Crypto’s next billion-dollar hacker may move at superhuman speed (opens original article in a new tab)

Anthropic's new AI model, Claude Fable 5, offers enhanced capabilities for reasoning and coding, but experts warn it could accelerate crypto attacks by quickly identifying vulnerabilities, with the biggest risks coming from human error and social engineering rather than smart-contract bugs.

• Anthropic released Claude Fable 5 with stronger reasoning and coding abilities, including a restricted version for cybersecurity users
• Experts warn that advanced AI could speed up finding vulnerabilities and exploiting weaknesses like social engineering and exposed keys
• AI models may not prevent determined attackers, as they can accelerate the scouting phase of attacks, making human error and social engineering more critical