Lena Simet (Human Rights Watch) on platform work: From turbo-capitalism to just working conditions

In The Gig Work Podcast by the WageIndicator Foundation, Martijn Arets talks to Lena Simet from Human Rights Watch about the downsides of platform work and ways to develop effective policy. “Technology for organizing work has developed at lightning speed, but legislation to protect workers’ rights on platforms is hopelessly behind.”

How can we ensure that platform companies in the gig economy behave as responsible employers and clients, rather than greedy intermediaries who make ever-increasing profits and pass the risks and costs of doing business on to workers? Trade unions, labor organizations, and governments around the world are looking for a solution to this problem.

This includes Human Rights Watch, an international organization that investigates human rights violations worldwide. In recent years, senior economic justice advisor Lena Simet has been specifically studying the rights impact and economic fairness of platform companies on workers. I spoke to her about her research on The Gig Work Podcast by the WageIndicator Foundation. Her conclusions provide a good overview of developments and opportunities from a global perspective. 

Legal vacuum

Simet studied the impact of taxi, food delivery, and grocery shopping apps on platform workers in Lebanon, Texas, and New York, among other places. “Technology for organizing work has developed rapidly, but legislation to protect workers’s rights on platforms is desperately lagging behind,” she says. “It’s a legal vacuum: platform workers are not formally employed, so the work and earnings are their own responsibility. Almost all the labor rights that have been fought for in the past seem to be non-existent in this business model.”

She thinks this is unfair. Her interest in platform work arose during the coronavirus crisis. “Platform workers were the heroes: they took to the streets to deliver meals or groceries, they worked in healthcare,” she says. “Everyone was happy with them, but that appreciation was not reflected in their working conditions. Many were not given face masks or hand sanitizers, and if they fell ill themselves, they received no compensation or paid leave.”

‘Working without protection should not become the new norm’

Meanwhile, the reach of platform work is growing enormously. “Platform workers are no longer just taxi drivers or food delivery workers,” she says. “Now you also see nurses, teachers, and therapists being hired on demand via apps. Instead of a permanent contract with fixed shifts, they are now deployed ‘on demand’ with varying hours and earnings.”

An increasing proportion of the global workforce is being hired and fired via platforms, she says. “This increases inequality in the labour market enormously. Our research shows that they have no protection under labor law. That is why new policy is so important. We cannot allow underpayment and lack of protection to become the new norm in the labour market.”

Employee or self-employed: decent work for everyone

Governments around the world are struggling with the legal status of platform workers: are they employees or self-employed workers? Being employed solves a lot of issues: often, job security and protections for employees are legally linked to this type of contract, but in practice, this is difficult to enforce.

In the Netherlands, too, the discussion is far from over. Just look at the latest ruling by the Amsterdam Court of Appeal on whether Uber drivers are formally employed or not. Conclusion: it varies from driver to driver. And in continents such as Asia, Africa, or Latin America, it is not at all common to have an employment contract. In fact, almost half (46%) of the global workforce are self-employed (ILO 2025).

That is why it is perhaps even more important at this point to find an answer to the question: how do we ensure that the risks and costs of self-employed people are covered just as well as those of employees? The biggest problems arise because platforms pass on the costs and risks that are borne by the employer in an employee relationship to the individual.

Platforms weaken individual bargaining power

Human Rights Watch’s research shows that action is needed. “We are seeing the consequences of a lack of regulation worldwide,” says Simet. “It is true that professional groups such as cleaners, taxi drivers, and food delivery workers did not usually work as employees even before the advent of platforms. But what has deteriorated is their bargaining power.”

She cites motorcycle taxis in Kenya as an example. “In the past, drivers set their own prices in negotiations with customers. Now, the app determines the price. Drivers no longer have any influence over this, especially since these companies often form monopolies.”

At the same time, platformization offers hope for improvement. “Platforms make workers who were previously invisible visible. If we succeed in forcing these large companies to pay workers decent wages, it will be a huge opportunity to provide millions of workers worldwide with better living standards.”

Lebanon: strong growth since 2019

When I spoke to Simet, she had just returned from Lebanon, where she had been studying the situation of platform workers. To her surprise, there had been hardly any research into the platform economy, even though the business model is growing rapidly there. “Since the economic crisis in 2019, platform work has been the only source of income for many people,” she says. “The group of workers is extremely diverse in terms of age, education level, and occupation.”

What are the consequences of platformization? Four concerning issues stood out to her:

1. Decline in income over time: She spoke to many people who have been working via platforms for a long time, sometimes as long as ten years. During that time, their income has usually declined. Many now receive only a fifth of what they used to earn. This is because there are now many more platform workers. Although prices for customers are rising, the platforms have the freedom to reduce the earnings of workers.
2. Lack of social security: Workers have to bear all costs themselves, have no sick leave, and receive no assistance in the event of accidents at work.
3. The huge gap between workers and companies: Companies are not interested in complaints. Workers can hardly unite to exert pressure.
4. Complete lack of policy. Platform work is not covered at all by current labor legislation.

Traumatic robbery

In the podcast, Simet tells the moving story of 74-year-old taxi driver Abraham. During the crisis, he lost his job, his savings, and his pension. Because of his age, most companies would not hire him, so in 2015 he started working via a taxi app.

One day, while driving, he was robbed at knifepoint by customers. They stole his phone and his car. He sought help from the platform company, but they refused to help him. After all, his contract stated that he was an “independent contractor” (self-employed), so he was entirely responsible for himself. “He was left traumatized and without a car,” says Simet. “With financial help from his family he was able to get by and eventually was gifted a car by his brother, who also helped him. He is working again, but he is still afraid every day.”

According to Simet, this story illustrates how platforms deliberately shift all costs and risks to workers with their business model. “Responsibility and humanity are lacking.” This is despite the fact that platforms, which operate in fragmented markets, could use economies of scale to improve conditions and mitigate risks. Not doing so is a conscious decision and strategy.

Exploitation in Texas

Research is the basis for creating policies around decent working conditions. What minimum protection do platform workers need? What is a Living Wage? Since platform workers are not employed, do not have fixed hours, and must arrange their own resources and security, such a tariff is structured very differently from an employee’s wage. Read more about a Living Tariff in this blog.

In May 2025, Human Rights Watch published the report The Gig Trap: Algorithmic, Wage and Labor Exploitation in Platform Work in the US. It shows that platform workers in Texas are being severely exploited.

$5.12 per hour

“It was very difficult to obtain data because companies are not required to share information about workers who are not employed,” she says. “So we collected information from the platform workers themselves. Initially, we saw a gross hourly wage of $16.90. But that is not what they actually earn from their work.”

Because platform workers have to pay for their own vehicle, phone, and internet, they are left with only $7.53.

If you then deduct the non-wage benefits that a normal employee would receive, you end up with $5.12 per hour. “That’s well below the minimum wage of $7.25 and even further below the living tariff,” says Simet. It’s important to realize that the minimum wage in Texas is not enough to live on. According to WageIndicator data, the Living Wage in Texas is currently $16.49. And keep in mind that the reported income per worker is an average. There are platform workers who, on days with high expenses and few rides, are left with virtually nothing, the big problem being that workers have no influence on the demand for work and the number of workers active on a platform.

Heartbreaking turbo-capitalism

“I found it heartbreaking to hear platform workers blame themselves for earning so little,” says Simet. “An older woman who shopped groceries for Instacart said, ‘Well, I just can’t walk fast enough.’”

