High risk · Annex III(2)

AI in Critical Infrastructure: High-Risk Under the EU AI Act

When AI helps run the systems a society can't function without, a failure isn't an inconvenience — it's a safety event. Annex III point 2 makes AI safety components in critical infrastructure high risk.

What Annex III(2) covers

  • Safety components in critical digital infrastructure
  • Road-traffic management
  • Supply of water, gas, heating and electricity

Who’s affected

Utilities, grid and pipeline operators, traffic-management authorities, and the vendors that supply AI safety components to them.

Why it’s high risk

AI acting as a safety component in these systems is high risk under Annex III(2). The emphasis on accuracy, robustness and cybersecurity (Art. 15) is especially heavy here, because the failure mode is physical harm.

Obligations to prepare for

Risk management system (Art. 9)
Data governance & bias checks (Art. 10)
Technical documentation — Annex IV (Art. 11)
Record-keeping / logging (Art. 12)
Transparency & instructions (Art. 13)
Human oversight (Art. 14)
Accuracy, robustness & cybersecurity (Art. 15)

FAQ

Does all AI in utilities count as high-risk?

No — only AI intended to be used as a safety component in the management or operation of the listed infrastructure. Back-office or purely administrative AI is judged on its own use case.

Being high-risk sets the obligations; your role decides who owns them. See provider vs deployer duties.

Not sure if your system falls under Annex III(2)?

The free checker classifies it deterministically in 2 minutes — no sign-up.

Run the free check