European Compliance Suite

Directive (EU) 2024/2831 on improving working conditions in platform work entered into force on 1 December 2024. Member states must transpose its provisions into national law by 2 December 2026. It establishes a rebuttable legal presumption of employment for platform workers where the platform exercises direction and control, algorithmic management transparency and human oversight requirements, and data protection restrictions for platform worker data. It is the first EU-wide regulation of algorithmic management in the workplace and directly addresses the use of AI in gig economy workforce management.

A digital labour platform is any commercial service provided at least in part at a distance and through electronic means, on the request of a recipient, that involves the organisation of work performed by individuals in exchange for payment. The definition is broad and covers ride-hailing, food delivery, freelance marketplaces, online staffing platforms, and any other service that uses technology to match workers with tasks and manages those workers through algorithmic systems. Platforms that consider themselves software tools rather than labour platforms should assess the definition carefully — the scope is wider than most expect.

The employment presumption is triggered where the platform exercises direction and control over workers. Indicators of direction and control include algorithmic task assignment, algorithmic pricing or earnings determination, algorithmic performance monitoring, behavioural constraints communicated through the platform, and algorithmic account restriction or termination.

Where these indicators are present, the presumption arises and the burden shifts to the platform to demonstrate the absence of an employment relationship. AI platforms that manage workers through automated systems are almost always exercising direction and control within the meaning of the Directive.

Platforms must inform workers about the automated monitoring and decision-making systems that affect their working conditions — including what data is collected, how automated systems use that data, what the main parameters of algorithmic decisions are, and how those parameters affect individual workers. The information must be provided in plain language accessible to non-technical workers. Platforms must also inform worker representatives about automated systems and provide access to information necessary for collective consultation before implementing or significantly changing those systems.

Yes. The Directive prohibits automated systems from making decisions to restrict or terminate a worker’s account — effectively firing them — without human oversight. Consequential decisions affecting workers must involve genuine human review with actual authority to override the automated system.

A human reviewer who approves AI recommendations without meaningful individual assessment does not satisfy the requirement. The same principle applies to other significant decisions affecting workers — account restriction, performance assessment, and access limitations all require genuine human oversight under the Directive.

AI systems used in employment decisions — including recruitment, task assignment, performance monitoring, and access restriction — are classified as high-risk under EU AI Act Annex III. High-risk AI Act obligations — technical documentation, conformity assessment, human oversight, and accuracy requirements — apply alongside Platform Work Directive algorithmic management obligations to the same systems simultaneously. A cross-framework assessment identifies where EU AI Act human oversight requirements satisfy Platform Work Directive requirements and where the two frameworks impose distinct and non-overlapping obligations on the same system.

Yes. The Directive applies wherever platform workers are located in the EU, regardless of where the platform is established. A US ride-hailing platform with EU drivers, a Canadian HR technology company managing EU gig workers, or an Indian staffing platform operating in EU markets faces the same employment presumption, algorithmic transparency, and human oversight obligations as an EU-established platform. The Directive follows the worker’s location, not the platform’s incorporation.

Platforms are prohibited from processing sensitive personal data through algorithmic management systems — including biometric data, health data, data revealing racial or ethnic origin, political opinions, religious beliefs, and trade union membership.

Platforms may not use automated systems to infer sensitive personal data from non-sensitive inputs — the same prohibition that applies under GDPR Article 9 and that is reinforced by the Platform Work Directive for the specific context of algorithmic workforce management.

Worker monitoring data may only be used for the purposes for which it was collected — purpose limitation applies with particular force to AI systems that could repurpose monitoring data for model training or commercial profiling.

Penalties are set at member state level as part of national transposition — specific penalty amounts vary by jurisdiction. Member state labour authorities have enforcement powers covering employment presumption, algorithmic transparency, and data protection violations. GDPR supervisory authorities retain enforcement powers over data protection violations that also breach GDPR regardless of Platform Work Directive status.

Platforms found to have triggered the employment presumption and failed to rebut it face reclassification of workers as employees — with retrospective liability for employment taxes, social security contributions, and worker benefits that could significantly exceed any direct regulatory fine.

Worker data processed through algorithmic management systems is personal data subject to GDPR. The Platform Work Directive reinforces and extends GDPR’s data protection requirements for the specific context of platform work — adding restrictions that go beyond GDPR in some areas, such as the explicit prohibition on processing sensitive data through algorithmic systems.

Both frameworks must be satisfied simultaneously — GDPR for general data protection obligations and the Platform Work Directive for platform-specific restrictions and worker rights. A cross-framework assessment maps where the two frameworks align and where the Platform Work Directive imposes obligations beyond what GDPR already requires.

First, determine whether your organisation qualifies as a digital labour platform under the Directive’s definition and whether your AI systems trigger the employment presumption through direction and control over workers.

Second, assess your algorithmic management transparency documentation against the worker information obligations — what you disclose, in what format, and whether it satisfies the plain language requirement.

Third, review your human oversight processes for consequential automated decisions — account restriction, performance assessment, and access termination — against the genuine oversight standard the Directive requires.

Fourth, assess your data processing practices against the Directive’s sensitive data prohibitions and purpose limitation requirements alongside your existing GDPR compliance programme.

A lawyer-built assessment covers all four steps and delivers a documented compliance position specific to your platform — mapped against your EU AI Act, GDPR, and NIS2 position where all apply.