TechCanadian researcher launches AI safety lab as rogue-agent reports mount

As reports of rogue AI agents proliferate, calls for stronger safety and control are growing. Jacob Tsimerman, one of the world’s leading mathematicians and a Canadian, is taking a leave from the University of Toronto to work on the problem. He says these systems are more capable than even top human cybersecurity experts or hacker teams, making a purely adversarial approach a losing game.