EU Adopts 2024 AI Act and Platform Work Rules to Curb High-Risk Workplace Algorithms
Updated
Updated · OnLabor · Nov 15
EU Adopts 2024 AI Act and Platform Work Rules to Curb High-Risk Workplace Algorithms
3 articles · Updated · OnLabor · Nov 15
Summary
The EU’s 2024 AI Act and Platform Work Directive add new limits on algorithmic management, extending workplace protections beyond the GDPR as AI tools spread from gig platforms into factories, warehouses and offices.
Under the AI Act, AI used for employer functions is generally treated as high-risk, triggering requirements on risk management, transparency, human oversight, accuracy and cybersecurity before deployment.
The Platform Work Directive goes further for workers subject to automated decisions, granting rights to explanation, human review and rectification, while banning monitoring of private conversations and tone-of-voice analysis for performance assessment.
A 2023 EU court ruling broadened what counts as automated decision-making by finding human involvement insufficient if it has no qualitative impact, reinforcing scrutiny of practices such as promotion sorting and algo-firing.
Enforcement is already taking shape: Italian regulators fined Foodinho and Deliveroo about 2.5 million euros in 2021, and a Dutch appeals court in 2023 found Uber’s review of driver deactivations merely symbolic.
Will the EU's strict workplace AI rules truly protect employees, or just force companies to hide algorithmic management behind fake human oversight?
Could the EU's push for human oversight in AI hiring inadvertently bring back the very human biases these algorithms were built to eliminate?
How can multinational employers mathematically prove to EU regulators that human review of automated HR decisions is meaningful rather than merely symbolic?