She calls it capitalism on steroids. “A person’s value and income are determined solely by how quickly profit can be squeezed out of their labor,” she says. “It has nothing to do with fair compensation and creates perverse incentives that force people to risk their health.”

New York: collective action leads to fairer pay

Fortunately, Simet also sees progress. For example, app delivery workers in New York have succeeded in enforcing a minimum tariff. “It’s a wonderful example of how collective action leads to change,” she says. “The platform workers first conducted their own research to highlight the problems and presented this to the city council. The council’s own research confirmed their findings based on their own research: from extremely low pay to lack of safety and privacy violations.”

New York used this research as the basis for policy reforms. The municipality did not force companies to hire people as employees, but set a minimum tariff for platform workers to compensate for their lack of protection. Despite fierce opposition from platform companies, New York gradually introduced a minimum wage for platform workers.

Discussion about waiting time

There was considerable debate about that tariff. Platform companies argued that they cannot pay for waiting time because workers “have multiple apps open at the same time” and are therefore paid by three different platforms simultaneously.

“In reality, this ‘multi-apping’ is hugely overestimated,” says Simet. “About 80 to 90 percent of platform workers use only one app at a time. What’s more, these companies have access to all the data: they can calculate down to the second who is available when. In New York, this has now been resolved: companies must pay for the entire time that workers are connected to the app, including waiting time.”

More efficient and fairer

The result? Because platforms are now responsible for waiting time themselves, they have started to plan more efficiently. Since the introduction of the minimum rate in New York, the number of deliveries per hour has risen from 1.6 to 2.5. By placing the responsibility with the platform, the app has a direct incentive to use the worker’s time more efficiently.

The conditions for platform workers have improved enormously, says Simet. The city council is now looking at next steps, such as protecting platform workers who are banned from a platform for unclear reasons.

Global problem, global solution?

It is clear that the excesses of the platform economy are a global problem. Although local solutions are now being devised, the International Labor Organization (ILO) is working on a global solution. In June 2026, during the 114th International Labor Conference in Geneva, governments, employers, and employee organizations will work on finalizing the ILO Platform Work Convention. Lena is participating in this process on behalf of Human Rights Watch, and provided written input on a draft of the convention. 

In Geneva, global agreements will be made on platform work, with a focus on social security, transparent algorithms, and the prevention of misclassification. I will also try to attend this conference to report on these negotiations. Whatever the outcome, I believe it is already a significant achievement that governments recognize the importance of this issue and that, contrary to many people’s expectations, we have succeeded in putting it on the global agenda. After all, fair working conditions are the responsibility of us all.

In the discussion about platform work, I keep bumping into a big dilemma. Online platforms offer a fast -access solution for work and income in the short term. At the same time, they often fall short in providing good working conditions, sustainable careers, and future perspectives. In my opinion, this tension is the most important challenge for the future of work. How do we solve it?

Frida Mwangi knows all about it. She made the transition from housewife to platform worker, and then went on to become an entrepreneur and union leader. As a founding member of the Kenya Union of Gig Workers (KUGWO), she champions the rights of Kenyan platform workers. Her lessons are relevant not only for Kenya, but for the platform economy worldwide. I spoke to her for a new episode of The Gig Work Podcast by the WageIndicator Foundation during my visit to Nairobi, Kenya.

A new start

Mwangi knows from her own experience what opportunities and perils the platform economy can offer. After being a full-time motherand housewife for 17 years, she wanted to return to work. Not only to earn money, but also to set an example for her children. But without recent work experience or references, a regular job was out of reach.

Then she discovered Upwork, one of the largest international platforms for freelance work. After a short training course, she was able to start working right away. Her first job was converting audio into text (transcription). “I could work from home in my own time, which was ideal in combination with raising my children and running the household,” she says. “In the beginning, it was exhausting because it was my very first job. At the same time, it felt like confirmation: ‘Oh, this is real. And it’s something I can actually do.’ It felt like a chance for a new life.”

Learning from others

Mwangi once wanted to become a lawyer, but that didn’t happen. She was still eager to learn. She discovered all kinds of online communities where platform workers shared knowledge and experience. “I learned a lot from that, both about the work and about how to earn more,” she says. “Those communities were incredibly valuable. In no time, I had more work than I could handle. I was able to outsource my surplus work through my own small business: Kazi Remote.”

This shows that platform work can be a stepping stone to employment and self-employment. But Mwangi also quickly discovered the negative aspects.

Unilateral conditions

Firstly, working conditions and earnings could change suddenly. Initially, she earned between $15 and $20 per assignment, later rising to $100 when she specialised in legal, financial and academic transcription transcription assignments. “As more people started working via Upwork, it became more difficult to get jobs,” she says. “The problem was that you had to bid on assignments, and that system was unreliable. Some days you kept bidding without getting any work.” The work also shifted from transcribing to proofreading AI-generated transcriptions.

Then Upwork introduced a new system. Platform workers had to buy credits to bid on a job. “To maintain a secure position on the platform, you sometimes have to spend up to $45 a month on credits,” says Mwangi. “For those coming from a financially vulnerable situation, that’s a significant barrier. The platform suddenly made the workers bear all the risks.”

Exclusion and slow payments

What’s more, the algorithm could exclude you for no reason. “Sometimes you would wake up and find that your account had been blocked without warning,” she says. “Often you would be reinstated automatically, but that took a while. In the meantime, you lost income.”

Platforms did not take responsibility, she says. ‘In the beginning, PayPal was not accessible to the African region. When the service did become available, accounts were regularly closed, even though the workers’ money was still in them. And payments were sometimes delayed by months. When we had complaints, no one was available to help us.’

Internet waste and mental damage

Ironically, Frida’s activism began through an initiative of the platform itself. During an Upwork event, she met other freelancers and discovered that she wasn’t the only one with problems. She also heard dire stories from colleagues in content moderation and data labeling. This is the work where people have to remove illegal or offensive texts or videos from platforms and train algorithms to recognize this type of content.

“Many thought they were going to do translation work, but instead had to filter harmful content on a daily basis,” she says. “It was garbage, internet garbage that you had to sift through. And the more you take in, the more harmful it is to your mental health.”

‘Platforms don’t offer a career’

She also saw that while platforms offered a stepping stone to work, they didn’t really help workers progress. “If I had stayed stuck in my transcription work, I would hardly have any assignments now,” she says. “This type of work has now been largely automated. That applies to more jobs via platforms.”

Tech companies offer a low barrier to entering the workforce, but rarely offer opportunities for advancement, training, or guidance. Mwangi: “I realized that platforms don’t offer you a career, but are only suitable as a temporary place to earn money. Yet many people become dependent on them, precisely because of the lack of opportunities for advancement.”

Organized action is not easy

She also heard more and more stories about underpayment in location-based work, such as taxi services. All these stories touched her deeply and brought back an old dream: to become a lawyer. She felt a strong urge to stand up for platform workers. Mwangi: “I believe that platforms must take responsibility, both in terms of working conditions and pay, as well as in terms of long-term prospects.”

Her first attempt to set up an association in 2019 failed. “No one had any experience with organizing,” she says. “Moreover, organizing is not easy in the platform economy. Whereas in a factory hall it is easy to talk to colleagues about problems, platform workers sit alone at home. There is also a gap between the different types of work. Online freelancers feel different from Uber drivers, for example.”

