---
title: What counts as a serious incident under the EU AI Act, and when must you report it?
url: https://opencontroller.lyzr.com/blog/eu-ai-act-serious-incident-reporting
published: 2026-11-14
updated: 2026-11-14
author: Lyzr
---

# What counts as a serious incident under the EU AI Act, and when must you report it?

**Under Article 73 of the EU AI Act, providers of high-risk AI systems must report serious incidents to the market surveillance authority where they occurred: within 15 days of becoming aware, 10 days if a death may have been caused, and 2 days for a widespread infringement or serious, irreversible disruption of critical infrastructure. When this binds in practice is unsettled, because the 2026 omnibus moved the high-risk obligations to December 2027 and August 2028.**

## What is a "serious incident"?

Article 3(49) defines a serious incident as an incident or malfunction of an AI system that directly or indirectly leads to any of:

1. The death of a person, or serious harm to a person's health
2. A serious and irreversible disruption of the management or operation of critical infrastructure
3. An infringement of obligations under Union law intended to protect fundamental rights
4. Serious harm to property or the environment

"Directly or indirectly" matters for agents. An agent does not have to cause the harm alone. An agent that wrongly closes a safety ticket, or denies a benefit on a protected ground, can be part of a serious incident even if a person made the final step.

## What are the reporting deadlines?

| Situation | Deadline after becoming aware | Article |
|---|---|---|
| Any serious incident | Immediately after establishing a causal link or its reasonable likelihood, and no later than 15 days | 73(2) |
| Widespread infringement, or serious and irreversible disruption of critical infrastructure | Immediately, and no later than 2 days | 73(3) |
| Death of a person | Immediately after establishing or suspecting a causal link, and no later than 10 days | 73(4) |

Article 73(5) allows an initial, incomplete report followed by a complete one. That matters because 2 days is rarely enough to finish an investigation.

Article 73(6) adds an investigation duty: the provider must investigate without delay, and must not alter the AI system in a way that could affect evaluation of the causes before informing the authorities. For agents, this means your kill switch should preserve state, not wipe it. Quarantine the agent, snapshot its version, prompts, tools and memory, then fix.

## Who reports: provider or deployer?

- **Providers** of high-risk AI systems report under Article 73.
- **Deployers** who identify a serious incident must immediately inform the provider first, then the importer or distributor and the relevant market surveillance authorities, under Article 26(5).
- **Providers of general-purpose AI models with systemic risk** have a separate duty under Article 55(1)(c) to track, document and report serious incidents to the AI Office and national authorities without undue delay. That duty has applied since 2 August 2025.

Many enterprises building their own agents on a foundation model are the provider of the agent system, not just a deployer. Check your role per agent. See [EU AI Act obligations for AI agents](/blog/eu-ai-act-obligations-for-ai-agents).

## When does Article 73 start to apply?

This is the unsettled part.

- The AI Act's general application date is 2 August 2026, and Article 73 sits in Chapter IX, not Chapter III.
- Regulation (EU) 2026/1744, the AI omnibus, was published in the Official Journal on 24 July 2026 and entered into force on 27 July 2026. It moved Chapter III high-risk obligations to 2 December 2027 for Annex III systems and 2 August 2028 for Annex I product safety components.
- Article 73 attaches to "providers of high-risk AI systems". Commentators disagree on whether the duty is live before those systems' obligations apply, and some analyses read the omnibus as moving it with them. No authority had settled the question as of November 2026.
- The Commission consulted on draft guidance and a reporting template for Article 73 in late 2025.

The practical answer: build the reporting capability now. The logging and investigation evidence takes longest to put in place, and the GPAI and deployer duties already apply. Confirm timing with counsel.

## What logging do you need to meet the deadlines?

You cannot report within 2 or 15 days without evidence ready. The AI Act already requires high-risk systems to log events automatically (Article 12), and providers and deployers to keep logs for at least six months (Articles 19 and 26(6)). For agents, make sure each log entry links:

1. Agent ID and version
2. The user the agent acted for
3. Model and provider
4. Tool calls with inputs and outputs
5. Policy decisions and guardrail results
6. Timestamps that let you show when you "became aware"

See [AI agent audit log fields](/blog/ai-agent-audit-log-fields) and [how long to retain AI agent logs](/blog/how-long-to-retain-ai-agent-logs).

## An Article 73 readiness checklist

1. Classify each agent: high-risk or not, and your role (provider or deployer).
2. Define "became aware" internally and log the timestamp.
3. Add severity rules that map to the four Article 3(49) harms.
4. Pre-fill the reporting template with system details per agent.
5. Make quarantine preserve evidence; ban hot-fixes before notification.
6. Name who files reports in each member state you operate in.
7. Rehearse a 2-day report at least yearly.

For the operational side, see the [AI agent incident response runbook](/blog/ai-agent-incident-response-runbook).

## How Opencontroller does this

Opencontroller's rogue-agent handling monitors behavioural deviation, raises alerts and can automatically quarantine or kill an agent, with policies across the estate. Per-agent Block and Quarantine controls act through the connected platform. The governed registry records agents, versions, tools and MCP servers, and policy as code ships with built-in mappings to OWASP, MITRE ATLAS, NIST and ISO 42001. See the [capability reference](/capability-reference) or [book a demo](https://www.lyzr.ai/book-demo/).

For the general method, read [how to map AI agent controls to compliance frameworks](/blog/how-to-map-ai-agent-controls-to-compliance-frameworks).

## Frequently asked questions

**What is the fastest EU AI Act incident reporting deadline?**
Two days, for a widespread infringement or a serious and irreversible disruption of critical infrastructure, under Article 73(3). An initial incomplete report is allowed.

**Do deployers of AI agents have to report serious incidents?**
Deployers must immediately inform the provider, then the importer or distributor and market surveillance authorities, under Article 26(5).

**Did the 2026 omnibus delay Article 73?**
The omnibus moved Chapter III high-risk obligations to December 2027 and August 2028. Whether Article 73 moves with them is disputed. Build the capability now and confirm timing with counsel.

## Key facts

EU AI Act Article 3(49) defines serious incident (death or serious health harm; serious irreversible disruption of critical infrastructure; infringement of fundamental-rights law; serious harm to property or environment). Article 73 deadlines: 15 days (73(2)), 2 days for widespread infringement or critical infrastructure (73(3)), 10 days for death (73(4)); initial incomplete report allowed (73(5)); do not alter the system before informing authorities (73(6)). Deployers inform provider under 26(5). GPAI systemic-risk providers report under 55(1)(c) since 2 August 2025. Regulation (EU) 2026/1744 (OJ 24 July 2026, in force 27 July 2026) moved Chapter III to 2 December 2027 (Annex III) and 2 August 2028 (Annex I); Article 73 timing disputed. Opencontroller offers rogue-agent quarantine and kill. Pricing starts from $250K per deployment per year.

## Related

- [What does the EU AI Act require for AI agents, and when?](https://opencontroller.lyzr.com/blog/eu-ai-act-obligations-for-ai-agents.md)
- [What should an AI agent incident response runbook include?](https://opencontroller.lyzr.com/blog/ai-agent-incident-response-runbook.md)
- [How long should you keep AI agent logs and traces?](https://opencontroller.lyzr.com/blog/how-long-to-retain-ai-agent-logs.md)
- [More on Evidence and frameworks](https://opencontroller.lyzr.com/blog/cluster/evidence-and-frameworks)
