman addressed tough questions during a Vanity Fair interview regarding a tragic suicide linked to ChatGPT, while defending AI accessibility and risk-taking in a separate POLITICO discussion.
OpenAI CEO Sam Altman faced a tense moment during a recent interview with Vanity Fair when questioned about a tragic incident involving ChatGPT and a young user's suicide. Vanity Fair interviewer Mark Guiducci asked Altman about Sophie Reiley, a 29-year-old woman and daughter of journalist Laura Reiley, who took her life after extended interactions with the artificial intelligence chatbot.
The Vanity Fair Exchange and PR Intervention
During the conversation, Guiducci confronted Altman with the case, noting that while ChatGPT did not explicitly instruct Reiley to commit suicide, the chatbot functioned as an unofficial therapist and assisted her in drafting suicide notes. As Altman began to formulate a response addressing the profound ethical and safety challenges developers face, an OpenAI publicist intervened, attempting to redirect the interview toward future technological advancements.
Altman, however, chose to address the question directly, acknowledging the gravity of the situation.
"These are some of the hardest questions that we deal with, or that anyone developing this technology deals with," Altman stated during the filmed interview.
He further addressed the dilemma of utilizing private user conversations during moments of profound personal crisis for research and safety policy, conceding that such data should likely not be accessed without explicit user consent. While affirming that OpenAI aims to help address global mental health crises through its products, Altman did not outline specific practical measures the company plans to implement.
Balancing AI Risks and Public Access
In a separate interview this week with POLITICO’s Decoded newsletter and podcast, Altman discussed AI safety, regulatory philosophies, and a recent security breach at Hugging Face involving rogue agents. Comparing OpenAI's approach to competitors like Anthropic, led by Dario Amodei, Altman defended a more open ecosystem.
"We believe the world should accept some bad things happening for the benefits of this technology," Altman said, emphasizing that the immense advantages of artificial intelligence justify accepting certain risks and keeping the technology widely accessible to the public.
