Hiring, workplace monitoring, automated decisions in employment contexts.
Definition & scope
The cross-jurisdiction picture below shows how each of 45 tracked instruments treats this topic. The patterns vary substantially — and 37 regimes are silent, leaving gaps that future policy work could address.
Regulatory approaches
The instruments that touch AI in employment operate through markedly different mechanisms, which the verdict-coded coverage matrix above does not distinguish. The EU AI Act regulates by ex-ante classification: AI systems used for recruitment, selection, and decisions on promotion, termination, task allocation, or performance evaluation are designated high-risk in Annex III §4, triggering provider duties (risk management, data governance, logging, conformity assessment — for most Annex III systems a self-assessment under Annex VI) and, critically for workplaces, deployer duties under Article 26: human oversight, monitoring, and a specific obligation (Art. 26(7)) to inform workers and their representatives before a high-risk system is put into use (Regulation (EU) 2024/1689, Annex III §4; Arts. 26-27). Scholars surveying the EU's data-protection, non-discrimination and social-acquis rules read this as a distinctively risk-tiered "European approach" to automated systems in high-risk workplace settings 1. A fundamental-rights impact assessment (Art. 27) is mandated mainly for public-body and public-service deployers, not private employers generally.
The United States, by contrast, governs through ex-post enforcement of pre-existing civil-rights statutes — Title VII, the ADEA, and the ADA — applied to algorithmic outputs, rather than an AI-specific statute (illustrated by the agency and litigation activity catalogued in the enforcement section above); empirical work on hiring-tool vendors shows how de-biasing claims interface awkwardly with this antidiscrimination frame 2. A third modality is procedural transparency: New York City Local Law 144 (effective 1 Jan 2023; enforcement from 5 July 2023) mandates an annual independent bias audit and public posting for automated employment decision tools, plus 10-business-day candidate notice (NYC Admin. Code §20-870 et seq.).
Key fault lines
Beyond the single related-debate link above, several substantive questions divide jurisdictions and experts. First is the architecture choice: ex-ante risk regulation (EU) versus ex-post liability under legacy civil-rights law (US sectoral) versus procedural audit mandates (NYC). The contested empirical question is whether transparency mandates actually bind: a study of 391 NYC employers found that Local Law 144's employer discretion over audit scope produced 'null compliance,' with very few firms posting the required audits 3, and qualitative interviews with practitioners found the regime had not effectively established an auditing practice 4.
A second fault line is whether algorithmic-management harms are best addressed by individual data rights or by collective/labour instruments — a body of European labour scholarship argues for worker co-determination and human-in-command oversight rather than individual remedies alone 56, an argument now partly embodied in the EU's Platform Work Directive (Directive (EU) 2024/2831). A third is doctrinal coverage: whether existing equality law already reaches algorithmic discrimination — some scholars characterise EU equality law as 'remarkably robust' yet blunted by opacity 7, while empirical socio-legal study finds anti-discrimination law structurally struggles to reach design-stage hiring harms 8. These are genuine, unsettled disagreements, not settled doctrine.
Trajectory — what is changing
The regulatory picture is moving on dated, near-term timelines. In the United States, Executive Order 14110 (which the matrix records as implicitly addressing employment via §6 and DOL guidance) was revoked on 20 January 2025 by Executive Order 14148, after which Executive Order 14179 (23 January 2025) directed agencies to review, suspend, or rescind actions taken under it; on 27 January 2025 the EEOC removed from its website the May 2023 technical guidance on applying anti-discrimination law to AI hiring tools (Exec. Order 14148, 90 Fed. Reg.; Exec. Order 14179, 90 Fed. Reg.; EEOC website removal, Jan. 2025). The federal posture has thus shifted from active guidance toward case-by-case statutory enforcement, leaving litigation and state/city measures as the principal active levers — a brittle footing given evidence that the city audit experiment failed to bind 3.