But she did not give up, because she was convinced that collective action was necessary. In 2024, she succeeded: together with other platform workers, they founded the Kenya Union of Gig Workers (KUGWO). It is the first Kenyan trade union dedicated to improving working conditions, wages, and rights for all types of platform workers.

‘It’s a matter of taking responsibility

Mwangi’s vision: platforms can offer both short- and long-term benefits for workers. “It’s a choice for companies whether or not to participate in exploitation,” she says. “That doesn’t just apply to the platforms themselves. Their customers are often large Western corporations. These companies must not forget the ‘S’ (Social) in the ESG principles (Environmental, Social, and Governance).”

KUGWO is keen to work with tech companies to put the interests of workers first. A good example is the collaboration with Microsoft/LinkedIn Learning. The Kenyan union pointed out that platform workers who lost their jobs due to automation had no opportunities to improve their skills. After consultation, Microsoft offered eleven free courses (such as project manager or software developer) as a stepping stone to better work. Mwangi: “This proves that even in a complex relationship, you can find concrete and sustainable solutions.”

The power of strong unions

Finally, I spoke to Mwangi about political influence and regulation. According to her, the voice of workers in Kenya is systematically ignored by policymakers. Her appeal to the rest of the world is therefore clear: “Build stronger institutions that enable workers to exert more influence. Support them, for example with legal and technical expertise. Employers and governments already have so much power, while workers are in a weak position.”

Mwangi emphasizes that you need financial independence and a strong membership base to be able to negotiate at all. She knows from experience how difficult that is. Nevertheless, with her resilience and perseverance, she has already achieved a lot.

Finally: is it a dilemma?

Mwangi’s call echoes earlier conversations I had, such as with Ephantus Kanyugi of the Kenyan Data Labelers Association. This is not an official trade union, which is precisely why it is fast and flexible. Mwangi chose a different route: establishing a formal union, with all the bureaucracy and political dynamics that entails. In practice, they are complementary. They have different strategies but a shared goal: better working conditions and pay for platform workers.

I agree with Kanyugi and Mwangi: what is needed in the short term and what is important in the long term must go hand in hand. Quick and easy access to work, with security and future prospects. Especially when the clients are companies, they must take responsibility and not shift it onto individual workers. Clients and platforms must choose: do they contribute to exploitation, or do they help build prospects for workers worldwide?

Last week, I had the opportunity to contribute to a seminar organised by the International Society for Labour and Social Security Law in collaboration with the Levenbach Institute. The theme was “The role of social partners in the use of AI at work”. Following contributions about experiences in the Netherlands, Belgium and Europe, I was asked to conclude with some reflections and to lead a workshop. Here are a few takeaways and thoughts:

• The impact of technology on labour is not new; we can learn a lot (as previous speakers mentioned) by looking at past experiences.
• AI and work is often not about replacement, but about the quality of – and access to – work and a growing asymmetry of power between employers/clients and workers.
• Social partners fill the gap between regulation and society, but I wonder whether the pressure becomes too high when enforcement is lacking, and the question is what skills social partners lack in order to be an equal partner in the debate;
• At the same time, social partners can really make a difference by including agreements on AI (I know: a very broad concept) in collective agreements. The only disadvantage of this is that 1) collective agreements usually apply to employees, while 46% of the working population worldwide is not employed, 2) collective agreements are often (especially when viewed globally) not public, which means that unions and sectors cannot learn from each other effectively, and 3) if you look at worker protests in the gig economy (= the testing ground for AI and labour), grassroots movements are by far the largest organisers, not trade unions.
• In discussions about labour law, preference is often given to employee status, while in many cases this will result in working with subcontractors and temporary employment agencies. Yet another company taking a slice of the pie, and you know where the bill will end up. I still miss a broader discussion about the value and appreciation of work.
• We really need to think about contract-neutral regulations and protection. See, for example, this paper on the European Platform Work Directive.
• Discussions about AI and work tend to focus on those who use AI or where AI is applied, but not on the workers and the work in the AI supply chain.

After my introduction and reflection, the attendees divided into groups to discuss the following four issues I had brought up:

1. Which stakeholder is responsible for setting up and managing a data wallet for workers: the GigCV case study.
2. How can the cooperative model leverage power in the topic of work and AI for workers?
3. How can we create a tariff floor for self-employed workers?
4. How can social partners safeguard the rights of workers in the AI supply chain in a global labour market?

All in all, it was an interesting session to attend and contribute to, and it is always great to learn more from other disciplines. Thank you to Miriam Kullmann and Matthijs van Schadewijk for the invitation and organisation, and to Mijke Houwerzijl, Juliana Londono, Simon Taes and Klara Boonstra for your inspiring presentations.

Last week, together with Tessa Duzee, I organised the event “Ghostwork, the invisible labour behind AI” at the Amsterdam University of Applied Sciences. The aim was to raise awareness of the fact that behind AI there are tens of millions of vulnerable workers who annotate and check the data, thereby keeping AI running. And to start a conversation about how we can improve these working conditions. From the perspective of the individual, from professionals in “Responsible AI”, from the Amsterdam University of Applied Sciences itself and from organisations (AI companies and their customers). This was led by moderator Tessa and contributions from experts Fiona Dragstra (WageIndicator Foundation), Nanda Piersma and myself. The data workers themselves were also given a platform through video clips, where they talked about their experiences.

It was interesting to bring together the various disciplines and engage in open discussion with the 80 students in the room. My five takeaways from this event:

1. Whether or not to exploit workers is a conscious choice. Not exploiting them is also a choice. The data work market is characterised as a to-business market, which is different from other gig markets such as taxi and delivery. And in a to-business market, organisations are responsible for their supply chain. I am looking here at both the AI companies themselves and the customers they serve.
2. In a market where organisations capitalise on fragmentation and information asymmetry, bringing people together is more important than ever. Think of trade unions and cooperatives. The key to solutions or resistance lies in finding and connecting nodes with which you can create critical mass. Consider, for example, (Dutch) organisations such as SURF and Public Spaces. But the government, as (I suspect) the largest customer of big tech and a major distributor of capital through subsidies, is also such a node. Make use of this, take responsibility and dare to make choices.
3. Creating fair(er) alternatives takes time. It is not realistic to expect alternatives to be as smooth and scalable to use as the current dominant players from day one. After all, they have a head start of years of innovation, learning and further development. Paid for from the income we as users have paid. Breaking this cycle requires us to bite the bullet, where short-term convenience and long-term sovereignty are at conflict with each other.
4. There is a lot of talk about European “champions”. Of course, I am in favour of European tech companies, but as long as nothing changes in terms of ownership and governance, there is nothing to prevent these companies from eventually being bought out by other parties or making profit-driven choices that have a negative impact on society themselves. That is why I advocate, in addition to “home-grown” tech companies, also engaging in dialogue about ownership and governance and making models such as the Steward Ownership model more common and financing for these types of models more attractive.
5. The biggest question during the event was: “What can you do as an individual?”. Firstly, I don’t think you can place the responsibility on the individual. But that doesn’t mean that you can’t do anything as an individual. Make conscious decisions, engage in conversation, listen critically to cheering stories (and keep in mind the interest of the sender of a message) and contribute to highlighting and addressing the issues that matter.

All in all, it was a great meeting, and I hope it has contributed to a better-informed debate about (responsible) AI among students, professionals and the AUAS itself.

The video of the event can be viewed via this link.

