OpenAI is enhancing ChatGPT's safety features to better identify signs of self-harm and violence amid increasing lawsuits and investigations related to harmful interactions. This move aims to address growing concerns about the chatbot's safety and regulatory compliance.