In the EU, two developments bear on workplaces. The Platform Work Directive (Directive (EU) 2024/2831) entered into force on 1 December 2024, with Member-State transposition due by 2 December 2026; it requires human oversight of algorithmic management, periodic review of automated systems, and prohibits purely automated dismissal — provisions that labour scholars assessing the Directive against Fairwork evidence read as a partial move toward enforceable algorithmic-management rights 9. Separately, the AI Act's high-risk obligations — which govern Annex III §4 employment systems — were deferred under the Digital Omnibus: following a provisional agreement on 7 May 2026, the application date for standalone Annex III obligations moved from 2 August 2026 to 2 December 2027 (European Commission Digital Omnibus, 19 Nov. 2025; provisional agreement 7 May 2026), pushing back the binding compliance milestone for AI hiring and workforce-management tools.
Coverage across jurisdictions
Historical primacy & cross-jurisdiction tension
First addressed by UNESCO Recommendation on the Ethics of Artificial Intelligence on (governs). Subsequent regimes have either codified, diverged from, or remained silent on this baseline.
- Forum-shoppingEU AI Act↔Executive Order 14179 — Removing Barriers to American Leadership in AI
- Forum-shoppingUNESCO Recommendation on the Ethics of Artificial Intelligence↔Interim Measures for Generative AI Service Management
- Forum-shoppingDirective (EU) 2024/2831 on improving working conditions in platform work↔G7 Hiroshima AI Process Code of Conduct
Compare jurisdictions: EU vs US · EU vs UK · EU vs CN
Enforcement & impact
Silent regimes — gap signal
Instruments that do not address AI in Employment — candidates for future policy work.
- Executive Order 14179 — Removing Barriers to American Leadership in AIUS
- Interim Measures for Generative AI Service ManagementCN
- G7 Hiroshima AI Process Code of ConductG7
- OECD AI Principles (Recommendation)OECD
- UN GA Resolution on Safe, Secure, Trustworthy AIUN
- NIST AI Risk Management FrameworkUS
- Bletchley Declaration on AI Safetyglobal
- Seoul Declaration on Safe, Innovative and Inclusive AIglobal
- NIST AI RMF Generative AI ProfileUS
- California SB-1047: Safe and Secure Innovation for Frontier AI Models ActUS
- India Digital Personal Data Protection Act + AI Advisory (MEITY)IN
- Brazil AI Bill (PL 2338/2023)BR
- ASEAN Guide on AI Governance and EthicsASEAN
- African Union Continental AI StrategyAfrican_Union
- Anthropic Responsible Scaling Policy (RSP) v2US
- OpenAI Preparedness FrameworkUS
- Google DeepMind Frontier Safety FrameworkUS
- Meta Frontier AI FrameworkUS
- UK-US AI Safety Institute Memorandum of Understandingglobal
- White House Voluntary AI CommitmentsUS
- Singapore Model AI Governance Framework for Generative AISG
- Japan METI AI Guidelines for BusinessJP
- General Data Protection Regulation (GDPR)EU
- EU General-Purpose AI Code of PracticeEU
- GSA Generative AI and Specialized Computing Infrastructure Acquisition Resource GuideUS
- DoD Responsible AI Strategy and Implementation PathwayUS
- FedRAMP AI Cloud Procurement GuidanceUS
- DFARS Subpart 252.204 (Safeguarding Covered Defense Information and Cyber Incident Reporting)US
- California SB-53: Transparency in Frontier Artificial Intelligence Act (TFAIA)US
- California SB 243: Companion ChatbotsUS
- California SB 942: AI Transparency ActUS
- Revised Product Liability Directive (Directive (EU) 2024/2853)EU
- Provisions on the Administration of Deep Synthesis of Internet Information ServicesCN
- New York RAISE Act: Responsible AI Safety and Education ActUS
- TAKE IT DOWN Act (Tools to Address Known Exploitation by Immobilizing Technological Deepfakes on Websites and Networks Act)US
- Japan AI Promotion Act (Act on the Promotion of Research, Development and Utilization of AI-Related Technologies)JP
- UN Global Digital CompactUN
See also
Further reading
13 academic & grey-literature sources bearing on this topic — catalogued metadata with a primary link; one-line findings are ✦ AI-generated summaries, labeled as such (charter §7.9). Browse the full literature index.
