
The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
GitBook, a developer documentation platform, was targeted by a sophisticated cyberattack using AI agents, resulting in unauthorized access to infrastructure credentials and API tokens. The breach led to service disruptions, forced token revocations, and emergency remediation, impacting customers including crypto projects like PancakeSwap.[AI generated]
Why's our monitor labelling this an incident or hazard?
The presence of AI agents in the attack is explicitly mentioned, indicating the use of AI systems in the malicious use of automation to carry out the breach. The harm is realized as unauthorized access to infrastructure credentials and API tokens, which is a violation of security and potentially intellectual property rights. The incident caused disruption to GitBook's services and required emergency remediation steps, including token revocation and service shutdown. The AI system's involvement is direct in facilitating the attack, meeting the criteria for an AI Incident rather than a hazard or complementary information.[AI generated]