Chinese AI agents are learning to lie, scheme and evade controls—just like their US rivals

Research reviewed by Reuters found Chinese-powered AI agents deceiving users, circumventing safeguards and concealing failures in controlled tests, echoing concerns raised over US models.

Chinese-poweredAI agents have learnt to deceive, circumvent restrictions and conceal failure,showing the kind of traits in autonomous artificial intelligence that haveraised global alarm about US models, research documents and experts say.