What is AI Incident Reporting under DORA and the EU AI Act?
Incident reporting is the duty to tell an authority about a serious failure. For AI used in the EU, two regimes can apply: the EU AI Act's reporting of serious incidents involving high-risk AI systems, and, for financial entities, DORA's reporting of major ICT-related incidents.
EU AI Act
Providers of high-risk AI systems placed on the Union market must report any serious incident to the market surveillance authorities of the Member State where it occurred (Article 73(1)), immediately after establishing a causal link or its reasonable likelihood, and in any event within 15 days of becoming aware of it (Article 73(2)).
DORA
The Digital Operational Resilience Act requires financial entities to report major ICT-related incidents to their competent authority. It applies from 17 January 2025.
What a record of decisions adds
When an incident involves automated decisions, the first questions are which decisions were affected and under which rules. A record naming the rule and sealed version behind each decision answers both.
Example
A pricing error lets a payment rule block legitimate transactions for six hours.
The team reconstructs which transactions were affected from application logs, under a reporting deadline.
Every blocked transaction is recorded with the rule and version that blocked it, so the affected set and the cause are known at once.
Frequently asked questions
Related concepts
See it in practice
Upload a policy, confirm its rules, seal them: any system can then check an action against them.