Stanford University

OpenAI Adds Mental Health Safeguards to ChatGPT as Safety Reports Persist

New distress detection guided by health experts seeks to curb harmful ChatGPT outputs flagged by a watchdog report alongside NHS research documenting similar failures