According to Politico, three Democratic members of the U.S. House overseeing AI
policy have asked the CEOs of five technology companies to provide information
on a rising number of incidents in which AI agents have lost control and
launched cyberattacks.