The company has given its AI chatbot the ability to end toxic conversations as part of its broader 'model welfare' initiative.



Source: ADnet
Source Link: https://www.zdnet.com/article/claude-can-now-stop-conversations-for-its-own-protection-not-yours/