Hollywood’s recent battles between screenwriters and the robots threatening their jobs should have been a wake-up call for millions of European workers. Who will protect our working rights from the algorithms, which increasingly act like our bosses?
The traditional landscape of our workplaces, from factories to offices, is being dramatically transformed by the proliferation of robots, algorithms and artificial intelligence. But it is the tech giants, not workers, that seem to be reaping most of the benefits. Work tasks are increasingly fragmented or distributed geographically along the global value chain. Algorithms now monitor work activity in real time, but they bring the risk of exploitation or discrimination if unchecked. The implications challenge both our established labour norms and the pillars of our social security system.
Amazon best represents a paradigm shift in which the algorithm becomes the boss. The company’s reliance on algorithms to oversee worker productivity in its warehouses is under increasing legal scrutiny, with workers complaining that the constant use of tracking technology is dehumanising. Algorithms are king at Amazon: hiring, evaluating and firing millions of people with little or no human supervision. One worker, prior to a July strike at Amazon’s BHX4 warehouse in Coventry, said: “They can monitor you, per minute, per task – it’s micromanagement.” Not only has the company integrated automation into the very core of its vast operations, it is also embroiled in other significant legal disputes, including a major antitrust lawsuit launched in September by the US Federal Trade Commission and 17 state attorneys general.
It is sometimes argued that the rise of new “algorithmic bosses”, responsible for automating tasks such as hiring , assigning tasks, determining salaries and even layoffs, can boost efficiency, especially at a time of stagnant productivity. However, absent from our public debate is how data extraction is used to increase surveillance, which risks reinforcing racial, gender and class inequalities, or the progressive loss of the right to access the data produced by workers and collected about them.
The entertainment industry’s recent confrontations with AI shed light on similar challenges. The Writers Guild of America, in a groundbreaking agreement, made it mandatory for film and TV studios to notify writers if they incorporate AI-generated content, ensuring that the intellectual property rights of writers remain inviolable. This sets an important precedent, emphasising workers’ prerogative in determining how AI should be integrated into their work.
The Hollywood strike follows countless battles over the past decade involving delivery workers or other professions in the platform economy. Writers and journalists in the US have filed lawsuits against artificial intelligence companies that feed their work into AI engines without giving them the option to opt out, or protect their intellectual property.
Workers and their unions across Europe have also taken the algorithm to court. The ruling that has shaped all subsequent equality actions involving workers and platform technology came in February 2021, when the UK supreme court ruled that Uber drivers should be classified as “workers”, not independent, self-employed “partners”. This classification entitled drivers to benefits such as minimum wage, paid holiday and breaks. Uber’s use of algorithms to manage and control drivers was a key factor in the court’s decision.
A court in the Netherlands sided with workers earlier this year in another case against ride-hailing giants Uber and Ola. The verdict emphasised that these platforms should not withhold data on automated decision-making, worker profiling and management assessments from drivers. This highlights the importance of human supervision in decisions such as dismissals, and the fundamental right of workers to access the data that algorithms utilise to determine their fate.
In June, a Palermo court ruled that the Spanish food delivery platform Glovo, the market leader in Italy, must disclose how its algorithm allocates tasks to workers. Central to this judgment is the demand for transparency over how these algorithms function, which is pivotal in championing the rights of platform workers.
As algorithmic management spreads, urgent questions arise: who is in control of the data that powers AI systems? What role do big tech monopolies play in shaping this landscape? The rapid expanse of the digital economy is constantly testing the boundaries of traditional labour rights, exacerbating the power imbalance in society.
The European Union has a decisive role to play in ensuring that this digital transformation aligns with democratic principles, with workers’ rights and collective bargaining at its centre. The proposed EU AI Act is a world-first attempt at horizontal regulation of AI systems. It incorporates several commendable aspects, such as prohibitions on mass biometric surveillance, AI systems used to recognise emotions, gender or sexual orientation and predictive policing systems.
Although the EU’s current overall legislative setup offers a world-leading example for reining in big tech, when it comes to protecting workers’ rights there is a disconnect. Neither the AI Act nor a proposed directive on platform work explicitly considers the impact of algorithms on working conditions, nor do they foresee a ban on using algorithms for the dismissal of workers.
For a genuinely fair digital workspace, Europe needs an all-encompassing strategy, one that covers both data accessibility and the democratic control of algorithms, and considers the impact of algorithms on working conditions. The European Data Governance Act, which came into force in 2022 in its present form, provides an initial blueprint. It allows AI “black boxes” to be opened and scrutinised by interested parties, for example, via the creation of data intermediaries, service providers that allow negotiation between data holders (workers) and those who collect it (companies).
For algorithms to truly be transparent and fair, however, traditional institutions, unions and relevant public bodies must be able to access the data gathered by tech companies, all the while upholding data privacy and business confidentiality.
We need a new public and independent data intermediary that serves the general interest: a new democratic institution to organise work in the 21st century. I am thinking of a “data trust for digital workers” to give employees more control over their data, establishing legitimate uses for the information collected as well as the circumstances under which it is produced and the purposes it can be used for.
Workers and unions must be able to actively participate in shaping the governance of data and algorithms to limit their negative impact and share the benefits of AI. If workers are engaged and empowered, productivity and efficiency will also grow. In the context of the twin digital and ecological transition, a form of algorithmic governance that contributes to guaranteeing the strategic autonomy and the competitiveness of European industry, but also the advancement of workers’ rights, is more urgent than ever.
Francesca Bria is an innovation economist and expert in data and digital policy
