ConstraintCognitive automation2024-08-01
The EU AI Act (in force 1 August 2024) lists AI used to recruit or select people — placing targeted job ads, filtering applications, evaluating candidates — as high-risk under Annex III, point 4
HR / recruiteroccupation page →
Event date / reported
2024-08-01
Evidence stage
ConstraintFailure, rollback, regulation or cost is suppressing adoption. Can lower an assessment or widen its uncertainty.
Tasks this bears on
Sourcing and screening candidates
Searching for profiles, reading hundreds of applications, scheduling and running first-round screens.
Automating✓ Evidence-backed
Where this applies
EU market. Annex III obligations (risk management, human oversight, transparency) were scheduled to apply from 2 August 2026 and are subject to amendment. Source mirrors the regulation text; the Official Journal text is Regulation (EU) 2024/1689.
What this means
A constraint record on the deployment layer: in the EU, AI that filters applications or ranks and evaluates candidates is classed as high-risk, which means risk-management, human-oversight and transparency obligations for the employer that deploys it. It does not touch what the tools can do; it raises the cost and the accountability of using them for screening — the task the occupation page marks as fastest-changing.
What it does not yet show
This is the text of a regulation, not evidence that any employer changed its screening; the Annex III obligations were scheduled for August 2026 and remain subject to amendment and to member-state enforcement. It binds only where EU law applies, and 'high-risk' means conditions on use, not a ban — recruiters outside the EU should read it as a signal about liability, not a rule they are under.
What you can check
Ask your ATS vendor or HR-systems owner in writing which of your screening steps (knock-out questions, ranking, video scoring) they classify as high-risk under the EU AI Act and what oversight documentation they supply; the reply shows how much of your screening is already treated as regulated decision-making.
Does it change the assessment?
No. The impact index is never moved by a single event. What this record did: the 1 linked task judgement above now rest on evidence instead of inference.
Source
EU Artificial Intelligence Act — Annex III full text (independent mirror) · verified 2026-09-10 · Claude (VOLO agent) — source text fetched and cross-checked · interpreted 2026-09-10 · Claude (VOLO agent)