High-risk AI guide · Annex III

Annex III point 4

High-risk AI in HR and employment

AI in HR needs early classification because recruitment, monitoring, task allocation and performance tooling can directly affect people.

What falls in scope?

Recruitment, selection, matching and candidate assessment.

Targeted job ads, filtering, ranking and interview analysis.

Promotion, termination, employment terms, task allocation or monitoring.

People analytics that affects individual opportunities or warnings.

When to review first?

CV screening or candidate ranking

Interview or assessment analysis

Performance scoring or behaviour monitoring

Vendor claims about AI Act or bias

What should a first review deliver?

Use-case map per HR workflow

First high-risk classification

Bias, GDPR and transparency questions

30/60/90-day evidence route

Recognisable situations

1

Recruitment

AI filters applications or ranks candidates for interview.

2

Monitoring

A tool assesses individual performance or behaviour patterns.

3

Task allocation

AI assigns work based on predicted productivity or suitability.

Based on official sources

This page is a practical guide for first classification and preparation. For legal decisions, a full system and context review remains necessary.

View official Annex III text

Make this concrete for your systems.

Start with the Gap Intake and share which AI systems, suppliers and processes are in scope. Then we can sharpen the classification and first evidence route.