Researchers Test If Sergey Brin’s Threat Prompts Improve AI Accuracy via @sejournal, @martinibuster