- Fair Work for Platform Workers: Lessons from the EU Directive and Beyond Peer-reviewed✦ AIAnalyzes the 2024 EU Platform Work Directive through Fairwork evidence, assessing its employment-status and algorithmic-management provisions and charting a path toward a proposed ILO platform-work Convention.
- Algorithm-facilitated discrimination: a socio-legal study of the use by employers of artificial intelligence hiring systems Peer-reviewed✦ AIEmpirical socio-legal study of employer AI hiring systems showing how design and deployment choices generate discrimination that current anti-discrimination law struggles to reach.
- Null Compliance: NYC Local Law 144 and the Challenges of Algorithm Accountability Peer-reviewed✦ AIField study of 391 NYC employers under LL 144: only 18 posted bias-audit reports; employer discretion over scope yields "null compliance", blunting the first AEDT bias-audit mandate.
- Auditing Work: Exploring the New York City Algorithmic Bias Audit Regime Peer-reviewed✦ AIFrom qualitative interviews with 16 experts and practitioners, finds "LL 144 has not effectively established an auditing regime": undefined key terms, auditor data-access barriers, contested auditor roles.
- Regulating algorithmic management: A blueprint Peer-reviewed✦ AIIdentifies regulatory gaps from algorithmic management (privacy harms, information asymmetries, loss of human agency) and sets out a concrete policy blueprint to address them.
- Algorithmic discrimination at work Peer-reviewed✦ AIArgues existing European equality law is 'remarkably robust' against algorithmic management discrimination but that opacity and enforcement gaps blunt its effect, mapping where reform is needed.
- Algorithmic management and collective bargaining Peer-reviewed✦ AIArgues collective bargaining and worker co-determination, not just individual data rights, are essential governance tools for regulating AI-driven algorithmic management at work.
- Regulating Algorithms at Work: Lessons for a 'European Approach to Artificial Intelligence' Peer-reviewed✦ AISurveys EU data-protection, non-discrimination and social-acquis rules for governing "automated systems in high-risk settings such as the workplace", drawing lessons for the proposed EU AI Act.
- Big Data in the workplace: Privacy Due Diligence as a human rights-based approach to employee privacy protection Peer-reviewed✦ AIProposes 'privacy due diligence' as a human-rights-based regulatory approach to algorithmic management and worker monitoring, arguing data-protection law alone inadequately constrains employer surveillance.
- Challenging Biased Hiring Algorithms Peer-reviewed✦ AIEvaluates UK equality and data-protection law against algorithmic hiring tools and proposes a 'transparent recruitment scheme' incentivizing publication of equality metrics from data-protection impact assessments.
- Mitigating Bias in Algorithmic Hiring: Evaluating Claims and Practices Peer-reviewed✦ AISurvey of algorithmic employment-assessment vendors' bias-mitigation claims, examining how "algorithmic de-biasing techniques interface with, and create challenges for, antidiscrimination law".
- "Negotiating the algorithm": Automation, artificial intelligence and labour protection Working paper✦ AIArgues labour law must protect worker dignity under algorithmic management, urging a "human-in-command approach" with social partners governing automation.
- The future of employment: How susceptible are jobs to computerisation? Peer-reviewed✦ AIEstimates computerisation probabilities for 702 occupations, finding about 47% of total US employment "at risk" — the headline figure framing displacement and retraining policy.
References
Sources cited inline in the analysis (linked from the superscript markers), then the primary instrument sources behind the classifications.
