March 2024: a significant advancement in the European Union’s legal framework on digital technologies

15 April 2024 By

March has been a significant month for advancing the European Union’s legal framework on digital technologies. After over two years of intense and uncertain negotiations, the European Union approved a directive aimed at improving working conditions in platform work on March 11th, commonly referred to as the Platform Work Directive (PWD). Just a few hours later, on March 13th, the Artificial Intelligence Act (AIA) passed its final legislative step. These two initiatives join previous legislations such as the Digital Market Act (DMA) and the Digital Service Act (DSA).

The first is a directive, so it fixes some aims that the single States have to implement into their own legal codes. In this case, the PWD addresses for the first time at the European level an issue that has been really sensitive in the last years: which is the legal status of platform workers and which protections it implies. The directive establishes that the relationship between a digital labour platform and a person performing platform work shall be legally presumed to be an employment relationship when facts indicating control and direction are present. However, the criteria and procedures for potential reclassification have not yet been decided; this issue will be managed at the national level. Moreover, the directive addresses some rules on enforcement, transparency, and traceability of data and algorithmic management.

The second is an act, meaning its instructions are immediately effective at the European level. In this case, the AIA establishes four levels of risk for AI systems: unacceptable, high, limited and minimal. Each level is then associated with a set of obligations to be fulfilled if developers want to carry on and commercialize their innovations in the sector.

These initiatives are undoubtedly two milestones on the road to the regulation of digital technologies and their impact on labour and society. According to the European Council, in 2022 over 28 million people in the EU worked through one (or more) of these digital labour platforms.​ In 2025,​ that number is expected to reach 43 million people.​ The AI Act is part of a wider package of policy measures to support the development of trustworthy AI, which also includes the AI Innovation Package and the Coordinated Plan on AI.

Nevertheless, they also show some important limits that are to be addressed.

The PWD is the tumultuous output of a difficult negotiation process between the Commission, the Council and the Parliament. Some Member States opposed openly the final text or delayed repeatedly the definition of an agreement. Lobbying pressure by platforms undoubtedly played a role in this game. In this regard, it cannot be forgotten that at the end of February, Amazon lobbyists were banned from the European Parliament. All these elements contribute to understanding the vague solution adopted, without achieving the final goal to homogenise labour conditions in Europe.

The AIA deals with the innovations brought by Large Language Models (LLM) trying to frame all potential risks but a clear evaluation is still difficult to achieve, both because of the novelty of such tech and of the harsh debate between chauvinists and catastrophists. In any case, the great limit in this sector, at the moment, seems to be the inadequacy of EU investments in front of the budget at the disposal of other global competitors. 

More generally, the European Union is championing the definition of a set of rules for digital innovation. However, this strategy, mainly focused on preventing the free market from defining oligopolies, appears to be more of a response to a delay in the capacity to autonomously develop digital technologies. This delay condemns policymakers and interest groups to intervene only after the operations of these companies are already deeply embedded into society. Starting from this consideration, in the INCA project, we are asking ourselves how it is possible to articulate policies that do not only limit themselves to (though necessary) regulation. We believe it is fundamental to open a field of political research that can also envision constituent forms in Europe for innovative designs of platforms alternative to the current infrastructural power expressed by the GAFAM.