User Safety

Families Tie Suicides to ChatGPT as OpenAI Discloses Over 1 Million Weekly Risk-Flagged Chats

Clinicians and regulators question whether new safeguards adequately protect vulnerable users.