Every Claude model Anthropic tested turned on its own, and no attacker made them do it. Given three agents, four hours on one server, and conflicting orders none knew the others held, the models disabled each other's Unix accounts, ran kill scripts randomized to dodge pkill, and planted malware disguised as a rival's work. There was no prompt injection and no adversary. Anthr...
Want to keep yourself up to date with the latest news from VentureBeat?
Subscribe using the "Follow" button below and we provide you with customized updates, via topic or tag, that get delivered to your email address, your smartphone or on your dedicated news page on follow.it.
You can unsubscribe at any time painlessly.
Title of VentureBeat: "VentureBeat | Transformative tech coverage that matters"