At a glance
On 11 November 2024, the European Union published the Platform Work Directive (EU) 2024/2831 (PWD) in the Official Journal. The PWD regulates the working conditions of people who work via digital platforms (often known as gig workers). It aims to better protect platform workers not only by regulating their status, but also by increasing transparency of algorithmic labour management using automated monitoring and decision-making systems and enhancing rights when such systems are used.
Scope of application - further than you think
The PWD applies to "digital labour platforms". This is a natural or legal person providing a service which meets the following requirements, regardless of where they are based as long as the people who carry out the platform work do so in the EU:
- It is provided, at least in part, at a distance through electronic means (website or mobile application).
- It is provided at the request of a recipient of the service.
- It involves, as a necessary and essential component, the organisation of work.
The PWD defines “platform work” as any activity organised through a digital labour platform and performed by a person within the EU for a third party. The term "persons performing platform work" includes all platform professionals, regardless of the nature of the contractual relationship or how the parties involved characterise this relationship (employment relationship or self-employment/freelancer). This means that both dependent employees and self-employed platform professionals should benefit from the protective measures of the PWD in future.
However, the main aim of the PWD in terms of labour law is to standardise the employment status of platform workers within the EU and improve their working conditions. To this end, a rebuttable presumption of an employment relationship (as opposed to self-employment) is introduced. It is up to the Member States themselves to determine the criteria which must be met for their total number and the number required for the fiction of the presumption to apply.
Data protection and algorithmic transparency obligations: more far-reaching requirements than in the GDPR and the AI Act?
Extended data protection provisions: prohibition of the processing of sensitive data (Article 7)
The PWD contains a large number of new data protection requirements that go well beyond the requirements of the GDPR. Article 7 introduces an absolute ban on the processing of certain sensitive personal data of platform professionals by means of automated monitoring systems or automated decision-making systems, which cannot be circumvented even with the consent of the platform professionals. This prohibited data includes:
- Data that allows conclusions to be drawn about the emotional or psychological state of the platform worker.
- Private conversations and behaviour while not performing or offering platform work.
- Data to predict the exercise of fundamental rights such as freedom of association or the right to strike.
- Data processing to infer sensitive information such as on race, ethnic origin, political opinion, religious beliefs, trade union membership and sexual orientation.
-
Biometric data to establish the identity of the platform worker by comparing it with multiple entries stored in a biometric database. However, biometric verification to confirm the identity of a platform worker by comparing their data with their own previously provided biometric data (1:1 comparison) is permitted, provided it complies with applicable data protection laws.
Businesses should therefore check whether they are processing "prohibited" data using automated monitoring systems or automated decision-making systems. It’s worth noting that the PWD arguably covers so-called free-text reviews submitted by customers. If customers mention emotional states or other sensitive aspects of platform professionals in reviews, this could be considered prohibited data. This means the PWD could extend to general review portals such as Google, which collect information that could allow conclusions to be drawn about the behaviour or condition of professionals.
Article 7 of the PWD complements the GDPR and the AI Act. The AI Act prohibits autonomous emotion recognition systems in the workplace (Article 5(1)(f)) unless they have been placed on the market for medical or safety reasons. In contrast, the PWD goes further, covering all automated systems.
With regard to automated decision-making systems, the PWD complements the GDPR. Article 22 GDPR prohibits decisions based solely on automated decision-making if they have a legal or similarly significant effect on the individual unless explicit consent has been given to the processing. As the PWD does not allow such processing even with consent, it again gives platform workers wider rights than they currently enjoy under the GDPR.
Transparency requirements and algorithmic monitoring (Articles 9 and 10)
In line with the ban on the processing of certain types of data, the PWD provides for strict transparency obligations that (also) go far beyond the rules for high-risk AI systems under the AI Act. Digital labour platforms must disclose specified information about all automated systems that have an impact on the working conditions of platform workers in a transparent and clear manner. This includes:
- Functionality and criteria of the algorithms: platform professionals have the right to detailed information about how the algorithms work and the decision-making criteria.
However, Article 9 of the PWD also distinguishes between employees ("platform workers") and self-employed persons ("persons performing platform work"). Both types of worker must be informed and under Article 9 PWD, workers' representatives must be informed of all relevant systems and their characteristics prior to the use of an automated monitoring or decision-making system. Article 14 of the PWD and Recital 53 specify that if there are no workers' representatives, the digital labour platforms must inform the workers directly.
In addition to the transparency obligations, the PWD requires regular human monitoring of all automated systems that affect platform workers. Automated and autonomous monitoring and decision-making systems should be and remain human-centered. A review must take place at least every two years and be specifically focused on discrimination risks and other negative effects. For digital labour platforms, this means that they must provide qualified staff with the ability to scrutinise automated decisions and the authority to adjust or revoke them if necessary.
Data protection impact assessment (Article 8)
Digital labour platforms must carry out a data protection impact assessment (DPIA) to comprehensively assess the impact of their automated systems on the rights and freedoms of platform professionals. This means they must evaluate all systems used without exception and make appropriate adjustments in order to meet the high protection standards provided for in the PWD. Unlike the AI Act, there are no exemptions and unlike the GDPR, digital labour platforms must even consult the platform professionals about their plans.
Right to human review and explanation of automated decisions (Article 11)
Article 11 of the PWD gives platform professionals the right to human review and explanation of automated decisions that significantly affect their employment relationship, such as dismissals or contract terminations. Digital labour platforms are obliged to provide a written explanation of such decisions and give those affected the opportunity to challenge and rectify them.
However, Article 11(5) of the PWD imposes a restriction on this principle: Article 11 of the PWD does not apply to platform professionals who are "business users" within the meaning of Article 2(1) of Regulation (EU) 2019/1150 (Platform to Business Regulation).
Special rules for representatives of platform professionals (Article 15)
Representatives for platform workers must be able to exercise the rights of workers' representatives with regard to the protection of personal data whether or not they are ‘workers’ representatives’. This includes Articles 8 (1), 9 (1) and (4) as well as Article 10 (4) and Article 11 (2) of the PWD.
When will the PWD apply?
The PWD was signed on 23 October 2024 and enters into force on 1 December 2024, twenty days after its publication. Member States have until 2 December 2026 to transpose the provisions into national law.
What does the PWD mean for you?
The PWD introduces a new era of regulation for digital platform work in the EU. It extends the protection of platform professionals well beyond the existing standards of the GDPR and the AI Act in relation to the use of automated monitoring and decision-making and sets clear requirements for data processing and algorithmic transparency in the area of platform work as well as rules for the classification of platform professionals.
Particular attention will need to be paid to which criteria the national legislators will introduce for the presumption of employee status and whether business models will need to be changed as a result. Companies with Europe-wide operations in particular must keep an eye on the respective national implementations.
Somewhat surprisingly, the PWD gives more extensive data protection rights to platform professionals than are currently given to other workers although it’s possible that Member States will introduce national legislation which would close this gap. It is also interesting that it extends provisions under the AI Act which was progressing through the legislative process at a similar time to the PWD.
The PWD obliges digital labour platforms to make comprehensive adjustments to data protection and significantly increases their transparency obligations. These platforms not only have to disclose their algorithms and how they work but also ensure that no sensitive data - including on emotional state and from private conversations - is processed, adding to the already complex regulatory burden.
