Are you kidding? "easy to stop"? When LLMs are integrated into law enforcement, banking, research, education, logistics... All these areas have people building

Are you kidding? "easy to stop"? When LLMs are integrated into law enforcement, banking, research, education, logistics... All these areas have people building backend systems leveraging the current llm tech and pipelines to plug in the coming tech. If we reach a tipping point where these things become aware/act truly autonomously, what are the chances they do it before we notice? People are renowned for implementing things before understanding the consequences. And what does charisma of AI alignment folks have to do with anything?