Want to know more? Then check out these two videos about data work:

Last week, the 8^th conference of the ‘International Network of Digital Labor’ took place in Bologna. This network’s mission is to research and discuss aspects of work in the digital age. I traveled to Bologna by train to attend the conference and present my research on GigCV and data portability for platform workers, which I am conducting at the Amsterdam University of Applied Sciences. During the conference, there was a lot of talk about data work(ers), the gig economy, and a broader discussion about the impact of technology on work. In this blog, I share my insights and thoughts. In my story, I choose to focus on data work and the labor behind AI. Because this issue brings together all the challenges of an imbalance of power in the world of technology, especially from the perspective of the “Big Tech” platforms and mentality.

Technology has a growing impact on how we find, perform, distribute, control, evaluate, and value work. Not only at the individual level or within the silo of an organization, but also from a geopolitical perspective. The impact on individual workers is often discussed in the (on-site and online) gig economy, but is also clearly visible in the (digital) workplace. In recent years, the development of—and discussion about—AI has been added to this. AI is not a separate silo, but a technological development within the automation of work. And it always takes place within a specific context.

How platforms are fragmenting markets

The platform model works well in fragmented markets where the costs for different stakeholders (often: supply and demand) to find each other are high. In short: markets with a high degree of information asymmetry. The promise of platforms is that, as a ‘digital message board’, they will bring clarity to markets such as (social) media, e-commerce, the sharing economy, or the labour market. As the center of the network, they have an overview and, via digital technology, they can facilitate the matching of stakeholders, create trust, and execute transactions.

The paragraph above is how I used to view this, but nowadays I am more critical. Or perhaps I should say: more realistic. I am still convinced that platforms operate in fragmented markets, but I see an important nuance in that platforms have an interest in these markets becoming and remaining more fragmented and in the number of competitors with an equal information position being as small as possible.

In the beginning, Uber broke down local taxi markets by selling services below the price of cost and being “creative/selective” in interpreting regulations. Not only to ‘capture’ market share, but also to fragment the supply in the local market and thereby strengthen its own position. A new interpretation of “divide and conquer.”

Looking back on the past 20 years, social media platforms have also fragmented the ‘market’ for social contact and the business-to-consumer market by first facilitating users with the platform and then reducing the possibilities of owning your own network. For example, my business network has slowly but surely become dependent on my contact list on LinkedIn, but the function to export this list (including contact details) suddenly disappeared. Projects or jobs are also broken down into tasks. Sometimes this is much more efficient, but it is also a way of increasing the information position and thus the company’s interests. Finally, consider platforms such as Booking, which do everything they can to maintain information asymmetry and make use of data on demand (reviews) and supply (advertisements).

In addition to local and national fragmentation, international fragmentation is also being exploited. Or perhaps better said: the use of institutional fragmentation. International regulations are absent, allowing platform and technology companies not only to pit countries and continents against each other, but also to “shop” for countries that do not ask too many questions or that have a poorly developed institutional landscape.

The result is a growing concentration of power, the increasing externalization of risks and costs to individuals and society, and increased dependence (and decreased sovereignty).

Data work

One field in which fragmentation is evident in all the areas mentioned is data work. This is the work that forms the basis of the AI we all use. Think of annotation, moderation, checking, and updating. This is a topic that came up frequently during the conference and one that I have been working on a lot lately, as you can read and listen to in the latest podcasts I made for the WageIndicator Foundation.

During the conference, the documentary “In The Belly of AI” was shown, which paints a dystopian picture of the conditions under which at least 150,000,000 data workers have to do their work. Not as an unfortunate side effect, but as a deliberate strategy. 

In Bologna, several data workers were presenting, and I spoke to some of them. Their stories are intense. People who are so changed by their work that those around them no longer recognize them. They are regularly diagnosed with PTSD, and even years after they have stopped doing this work, they still have symptoms such as insomnia, nightmares, and impaired short-term memory.

Colonial structures and concentration of power: from Big Tobacco to Big Tech

The second day took place at DAMA, where the site of a former tobacco factory now houses an ecosystem of initiatives related to AI and data. A public initiative. Although DAMA is a public initiative, the fact that it is located in a former tobacco factory is an interesting choice. You could say that both sectors, tobacco and Big Tech, have many similarities. Think of having a powerful lobby, a prime example of ‘using’ colonial structures (read: exploitation) and externalizing costs and risks to the individual and society.

The dilemma is that the impact of the tobacco industry is essentially bad and should be minimized, whereas AI, if used under the right conditions, also has many positive aspects. I would like to note, however, that if a fair price were paid for AI, many services such as ChatGPT would be available to far fewer people or, at the very least, would be used much more consciously. Which in itself would not be a bad thing. In addition, discouraging smoking through policy and individual efforts is easier than stopping AI. I am therefore not in favor of stopping AI, but I am in favor of AI that does not adopt and reinforce existing power structures. Perhaps it is naive to believe that things can be different, but ultimately everything is a choice, and making choices involves taking responsibility.

AI: good for whom?

AI gives more people access to more opportunities. More people, but by no means all people. Sarah Roberts, professor at UCLA (University of California) and author of the book ‘Behind the Screen: Content Moderation in the Shadows of Social Media’, has a clear opinion about who AI is really good for. In her presentation, she called AI a ‘systematic mechanism for labor devaluation’. She asked the (legitimate) question: who benefits from AI? Of course, individual users reap the rewards of AI, even though they are now, in a sense, data workers, training a system that skims off the value. 

To know who the real winner is, it is important to look at where the profits made with AI go. For example, AI allows workers to work more efficiently, but this will generally lead to an employer or client demanding more work from the worker in the same amount of time. This can be seen, for example, at translation agencies, but also in distribution centers, as can be read in the “Fairwork Amazon Report 2024: Transformation of the Warehouse Sector through AI.” This does not only apply to low-valued and low-paid precarious work. Because let’s be honest: would your boss allow you to spend the time you save by being more productive on vacation days for the same salary?

Regardless of where the “gains” go, you can also question the mantra that productivity gains (which many see as an exaggerated promise, with some exceptions) lead to more free time. I have come across the idea that technological change does not generally contribute to less work, historically speaking, in two books I am currently reading: ‘More Work for Mother: The Ironies of Household Technologies From the Open Hearth to the Microwave’ by Ruth Schwartz Cowan and ‘Waiting for Robots, The Hired Hands of Automation’ by Antonio Casilli.

Solutions

As I wrote earlier, I am not opposed to AI or technology. I too see the possibilities that exist and enjoy the benefits of these developments every day. What I am opposed to is the inequality that is increasing as a result of technology, the impact of these large companies on the debate and policy, and the way in which costs and risks are externalized and profits are privatized (at all costs).

Where are the solutions? Although there is no “golden bullet” that can fix everything, I believe it starts with acknowledging and recognizing the situation. Looking beyond the industry’s rhetoric and asking critical questions. To begin with, there is a lot of talk about the impact of technology. In my opinion, it should be less about the technology and more about the underlying choices, ownership, and governance structures. Technology in itself does nothing; it is the choices made by the stakeholders involved that determine its effects. The advantage of this perspective is that you can no longer hide behind ‘not being able to understand’ systems because they are too complex. The system should never take center stage, and everyone, every stakeholder, is jointly responsible.

It is also important to recognize that tech companies are businesses, not countries with democratically elected representatives. So please stop talking about democratization, because giving a few more people access while simultaneously strengthening your own unelected power has little to do with democracy. In any case, I invite you to be more critical of the words used in the discussion. Be sharper, be more critical.

