This article was authored by Daniel Ilan, Rahul Mukhi, Prudence Buckland, and Melissa Faragasso from Cleary Gottlieb, and Brian Lichter and Elijah Seymour from Stroz Friedberg, a LevelBlue company.

Recent disclosures by Anthropic and OpenAI highlight a pivotal shift in the cyber threat landscape: AI is no longer merely a tool that aids attackers, in some cases, it has become the attacker itself. Together, these incidents illustrate immediate implications for corporate governance, contracting and security programs as companies integrate AI with their business systems. Below, we explain how these attacks were orchestrated and what steps businesses should consider given the rising cyber risks associated with the adoption of AI.