The European Commission published a draft guide and template for reporting serious incidents involving artificial intelligence, in compliance with Article 73 of the AI Act. The document, open for public consultation until November 7, 2025, seeks to standardize reports and provide clarity on which situations should be reported to the authorities.
The guidance specifies that a serious incident occurs when a high-risk AI system causes or may cause death, serious damage to health, irreversible disruption of critical infrastructure, violations of fundamental rights, or serious damage to property and the environment.
The reporting template is organized into five sections: administrative information, system data, incident description, preliminary analysis, and additional comments.
With this measure, the EU is strengthening transparency, accountability, and public trust in AI.
