OpenAI Enhances ChatGPT Safety Features for Mental Health Support
OpenAI is upgrading ChatGPT’s safety features to better address user mental health crises. The AI company aims to enhance detection of various mental distress forms and strengthen safeguards during extended conversations, where current protocols may falter. Planned enhancements include improved mental health responses, one-click access to emergency services, and connections to therapists. New parental controls will allow monitoring of teens’ interactions with ChatGPT, and mental health resources will be localized for global users.
ChatGPT already includes measures like recognizing self-harm expressions, blocking harmful responses, and referrals to hotlines, backed by a network of over 90 mental health experts. With a user base exceeding 700 million, ChatGPT has shifted from basic conversation to offering personal advice and emotional support. Recent updates leveraging the new GPT-5 model have shown a 25% reduction in problematic responses during mental health emergencies, demonstrating OpenAI’s commitment to user safety.
