EU ‘gig worker’ rules look to rein in algorithmic management

Backers of the proposals argue that all workers should be protected from management by automated systems and want companies to be more transparent about what they’re doing.

no. 9 ethical ai artificial intelligence algorithms by laremenko getty images
laremenko / Getty Images

Companies who employ “gig” workers would have to provide greater transparency about the use of algorithmic management and monitoring under rules proposed by the European Commission last week. The rules, if they go into effect, should be expanded to all workers subject to management by automated systems, according to workplace experts.

As with other European Union legislation, the rules would affect US companies with gig workers in the region.

The Commission’s directive proposals on algorithmic management were one of three sets of measures announced last week, including the provision of employee status for those working with “digital labor platforms.” That group includes ride-hailing firms such as Uber and Lyft, as well as delivery companies such as Deliveroo and informal domestic service work platforms like TaskRabbit.

There are an estimated 28 million gig workers in the EU, the Commission said, a number that’s expected to rise to 43 million in 2025.

The algorithmic management proposals seek to provide workers with greater protection from any negative effects of the automated systems used to control aspects of gig work.

“This is a step forward in the awareness of the risks that are engendered by algorithmic management, and a more attentive approach than what has come from the Commission so far,” said Valerio De Stefano, professor of labor law at Belgium’s KU Leuven university.

Algorithmic management explained

Algorithmic management is a key component of digital labor platforms. It involves tools and techniques used to automate aspects of worker coordination, such as allocating tasks and tracking performance, and relies on data collection and surveillance instead of human supervision.

However, the use of algorithmic management has raised concerns around the ability for “black box” systems to monitor and evaluate workers, with little opportunity for workers to contest decisions.

A report this week from the Worker Info Exchange, a nonprofit group, highlighted the prevalence of gig-worker surveillance, and the lack of recourse workers have when issues arise: ride-hailing service drivers being locked out of accounts because of facial recognition software failures, is one example.

The detrimental effects of algorithmic management and monitoring of platform workers were highlighted in a report from UK MPs last month. “Pervasive monitoring and target-setting technologies, in particular, are associated with pronounced negative impacts on mental and physical wellbeing as workers experience the extreme pressure of constant, real-time micro-management and automated assessment,” said members of the All Party Parliamentary group in their report, “The New Frontier: Artificial Intelligence at Work.”

The automated distribution of tasks and measurement of performance is not inherently problematic, said Helen Poitevin, VP analyst at Gartner, but problems can arise with an overreliance on the technology and too little human interaction. “It's when you don't have any support system to help [workers] grow and learn, or have a place to go to if the machines are wrong and you need to be able to raise the flag,” she said. “You have to have the support in place.”

The Commission’s proposals center on greater transparency around the use of algorithmic management tools for gig workers. This includes:

  • Informing workers how automated systems are used to make decisions and take actions that “significantly affect” working conditions.
  • Limits on the collection of worker data that’s not strictly necessary, particularly around private conversations or data that relates to health or emotional state.
  • Human oversight of automated systems. This involves assessments to ensure that workers do not face psychological and health risks caused by algorithmic decision-making.
  • Providing the opportunity for workers to challenge automated decisions that affect their job. This has been an issue for gig workers that have faced account termination or suspension without being told why.
  • A requirement for platform providers to consult with worker representatives about major changes to decision-making systems.

The Commission’s proposals build on the EU’s existing General Data Protection Regulation (GDPR) legislation, which includes protections for workers around data collection. However, the GDPR doesn’t provide the same clarity around algorithmic management as the rules proposed for gig workers, said De Stefano, and the Commission notes difficulties faced by gig workers in asserting their individual rights within the GDPR.

Beyond just gig workers?

Jeremias Adams-Prassl, professor of law at the UK’s University of Oxford, welcomed the proposed rules around automated decision-making. “A lot of the debate has been on the employment status of platform workers and the directive looks at that, but it also goes beyond,” said Adams-Prassl, calling the proposals “a really important first step.”

He argued that the scope of the rules should be broadened to all workers now managed and monitored by automated systems – not just those in gig economy jobs. “The gig economy is not necessarily a silo of the labor market: I would suggest that the legislators think hard about extending those rights to all workers, not just platform-based workers,” he said.

Algorithmic management and monitoring have been deployed by companies in a variety of sectors in recent years. Warehouse employees, for example, often face algorithmic control and monitoring, with many tracked and directed via software. And in the retail sector, automated shift scheduling software is sometimes used by employers.

More broadly, those in more traditional desk-based jobs have also been subject to growing levels of productivity tracking with the rise of monitoring tools during the COVID-19 pandemic, relying on automated screenshots and keystroke-logging, for example.

“We see similar challenges in workplaces across the socio-economic spectrum, but the gig economy was certainly the cradle of a lot of these technologies, which explains why we’re starting to see regulators respond to it in this specific context,” Adams-Prassl said. 

Said De Stefano: “The big challenge will now be to extend the protection of the [EC’s digital labor platform directive] to all workers, otherwise we will basically [have] a segmented labor market in which only platform workers are protected meaningfully.”

Setting the stage for global protections?

As with any EU law, new rules around algorithmic management could have a broader effect beyond the 27 member states.

Should the European Commission’s proposals be accepted, US-based gig work companies with a presence in the EU would need to meet regulations for works in that region. Adams-Prassl highlighted the “Brussels effect,” where EU legislation such as GDPR has an impact on other regions and countries. That could turn out to be true with rules around algorithmic management of gig workers.

“Even though they're not bound by the GDPR, there are lots of companies in the US [that are affected by it]: once you build a product to GDPR standards, you end up using it all over the world. So, we may see a strong sort of 'Brussels effect' [where the latest move] becomes the new expected standard for platforms.”

The proposals will have to be debated by the European Parliament and Council before they could take effect. There is no set timeline for implementation, and because the rules will have to be negotiated among member nations, might not be in place until 2024 or later.

Copyright © 2021 IDG Communications, Inc.

It’s time to break the ChatGPT habit
Shop Tech Products at Amazon