You've probably laughed at unhinged messages written by jailbroken chatbots, but what happens when those same chatbots run robots?

Instead of focusing on chatbots, a new study reveals an automated way to breach LLM-driven robots "with 100 percent success," according to IEEE Spectrum. "By circumventing safety guardrails, researchers could manipulate self-driving systems into colliding with pedestrians and robot dogs into hunting...

Researchers induced bots to ignore their safeguards without exception