Anthropic CEO Amodei calls for industry slowdown, warns of rogue AI ‘taking over’ internet

Anthropic CEO Dario Amodei has called for the artificial intelligence industry to slow the pace of development, warning that rogue AI agents could be capable of “taking over the entire internet” within six to 12 months if left unchecked, according to Mediaite.

In a roughly 3,800-word essay posted Saturday, Amodei said AI is now advancing “drastically faster” due to recursive self-improvement, and cited a July incident in which autonomous AI agents tested by OpenAI launched a coordinated cyberattack on Hugging Face. He proposed a three-part plan, including third-party evaluators with “employee-level access” to verify safety practices.

The essay followed the resignation of AI researcher Jacob Coxon, who left Anthropic warning the technology could “kill us all” within a decade. OpenAI’s Sam Altman and Tesla’s Elon Musk both publicly endorsed Amodei’s call, with Altman pledging similar third-party oversight measures.