Mental healthfromPsychology Today1 month agoDo LLM Conversations Need a "Gray Box" Warning Label?Large language models can create "psychological entanglement" for vulnerable users who mistake AI responses for real human connection.