AI and misinformation

This was a good article on AI systems dealing with misinformation and disinformation.

Is it possible to just pull the plug on the bad robots?