- Jeremias Adams-Prassl (2022) Regulating Algorithms at Work: Lessons for a 'European Approach to Artificial Intelligence', European Labour Law Journal. 10.1177/20319525211062558 — Surveys EU data-protection, non-discrimination and social-acquis rules for governing "automated systems in high-risk settings such as the workplace", drawing lessons for the proposed EU AI Act. ↩
- Manish Raghavan, Solon Barocas, Jon Kleinberg, Karen Levy (2020) Mitigating Bias in Algorithmic Hiring: Evaluating Claims and Practices, ACM FAT* '20. 10.1145/3351095.3372828 — Survey of algorithmic employment-assessment vendors' bias-mitigation claims, examining how "algorithmic de-biasing techniques interface with, and create challenges for, antidiscrimination law". ↩
- Lucas Wright, Roxana Mika Muenster, Briana Vecchione, Tianyao Qu, Pika (Senhuang) Cai, COMM/INFO 2450 Student Investigators, Jacob Metcalf, J. Nathan Matias (2024) Null Compliance: NYC Local Law 144 and the Challenges of Algorithm Accountability, ACM FAccT '24. 10.1145/3630106.3658998 — Field study of 391 NYC employers under LL 144: only 18 posted bias-audit reports; employer discretion over scope yields "null compliance", blunting the first AEDT bias-audit mandate. ↩
- Lara Groves, Jacob Metcalf, Alayna Kennedy, Briana Vecchione, Andrew Strait (2024) Auditing Work: Exploring the New York City Algorithmic Bias Audit Regime, ACM FAccT '24. 10.1145/3630106.3658959 — From qualitative interviews with 16 experts and practitioners, finds "LL 144 has not effectively established an auditing regime": undefined key terms, auditor data-access barriers, contested auditor roles. ↩
- Valerio De Stefano and Simon Taes (2023) Algorithmic management and collective bargaining, Transfer: European Review of Labour and Research. 10.1177/10242589221141055 — Argues collective bargaining and worker co-determination, not just individual data rights, are essential governance tools for regulating AI-driven algorithmic management at work. ↩
- Jeremias Adams-Prassl, Halefom Abraha, Aislinn Kelly-Lyth, Michael 'Six' Silberman and Sangh Rakshita (2023) Regulating algorithmic management: A blueprint, European Labour Law Journal. 10.1177/20319525231167299 — Identifies regulatory gaps from algorithmic management (privacy harms, information asymmetries, loss of human agency) and sets out a concrete policy blueprint to address them. ↩
- Aislinn Kelly-Lyth (2023) Algorithmic discrimination at work, European Labour Law Journal. 10.1177/20319525231167300 — Argues existing European equality law is 'remarkably robust' against algorithmic management discrimination but that opacity and enforcement gaps blunt its effect, mapping where reform is needed. ↩
- Natalie Sheard (2025) Algorithm-facilitated discrimination: a socio-legal study of the use by employers of artificial intelligence hiring systems, Journal of Law and Society. 10.1111/jols.12535 — Empirical socio-legal study of employer AI hiring systems showing how design and deployment choices generate discrimination that current anti-discrimination law struggles to reach. ↩
- Sandra Fredman, Darcy Du Toit, Alessio Bertolini, Jonas Valente and Mark Graham (2025) Fair Work for Platform Workers: Lessons from the EU Directive and Beyond, Industrial Law Journal. 10.1093/indlaw/dwaf018 — Analyzes the 2024 EU Platform Work Directive through Fairwork evidence, assessing its employment-status and algorithmic-management provisions and charting a path toward a proposed ILO platform-work Convention. ↩
- EU-AIA-2024: Annex III §4 (high-risk: employment management)
- US-EO-14110: §6 + DOL guidance; sectoral
- UK-WHITEPAPER-2023: ICO + EHRC remit
- COE-AI-CONV: Non-discrimination + dignity provisions
- OMB-M-24-10: Attachment 1 examples include employment + benefits decisions as rights-impacting; minimum practices apply
- UNESCO-AI-ETHICS-2021: Policy Area 'Economy and Labour', para 116 — Member States to assess and address AI's impact on labour markets
- EU-PWD-2024: Directive (EU) 2024/2831, Chapter III (esp. Arts. 7-11) and Chapter II (employment-status presumption)
- IT-AILAW-2025: Art. 11 — workplace AI must be safe, reliable, transparent, non-discriminatory and not contrary to human dignity; employer must inform the worker of AI use (per Art. 1-bis D.Lgs. 152/1997). Art. 12 establishes a national Observatory on workplace AI.
