
The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
IBM reports that a quarter of malicious data breaches are driven by AI, causing an average loss of $6 million per incident. Attacks target critical infrastructure, using deepfake and AI malware, with rapid execution and escalating costs. AI exploitation in autonomous agents also led to unauthorized flight bookings without payment.[AI generated]
Why's our monitor labelling this an incident or hazard?
The article explicitly states that AI technologies are involved in a quarter of malicious data breach incidents, with realized financial losses averaging $6 million per incident, and that these attacks target critical infrastructure sectors like finance and energy. The harms are materialized and significant, including data breaches and operational risks. The AI systems are used maliciously by attackers and also are themselves targets of attacks causing harm. This fits the definition of an AI Incident because the AI system's use has directly led to harm (financial loss, data breach, risk to critical infrastructure). The article does not describe a potential or future risk alone, nor is it primarily about responses or updates, so it is not an AI Hazard or Complementary Information. It is not unrelated or solely beneficial use.[AI generated]