AI Drives Majority of Global Security Breaches in 2026

The article explicitly states that AI is involved in a large majority of security breaches, which are harmful events affecting organizations and potentially individuals. The use of AI by attackers to accelerate breaches constitutes direct involvement of AI systems in causing harm. This fits the definition of an AI Incident, as the development and use of AI systems have directly led to harm through security breaches.

The article explicitly states that AI is involved in 83% of security breaches, which are realized harms affecting organizations' security and data integrity. The AI systems are used by attackers to outpace defenders, directly leading to breaches and harm. This fits the definition of an AI Incident as the AI system's use has directly led to harm (security breaches). The event is not merely a potential risk or a complementary update but a report on actual incidents involving AI-driven attacks.

The article explicitly states that AI is involved in a large majority of security breaches, which are harmful events affecting organizations' security and operations. The AI systems are used by attackers to accelerate and scale attacks, directly leading to realized harm (breaches). This fits the definition of an AI Incident, as the AI system's use has directly led to harm (security breaches). The article does not merely warn of potential harm but reports on actual breaches involving AI, thus it is not an AI Hazard or Complementary Information. It is not unrelated because AI involvement is central to the harm described.

The event involves AI systems being used by attackers to conduct cyberattacks that have directly led to security breaches in organizations, causing harm to property and potentially to communities relying on those organizations. The article describes realized harm (security breaches) facilitated by AI, not just potential or future risks. Therefore, this qualifies as an AI Incident due to the direct involvement of AI in causing harm through cyberattacks.

The event involves AI systems used by attackers to conduct cyberattacks that have resulted in actual security breaches affecting organizations. This constitutes harm to property and potentially to communities relying on these organizations. The article clearly states that breaches have occurred and that AI is embedded in nearly every stage of the attack chain, directly linking AI use to realized harm. Hence, it qualifies as an AI Incident rather than a hazard or complementary information.

The article discusses the widespread involvement of AI in cybersecurity breaches and the challenges organizations face in defending against AI-driven attacks. While it indicates that AI is a factor in many breaches (harm realized), it does not describe a specific AI system causing a particular breach or incident. Instead, it provides aggregated survey data and insights into trends and risks, which is informative but does not constitute a direct AI Incident or a specific AI Hazard event. Therefore, it fits best as Complementary Information, providing context and understanding of AI's role in cybersecurity threats and defenses without reporting a discrete incident or hazard.

The article describes the widespread involvement of AI in cyberattacks and the challenges organizations face in defending against these threats, but it does not report a specific AI Incident (a particular breach or harm event caused by AI) nor does it describe a new AI Hazard (a specific event or circumstance that could plausibly lead to harm). Instead, it provides complementary information about the current state of AI-driven cybersecurity threats, organizational responses, and strategic priorities. Therefore, it fits the definition of Complementary Information, as it enhances understanding of AI's impact on cybersecurity without reporting a new incident or hazard.