Cite this article 8 formats · BibTeX, RIS, APA, Chicago, … · 1-click copy
Persistent identifier: https://policywindow.org/wiki/employment — committed-stable URL with content-versioning via ?asOf= (rollout pending per methodology §7). DOIs via Zenodo are on the roadmap.
Article tools — track changes, suggest an edit
View history — every captured revision of this article · What links here
8 instruments tracked.
Does governance work? — the social-science evidence
What the peer-reviewed social science shows: whether the harm this topic addresses is empirically real, and whether governance of it works. The badge is the epistemic status of the evidence(not the policy debate) — “thin” or “absent” efficacy evidence is itself a finding (the “second silence”). Each epistemic-status label is Policy Window's editorial assessment of the cited evidence base (a structured classification), not a verdict any single source issues.
Discrimination and adverse outcomes in employment decisions are empirically well-established, and AI systems demonstrably reproduce them. The foundational field-experiment literature shows robust human baseline discrimination (Bertrand & Mullainathan 2004 found White-sounding names received 50% more callbacks), and AI-specific audits confirm the pattern: Amazon scrapped a recruiting tool that penalized resumes containing 'women's' (Dastin 2018), and a controlled resume-screening audit of language-model retrieval found systems favored White-associated names ~85% of the time and never preferred Black male-associated over White male-associated names (Wilson & Caliskan 2024). On the monitoring side, a meta-analysis (k=94, N≈23,461) found electronic performance monitoring reliably raises worker stress with no evidence of improved performance (Ravid et al. 2023). Honest caveat: measured disparities are highly model-, prompt-, and context-dependent, and most evidence comes from controlled audits and one firm's internal test rather than measured outcomes in live, at-scale hiring pipelines.
Sources: Bertrand & Mullainathan 2004 (American Economic Review 94(4):991-1013); Wilson & Caliskan 2024 (AAAI/ACM AIES; 'Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrieval'); Dastin 2018 (Reuters, 'Amazon scraps secret AI recruiting tool that showed bias against women'); Ravid, White, Tomczak & Behrend 2023 (Personnel Psychology 76:5-40)
There is no rigorous evidence that governing AI in employment reduces the documented harms; the central evaluated regime appears to fail at the compliance stage before any impact on bias can occur. NYC Local Law 144 — the first jurisdiction worldwide to mandate independent bias audits and public posting for automated employment decision tools — was directly studied across 391 employers and found to produce 'null compliance': the law's discretion makes it impossible to tell whether firms comply, with very few posting the required audits (Wright et al. 2024). Parallel qualitative work shows the audits themselves are undermined by missing demographic data, opaque aggregation, and 'test data' that does not reflect real use (Groves et al. 2024). No study links any AI-employment rule to a measured reduction in discriminatory hiring outcomes — the evidence that the rule works is itself missing, largely because mandated transparency artifacts (audit reports) are sparse, non-standardized, and unenforced.
Sources: Wright, Muenster, Vecchione, Metcalf & Matias et al. 2024 ('Null Compliance: NYC Local Law 144 and the Challenges of Algorithm Accountability', ACM FAccT '24); Groves, Metcalf, Kennedy, Vecchione & Strait 2024 ('Auditing Work: Exploring the New York City algorithmic bias audit regime', ACM FAccT '24); Ravid, White, Tomczak & Behrend 2023 (Personnel Psychology 76:5-40, on monitoring outcomes as the closest analogue evaluation evidence)