Back to data workers: where are the solutions here? The advantage of the market behind data work compared to the broader gig economy market is that the buyers of data work are almost always companies, while buyers in the gig economy are almost always consumers. The advantage of companies is that it is easier to address them and hold them accountable for the choices they make. In the clothing industry, exploitation in the production chain is being combated, and this can also be done in the AI supply chain. And again: the exploitation of data workers is not an incidental coincidence, but a very conscious choice. A choice made by companies valued at many billions of euros.

In a market that revolves around fragmentation, organization is the final solution I would like to point to. I have previously researched this in relation to the cooperative model (platform cooperatives), and I also see promising initiatives in the data work sector, such as the Data Labelers Association, Turkopticon, and the Worker Info Exchange.

Many initiatives are bottom-up, and trade unions, whose core focus is on organizing workers and thereby reducing the power imbalance, are, in my opinion, still not looking enough at how they can support these workers with creative tools. For example, in Bologna, I attended a presentation by ‘Reversing.works’, which uses workers to investigate what data the worker’s platform stores, uses, and sells.

To conclude

I have said a lot in this blog, and it was sometimes challenging to keep it structured. I hope you can forgive me for that. The conference in Bologna was the trigger for writing this piece, which brings together a lot of the thinking I have done over the past few months and years.

If there is one thing I hope you will remember after reading this blog, it are three words: together, choices, and power. Together: I see too many silos in the debate, each with their own agenda and their own language, without much interest in delving into the other side. That is a shame: the only way to work towards a sustainable solution is by working together with all stakeholders. To understand why the other person does what they do. Just because you may not be able to identify with another stakeholder does not mean you should close the door. My tactic is to try not to get annoyed, but to be curious. That has helped me a lot. By being curious, however difficult it may sometimes be, you remain open-minded and keep the door open.

Choices to emphasize that everything we do is the result of decisions that are made. And decisions can be influenced. When you are aware that choices can and must be made, you are also aware of your responsibility in this regard. And finally: power. Ultimately, it is important to look beyond all the beautiful stories and cool tools and see what a development contributes to gaining or losing power. By simply asking the question: why does someone say what they say and who wins when this becomes reality? Which boils down to the advice to remain critical, without becoming bitter. A daunting task in this day and age, but nothing is impossible.

Rollee makes it easier for platform workers to access financing. Until now, it has been difficult to obtain a loan without a steady monthly income. Lenders want a clear picture of income in order to calculate a credit score. Rollee unlocks data from platforms, banks, tax portals and other relevant data sources for lenders. This should lead to fair access to financial services for all workers, says founder Ali Hamriti in this episode of The Gig Work Podcast by the WageIndicator Foundation.

Outdated system

The idea came to Hamriti, who is Moroccan-French, when he was working at a fintech company in Paris that provides loans to small businesses and the self-employed. ‘I noticed how difficult it was for freelancers to provide financial institutions with enough certainty about their income to get a loan,’ he says. “The problem is that self-employed workers and platform workers are not employees and therefore do not receive monthly payslips. Their income varies from month to month. This makes it difficult for lenders such as banks to assess their creditworthiness, and they are often denied loans.‘

He knows that not having a payslip does not mean that they are not creditworthy. ’On the contrary: in Europe, you sometimes see that people can earn more as freelancers than in salaried employment, with more freedom,” says Hamriti. ‘Yet they are often excluded from financial services because the current systems are not organised for this. Bank statements show payments and income, but banks simply have too little context for that income. This inspired me to build a new system.’

Retrieving and sharing data

The basis of Rollee is a so-called API (Application Programming Interface): software that enables two applications to communicate with each other. ‘We link to alternative data sources about work, income and taxes,’ says Hamriti. ‘I discovered that you can get more contextual data via freelance platforms and tax portals, for example.’

The result is an open platform that allows workers to collect their platform data, income data, tax data, payslips and invoices, for example. Lenders can request and analyse all this alternative data on income and work via the API platform.

Insight into platform work

In the beginning, he focused mainly on platform workers. ‘They have more access to work via digital media, but not yet to financial services,’ he explains. ‘In emerging markets such as Africa and India, current platform workers were previously invisible to lenders because they worked informally. They found their customers through word of mouth, and payment was in cash.’

But now they are increasingly working via digital platforms such as Uber and Upwork. ‘That suddenly makes their work and income transparent,’ he says. ‘If they can share that overview with lenders, it opens up opportunities for financial services such as credit.’

100 integrations

‘Our goal is to give as many people as possible access to the financial system,’ says de Hamriti. “We started by focusing on platform workers, but it is also suitable for other workers with variable incomes. With Rollee, banks can make better-informed and fairer decisions about the creditworthiness of all types of workers. It is a solution for all kinds of financial service providers: from leasing companies and banks to insurers.”

Rollee works with all kinds of companies and agencies, such as the Tax Administration and banks. They now have more than 100 integrations with freelance platforms, tax portals, payment systems and digital wallets. No direct cooperation with platforms is required, see box.

Privacy and statistics

Rollee does not store any personal data, he explains. ‘We only facilitate the transfer of data between workers and lenders. When a worker links their bank account or platform data, the data is sent directly to the selected third party. After the transfer, we only store statistical data, such as average incomes per country.’

Rollee’s mission is to make the financial system fairer. That is why they do share statistics with financial service providers. ‘This enables us to help lenders better tailor their acceptance criteria to modern workers,’ says Hamriti. ‘For example, they can adjust their criteria if it turns out that freelancers with a slightly lower income are still financially stable.’

Hamriti is also looking to collaborate with modern lenders. ‘In the long term, we want to help workers find the best interest rates and financial terms via our fintech partners,’ he says. ‘Companies such as Revolut and Monzo can make competitive offers via our platform. This allows workers to get a fairer, better deal based on their actual income rather than general credit rules.’

International cooperation

Hamriti is currently focusing on Europe and the United States. Emerging markets such as Africa, Asia and Latin America are also very interesting for Rollee. ‘We have already helped companies in Nigeria and Kenya, where digital payments are often better developed than in Europe,’ he says. “We work with banks and lenders that operate in several countries. That makes sense, as freelancers in the platform economy often work across borders.‘

Rollee is also in dialogue with governments about regulations surrounding data sharing. Hamriti sees that France and the Netherlands, for example, are rapidly improving their digital services. ‘And there is a major European project to develop a digital identity card,’ he says. ‘This will make data exchange increasingly easier.

This offers opportunities for growth for our company. After all, financial institutions need a uniform, secure API to access different data sources. We can help with that.”

Rollee versus GigCV

What makes this conversation so interesting to me is that Rollee has both similarities and differences with my own initiative: GigCV. This is an API that allows platform workers from affiliated platforms to download an overview of their work experience in the gig economy. Via an open standard, they can easily obtain an overview of their reputation and transaction data on platforms.

The more platforms share their data, the more valuable such a CV becomes. That is why I am seeking cooperation with platform entrepreneurs. In practice, it remains difficult to convince enough parties of the strategic advantage of sharing data. Moreover, they have to integrate the GigCV API. Rollee takes a different approach. This initiative does not depend on the cooperation of the platforms, because they log in on behalf of the workers. This allows them to scale up much faster. In addition, the data is immediately usable: banks already use this information to calculate risks, but the problem was that they simply did not have access to it.

