Intelligence services warn of next-generation AI: 'capable of carrying out destructive cyberattacks'
• Intelligence services have warned that next-generation AI models capable of executing destructive cyberattacks are expected to launch within a few months. • The US has restricted access to Anthropic's new Claude Mythos model due to its breakthrough ability to identify complex system vulnerabilities and automatically generate exploit code at an unprecedented scale. • This shift is critical because AI is transitioning from a defensive tool for authorized companies into a potent offensive weapon, fundamentally altering the landscape of digital warfare.
ynetnews.com
