This case raises questions about what should happen when chatbots identify a potential threat before a crime has been committed.

Advertisement