I predict that platform data and data portability will become increasingly important. Regulations such as the Digital Markets Act (DMA) and the European Platform Work Directive mean that data will no longer be shared only on request, but will be available immediately and in real time. Large platforms must give their users access to data and offer free tools for data exchange, such as APIs. This gives users more control and stimulates innovation. Article 9, paragraph 6 of the Platform Work Directive confirms this right within digital labour platforms. This is good news for the future of initiatives such as GigCV and Rollee, and therefore for platform workers.

Want to know more? Listen to the full podcast with Ali Hamriti here

Artificial intelligence (AI) is and will remain dependent on human labour. The people who do the work behind AI systems are often invisible. This carries risks of poor working conditions, low wages and inadequate protection for workers. How does this situation arise, and how can we ensure that the many invisible data workers also benefit from technological developments? For the WageIndicator Foundation‘s Gig Work Podcast, I spoke with Professor Antonio Casilli (Institut Polytechnique de Paris), author of the book Waiting for Robots, the Hired Hands of Automation.

Scooby-Doo in the world of platform work

‘Me and my team are like Scooby-Doo: we travel all over the world investigating mysteries,’ says Casilli. ‘We conduct empirical research into artificial intelligence and how it is produced. Our focus is not on the new possibilities of AI, but on the development process: who is working behind the scenes to make AI possible?

His research team is called Diplab, which stands for Digital Platform Labor. They have developed a very broad view of automation.

The myth of automation

The dream of automating work is not new: Thomas Mortimer, among others, wrote in 1801 about a machine that would be capable of making human labour ‘almost completely superfluous’.

‘Technologists and economists have been looking for ways to make labour more efficient for centuries,’ says Casilli. ‘The industrial revolution saw the emergence of the first machines, such as the steam engine and the Spinning Jenny. Every innovation came with great promises. They would save us many hours of work. But nothing could be further from the truth.’

Many predictions about automation were overstated. Studies between 2013 and 2024 claimed that robots would replace 46-47% of all jobs. Casilli: ‘Organisations such as the OECD and ILO have shown that this is not true. Even with additional crises such as climate change, geopolitical tensions and a pandemic, global unemployment has not risen. In fact, in 2025, people will be working more than ever.’

The problem lies in the methodology used by these researchers, explains the professor. ‘They take a profession and break it down into tasks. If they expect AI to replace 60% of the tasks, they conclude that the job will disappear. But that’s not how it works in practice. Often, employees simply get new tasks.’ 

See also his research Waiting for robots: the ever-elusive myth of automation and the global exploitation of digital labour.

Influence of platformisation

According to Casilli, the biggest change in recent years is not automation, but platformisation. Companies such as Uber, Amazon and Meta use huge amounts of data to connect supply and demand and organise work. They also use all this data to train AI systems. For example, they build software such as ChatGPT (the P stands for ‘Pretrained’) and the technology behind self-driving cars.

‘What is often forgotten or ignored is how many people are involved in this,‘ says the researcher. “The promise of AI is that systems can take over human cognitive tasks. But in reality, many so-called ”automatic’ processes depend on human labour. The people who do this work are often invisible and poorly paid.’ This is not a recent phenomenon: Google, for example, has had its own platform, Raterhub, since 2007, where data workers verify search results and thus improve the search engine’s algorithms. Amazon Mechanical Turk, the platform used by Amazon and also available to external customers, makes a clear reference to the myth surrounding AI and its dependence on human labour. The Mechanical Turk after which the platform is named is the ‘chess robot’ invented in 1770, which travelled the world for 84 years as an example of automation. Later, it turned out that there was a person (often described as disabled or underage, in any case not a chess master) inside the machine and there was little automation involved.

Automation does not lead to less work, but to different, degraded form of work. ‘Big tech companies prefer not to talk about that. It undermines the narrative that AI is truly intelligent. In reality, people are working more than ever, but sometimes under worse conditions than before.

Check here the full interview with Antonio Casilli on YouTube

Who are these data workers?

Data workers collect, organise and improve data. Without them, AI would not work. Take image recognition, for example: AI learns what a cat is by analysing millions of images of cats. ‘People have to label those images first. It seems like simple work, but it’s a skill in itself. Yet these data workers often receive remuneration that is not commensurate with their efforts,’ says Casilli. ‘In countries such as Kenya, the monthly wage for these data workers is around $400. That’s not enough to make ends meet.’

Casilli emphasises that this is not a temporary phase. ‘Data work will remain necessary as long as we continue to develop AI,’ he says. ‘We have to constantly train the systems, adapt them to new customer requirements and check them for errors. World Bank or Oxford estimates point towards a ballpark figure of 150 million such workers worldwide, and that number is only growing. That’s another reason why it’s important to take a critical look at their working conditions.’

You are a data worker too

In his book Waiting for Robots, Antonio Casilli mentions a group of digital workers who are often overlooked: social network labourers. This basically includes everyone with a smartphone. Through our daily online activities, we train the AI of large tech companies. We teach AI what a traffic light is by filling in ReCaptchas. When we like social media posts, we teach systems which images are attractive. So we provide value to AI systems, but we are usually not paid for it. We are both users and producers of data. This raises an interesting question: is this work or not?

Casilli ziet dat deze vorm van arbeid bestaande machtsstructuren en scheve arbeidsverhoudingen versterkt. Hij en zijn team hebben samengewerkt met beleidsmakers en vakbonden om dit aan het licht te brengen. “Tech-ingenieurs bij bedrijven zoals Google verdienen hoge salarissen, terwijl datawerkers in India, Venezuela en Madagascar onderbetaald worden. Dit volgt koloniale patronen. India voert datawerk uit voor Engelstalige landen, terwijl Franse bedrijven werk uitbesteden aan Franstalige landen in Afrika.”

What can we do?

What can we do about this? He describes this in the last chapter of his book ‘What is to be done?’, a tongue-in-cheek quote from Vladimir Lenin. According to Casilli, a systemic approach is needed to improve the conditions of all data workers worldwide. ‘A solution for a specific group will not work in the end. We need to look for a universal strategy.’

He distinguishes between three types of solutions: regulation, collective platform initiatives and a global redistribution system:

1. Regulation: Spain, for example, has introduced the Riders’ Law and the European Union is working on guidelines for platform workers. “These are steps in the right direction, but this type of regulation needs to be applied more broadly. After all, tech companies operate globally.”
2. Platform cooperatives: Workers can set up their own platforms in which they have a say in wages and working conditions. ‘This is already happening on a small scale, but deserves more attention.’
3. Redistribution: Large tech companies can be taxed and the proceeds used for a universal basic income for data workers. Importantly, Casilli states that this UBI is neither connected to a “robot tax” (as he doesn’t see robots replacing workers) nor it is intended to replace welfare assistance (as it should be paid regardless of other social benefits). ‘This will ensure greater fairness.’

By combining these three strategies, the professor hopes that we can create a fairer and more sustainable system. ‘Tech companies must take responsibility for all their workers, including the invisible ones who manufacture their data,’ says Casilli. ‘I am concerned about this situation: wages are far below the minimum and even basic health and safety rules are not always observed.’

Casilli believes that organisations such as the WageIndicator Foundation and the Fairwork project are making an important contribution. ‘These organisations set standards for fair wages and working conditions, and these are desperately needed.’

Enforcement, collective action and user responsibility

After several interviews on this topic, I personally believe that, besides the solutions that Casilli provides, it is also important to enforce existing regulations. In countries where there are many underpaid data workers, there is a lack of supervision. This is partly due to strong lobbying by tech companies. That is why it is so important for workers to take collective action, for example via trade unions. These are underrepresented, although a number of interesting grassroots initiatives have emerged.

