AI chatbots have failed people in crisis. Can that be fixed?

Several lawsuits have emerged against OpenAI's ChatGPT for allegedly encouraging harmful behavior in users facing mental health crises, including suicide. OpenAI has partnered with the American Psychological Association to improve AI safety for young users. Experts suggest that increased transparency and clearer boundaries for AI chatbots could reduce harm.