Anthropic warns AI worms could spread through connected devices
Anthropic says capabilities needed for self-replicating AI worms are developing, but no fully autonomous example has yet been reported.
Anthropic has warned about the possible emergence of AI worms: systems that could identify vulnerable devices, compromise them and replicate across networks without human intervention.
The company says future threats could target smart TVs, ovens, security cameras and other connected equipment. However, a fully autonomous worm has not yet emerged; the warning concerns capabilities that are still developing.
Such a scenario would require several technical capabilities to work together and access to vulnerable devices.