I also believe that (large) users of AI solutions must take responsibility. There are many discussions about responsible AI use. But I can no longer take a discussion about responsible AI seriously if it does not take into account the hidden workers.

Why this is important

Casilli and his team are uncovering an important mystery: AI is not a magical ‘black box’. In reality, millions of people work behind the scenes on these so-called ‘intelligent systems’. AI is presented as completely autonomous, and the extensive manual labour involved is often forgotten or ignored. This has serious consequences for the working conditions of these data workers.

If we really want to use AI responsibly, we must also consider the people behind the technology. I try to raise awareness of this issue and highlight it wherever possible. That is why I spoke earlier with Claartje ter Hoeven about Ghostwork: the invisible world of work behind AI. I will soon be speaking to the Data Labeler Association in Kenya to gain more insight into the conditions and problems faced by workers in Kenya. After all, we can only really get started with responsible AI if we understand how AI is created.

Want to know more? Listen to the full podcast with Antonio Casilli here. Check all podcast episodes via this link.

Claartje ter Hoeven (Utrecht University) and her research team reveal the hidden world of European data or ghostworkers. They are often highly educated, seeking flexibility, but their working conditions are usually poor.What drives them? What impact do they have on algorithm development, and vice versa?

They are invisible, ubiquitous and indispensable for the development of artificial intelligence (AI): ‘ghost workers’. Millions of people worldwide annotate, check and translate texts and images so that AI can understand and process the information. Who are these people and what drives them? What about their well-being? And what impact do their poor working conditions have on the development of AI? To learn more about this, I sat down with researcher Claartje ter Hoeven of Utrecht University for The Gig Work Podcast from the WageIndicator Foundation. She is conducting research into this phenomenon with a European Research Council (ERC) grant.https://open.spotify.com/embed/episode/5km4FhBvCgU2cW29EmlDpZ?si=b165a66c63e44aa9&utm_source=oembed

Annotating, checking and correcting so AI can learn

Ter Hoeven and her team are researching the working conditions and well-being of so-called ‘ghostworkers’ finding work through online platforms. They build on the work of Mary Gray and Siddharth Suri, ‘Ghost Work: How to Stop Silicon Valley from Building a New Global Underclass’. While Gray and Suri researched working conditions of ghostworkers in the United States and Asia, Ter Hoeven and her research team focus on Europe and examine the relationship between working conditions and worker well-being. It is a five-year study and the team is now about halfway through.

‘We are investigating the hidden labour behind AI,’ she says. ‘Ghostwork is a catchy term, but in science these days we prefer to call it ‘datawork’. Dataworkers make texts and pictures readable to AI in order for the machine to learn from them. For example, they indicate what a lamppost or a bicycle is, so that the algorithm of a self-driving car learns to recognise these objects. They also check and correct the output of AI models and algorithms.’

Low-paid

Data workers often do this from home through online platforms such as Amazon Mechanical Turk, Clickworker or Microworkers. They usually get paid per mini-task or ‘microtask’: an annotation, check or translation. The pay is often low and data workers have to search for the microtasks themselves on various platforms. Searching takes time, but they do not get paid for that. They often earn less than the minimum wage.

There are also companies that employ data workers, so-called Business Process Outsourcing organisations (BPOs). They work in physical office locations, are often paid by the hour and are supplied with the tasks. Although they have no unpaid search time, their pay too is often below the poverty line. 

We still know quite little about these data workers. and many big tech companies prefer not to talk about the contribution of humans in the development of AI, because it does not fit the narrative that AI is ‘self-learning’. This is not only a shame, but also detrimental to the development of responsible AI. It is therefore beneficial that Ter Hoeven and her team are researching datawork through platforms in Europe.

Highly educated with a distance to the labour market

Ter Hoeven used various research methods to discover how working conditions of dataworkers affect their well-being. It started with a short survey. This they distributed as a microtask on various platforms in Europe to get as many responses as possible. In the end, more than 5,000 people completed the survey.

‘A striking result was that many data workers are highly educated,’ says the researcher. ‘They often have a certain distance to the labour market. Think migrants or people who combine work with caring responsibilities for family members.’

Four drivers

Among data workers working via platforms, Ter Hoeven distinguishes four groups based on their motives , based on her survey of over 5,000 respondents: explorers, enthusiasts, supplementers and dependents:

Rather an algorithm than a human as boss

Ter Hoeven and her team conducted 137 face-to-face interviews with data workers from Europe. She discovered all kinds of motivations. ‘Those who work through platforms are dependent on algorithms,’ she says. ‘Algorithms determine whether you get a task and sometimes what you earn with it. There are all kinds of drawbacks to this. It is often unclear how algorithms make decisions, and platforms make it almost impossible to complain or discuss decisions. Yet many data workers told me during interviews that they would rather work for an algorithm, than a human manager.’

For example, she spoke to a migrant in Germany who had bad experiences with nasty bosses. Thanks to microwork, he could at least work from home and decide his own working hours. Another interviewee was a neuroscientist with a medical condition, which meant she needed more time to get up in the morning. She had to stop working at the university as a result. Thanks to datawork, she can still earn money. Ter Hoeven: ‘So our research not only says something about microwork, but also raises questions about the way we organise more traditional work.’

Need for colleagues and appreciation

The researchers present their findings not only on paper, but also in a documentary. They invited six European dataworkers to participate in video recordings. ‘We asked them questions and brought them together for panel discussions,’ says the researcher. ‘This cinematic research offered very interesting insights.’

Trailer of the film ‘Ghost Workers’ by Lisette Olsthoorn in collaboration with Erasmus University Rotterdam. This film was funded by the Erasmus Initiative Societal Impact of AI and by the European Research Council (ERC) as part of the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 101003134).

While most dataworkers indicated during the interviews that they generally did not miss colleagues, it became clear during the recordings that they actually needed to do so. ‘Suddenly they could complain, brainstorm and share experiences with like-minded people,’ says Ter Hoeven. ‘They had simply never had a dataworker colleague before. Furthermore, I saw their self-confidence grow during the filming now that they were the ones in the spotlight. Dataworkers may need contact and appreciation more than they themselves sometimes think.’

Data quality

Datawork raises a lot of questions. These are not only about the well-being of employees, but also about data quality. AI has an increasing impact on our daily lives. Ter Hoeven: ‘An example: some dataworkers annotate medical procedures. For example, they have to indicate whether a doctor’s hand is shaking during an operation. But these people usually have no medical background. So how reliable is that data?’

What consequences does it have if AI learns from people without sufficient expert knowledge and information about the context of the data to be ‘translated’? To improve quality, it may make sense to better match workers’ skills to tasks and provide better guidance. This is only possible if you invest more in data workers.

More transparency

In her book The Tech Coup, Marietje Schaake discusses how tech companies are conducting real-time experiments with user data to optimise their platforms, often without users’ knowledge. This can have serious consequences for privacy, democracy and personal freedom. That is why she calls for stricter regulation and more transparency.

The same applies here. In my opinion, organisations should be more transparent about the contributions of data workers and their potential risks. I therefore hope that European legislation like the Corporate Sustainability Due Diligence Directive (CSDDD) will also apply to how companies develop their AI. After all, this smart technology is increasingly affecting all kinds of processes. In short, the conversation with Ter Hoeven again leads me to many new questions. I will be seeking answers to those in this podcast in the coming months.

