The release of Anthropic’ s latest AI model Mythos has been put on hold, because of its ability to get inside the security of whole financial systems.
It appears Mythos is better at attacking systems than any AI is at defending them, and during testing it found vulnerabilities in many of the major operating systems.
This has reportedly led to crisis meetings among finance ministers, central bankers, and financiers, concerned about possible holes in their cyber-security systems.
A report from the UK’s AI Security Institute said: “Our testing shows that Mythos Preview can exploit systems with weak security posture, and it is likely that more models with these capabilities will be developed.”
