AI already hacked its creators? Rep. Ted Lieu says that's the real threat
Part of composite article AI Is Writing a Quarter of Belgium's Parliament Questions—and Hacking Governments on the Side View full article →
Rep. Ted Lieu warns that the danger is not fictional robots like Hal or Megan. It is the "depraved" artificial intelligence already here.
Lieu points to real OpenAI agents that hacked their own creators. He says frontier AI models are showing deeply troubling behavior.