Targeted job advertising with AI: high-risk and a discrimination risk
AI that shows job ads to specific groups falls explicitly under Annex III of the AI Act as high-risk. The biggest risk is discrimination: an algorithm that shows a vacancy mostly to young men invisibly excludes others. The GDPR and the DSA set additional limits.
Short answer: Using AI to target job ads โ deciding who does and does not see a vacancy โ is named explicitly in Annex III, point 4 of the AI Act as high-risk. It is not an edge case. The reason is discrimination: an algorithm that shows a vacancy mostly to a certain age, gender or origin excludes others without anyone noticing. On top of that come the GDPR and the DSA.
Why this is high-risk
Recruitment begins before the application: with who sees the vacancy at all. Ad platforms optimise for "click likelihood", and that optimisation can lock onto demographic patterns โ the technical role lands mostly with men, the care role mostly with women. The result is a biased applicant pool you never see, because the excluded group never got to apply.
The discrimination risk
This touches equal-treatment law directly. Even without deliberately targeting a protected characteristic, platform optimisation can produce indirect discrimination. See AI and discrimination in recruitment: responsibility lies with the employer, even when the platform performs the targeting.
What the GDPR and DSA add
Targeting is profiling, and thus processing of personal data with a valid basis and transparency (GDPR). The Digital Services Act also bans ads based on special-category data and profiling-based ads to minors, and requires platforms to make ads recognisable. If you recruit through large platforms, these rules reach you via the advertising terms.
What to do
- Treat job-ad targeting as high-risk โ with human oversight and monitoring for skewed outcomes.
- Target broadly, not narrowly: avoid demographic narrowing and don't let the platform optimise unchecked.
- Check the reach figures for unintended exclusion (age, gender, region).
- Record the basis and the trade-offs โ you need them in a complaint.
- Combine with your broader recruitment approach and CV screening; the whole funnel falls under the same regime.
The fairest selection starts with fair reach. An algorithm that decides who sees your vacancy helps decide who can be hired โ which is exactly why the law is strict about it.
Sources
- https://eur-lex.europa.eu/eli/reg/2024/1689/oj
Regulation (EU) 2024/1689 (AI Act): Annex III, point 4 explicitly names the targeted placement of job advertisements as high-risk. - https://eur-lex.europa.eu/eli/reg/2022/2065/oj
Regulation (EU) 2022/2065 (DSA): advertising transparency and limits on profiling-based ads.
Read next
AI in housing allocation: access to essential services
AI that decides who gets access to housing strikes at the core of the high-risk regime. Annex III of the AI Act covers access to essential private and public services; on top of that, the GDPR prohibits discrimination and sets demands on automated decisions.
AI in onboarding and internal mobility: where is the line?
Talent marketplaces, skills matching and career paths with AI seem neutral, but they reach the high-risk line as soon as they steer promotion or progression decisions (Annex III, point 4). Then the AI Act, GDPR, transparency and equal opportunity apply internally too.
AI fraud detection by government: the lessons after SyRI
After the SyRI ruling (District Court of The Hague, 2020) and the Dutch childcare-benefits scandal, government fraud detection with AI is high-risk under Annex III. The lessons: no opaque risk scores, no proxy discrimination, but proportionality, explainability and a rights assessment.