Annex III of the AI Act

Annex III lists the uses of AI that lawmakers consider high-risk because of their impact on people's rights. There are eight areas:

• Biometrics (remote identification, categorisation, emotion recognition).
• Critical infrastructure (water, gas, electricity, traffic).
• Education and vocational training (admissions, assessment, cheating detection).
• Employment and worker management (CV screening, candidate assessment, promotion or dismissal decisions).
• Access to essential services (creditworthiness, life and health insurance pricing, public benefits, emergency calls).
• Law enforcement.
• Migration, asylum and border control.
• Administration of justice and democratic processes.

For most companies the common cases are HR and credit or insurance. If you use AI in these processes, even a tool bought from a vendor, you have obligations as a deployer.