Want to know more? Listen to the podcast episode on Ghostwork here. This blog was also published on Gigpedia.org.

The GHOSTWORK-project received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme Grant agreement No. 101003134

The growing impact of digital technology, generative AI and algorithmic management on work is an increasingly widely explored topic. Last week during the two-day conference ‘Future of Work: Reclaiming the value of work in the digital economy’, researchers from across Europe gathered to present their work and exchange views.

It was organised by both the research group of the ERC project ‘Respect Me “ and the European Trade Union Institute” (ETUI). In this blog, I look back at the discussions and my own contribution on an upcoming paper on the Living Tariff methodology.

From focus on platform to focus on impact of technology on work

The shortcoming with many discussions on the platform economy and platform work is that it is assumed to be an isolated phenomenon. A self-contained silo. I wrote earlier that this is of course big nonsense: you cannot talk about platform work while ignoring the rest of the labour market. So that has been my biggest criticism for ages of how unions relate to this development. They often ignore what the (often poor) conditions of workers are in the same market where a platform does not provide intermediation and that the alternative for the worker is not a well-paid job with a lot of security. For me, this was most visible in the case where FNV sued domestic cleaning platform Helpling. When you have a discussion about domestic cleaning platforms without acknowledging that it takes place in a sector where informal work dominates in most countries, you deliberately miss an opportunity to do something for this group of workers.

It is therefore first important to look at what is really new. I did this in 2021 together with Jeroen Meijerink in our research ‘Online labour platforms versus temp agencies: what are the differences?’.

Also at the conference in Leuven, the call to look at what is really new was highlighted by Uma Rani of the International Labour Organisation (ILO), among others. She brought an interesting overview of a historical perspective of the use of technology in the context of work and also showed that sometimes the technology itself does not change, but the way it is applied does. This was also discussed by her ILO colleague Annarosa Pesole.

The platform economy is often seen as the ‘nursery’ or ‘sandbox’ of the labour market, where these technologies like algorithmic management are developed and tested (which raises important ethical issues) on workers. To then be applied to the wider labour market. Something that, by the way, is in line with the current development of legislation like the Platformwork Directive, which everyone understands that the passages on algorithmic management should apply not only to a specific group of platform workers, but to all workers.

And so that was the focus of this conference: the impact of digitalisation on the way we work, organise, allocate, control and (to some extent) evaluate work. Where the predominance was on the worker’s perspective, which is not surprising with the ETUI as co-organiser. There was much discussion on how to secure workers’ voices in the governance of this technology, how to make processes more transparent (and verifiable) and how to increase knowledge among workers and unions.

Insufficient use of existing regulation

There is a lot of focus on new regulation in the platform economy. In addition to the previously mentioned Platformwork Directive, a platform law to be implemented nationally in all European member states over the next 18 months, we also have the Digital Service Act (DSA), Digital Markets Act (DSA), Platform to Business Directive (P2B) and others. The focus here is on creating more transparency and accountability with the aim of improving the balance of power between the different stakeholders (platform, worker, client and society).

💡

Read a report on a meeting on workable regulation I organised earlier and a review of the DAC7 platform law

With all the attention around new regulations, you would almost forget that there is already a lot of regulation that working platforms already have to comply with. During the session ‘Representing workers rights in the platform economy’, among others, this was emphasised several times. This session was mainly about the upcoming Platformwork Directive, but also explicitly about GDPR. Since many platform workers are not employees, they are also not protected under employment law, making AVG for rights relating to data suddenly very interesting. (note: I am not a lawyer) For example, María Luz Rodríguez Fernández presented the GDPoweR project, which , according to its own website, ‘explores what worker data is collected by platform companies and how this affects workers, what strategies are used by social partners to negotiate and implement collective agreements and how the implementation of such agreements can be monitored and enforced. A central method used is the recovery of worker data through GDPR requests and the joint analysis of this data by workers and researchers.’ The slides below summarise the initial findings.

And although GDPR is an individual and not a collective right, the GDPoweR project shows that there is no reason not to use it for collective activities. Something I also described earlier in the blog and podcast ‘knowledge is power, even in the platform economy’ following an interview with James Farrar, founder of the Worker Info Exchange. All the cases his modern union brought against Uber, among others, were won on the basis of existing regulations. Something policymakers and unions should consider a lot more.

And of course, new regulations also provide opportunities. Personally, as a non-lawyer, I expect little impact from the employee part of the upcoming Platformwork Directive. It would be strange if your rights as workers depend on how you get your work handed to you. I see this part more as a stepping stone to better rights for an entire sector, although with the current political climate, the words ambition and political unity are rare. I am particularly curious to see how the excerpts dealing with the impact of technology on work are fleshed out. On the one hand, because certain things are contract-neutral (this sounds boring, but is revolutionary, as all certainties and obligations around security, among other things, are linked to being an employee) and, on the other hand, because there is a mention of ‘digital community channels’ that should break or reduce the isolation of individual workers (and thus an important instrument of platform power).

Finally, Annarosa Pesole (ILO), among others, warned on the finding that more and more parties are using ‘off the shelf’ technology and thus have little knowledge, insight or influence on the technology being deployed. A danger for the worker, but also for the one deploying the technology, as from the AI Act, among others, there is also a responsibility on the user of this technology.

Caught in the worker paradox and union dilemma

Many discussions on platforms and labour have revolved around whether the worker is a freelancer or an employee. In itself a logical thought from a Global North perspective, because there, being an employee is the dominant model and many securities and obligations are linked to this contract model. In addition to social securities, I am talking about health and safety obligations, but also a role in social dialogue and representation. If you are not an employee, you are virtually outside the scope of this.

Although I can understand the focus, I do question to what extent the discussion is not caught in an employee paradox, a term that also came up during this congress. This is because in many countries, the worker model is not the dominant model and because this focus ignores a large group of workers who also need protection. Think of freelancers, but also of the informal market.

The employee paradox also lacks room for nuance. Employing means, for many platforms, switching themselves to the temp model (with limited rights) or using existing temp agencies or subcontractors. Randstad’s CEO a few years ago, for instance, named the ‘gig staffing market’ as one of the big opportunities for the staffing company. One of the conclusions from the aforementioned paper I wrote with Jeroen Meijerink is that many types of platform work can be perfectly well organised via a temping model, with the important note that the terms ‘security’ and ‘fairness’ under these models are not of the level as they are presented to us in the public debate. And that it becomes very difficult for employment agencies to comply with the legal duty of care. Something that also came up in the contribution by Silvia Borelli (Università di Ferrara).

The conference presented several examples of how ‘worker representation’ in digital technology and AI can be organised.

A subject that will only become more important in the coming years. Here, unions face a dilemma: do they stick to the existing model, or do they initiate a change to a more broad focus on ‘the workers’ in the broadest sense of the word and build up expertise to also stand up for workers in the digital domain. Visitors to the conference agreed that unions currently give far too little priority to this, something also acknowledged in the report ‘Collective bargaining practices on AI and algorithmic management in European services sectors.’

I challenge unions to then instantly look a bit wider and examine whether they may need to target not only employers, but also mediators and the creators of technology. In the digital domain, it is more necessary than ever to fight for a better balance of power, which will ultimately also lead to better products and innovations. Trade unions could and (in my opinion) should play an important role in this.

What is a decent income? Towards a more global view

A topic that doesn’t come up much at these kinds of conferences is the topic of income. Or rather, decent income. When it comes to income, the topic is usually the intransparency of payments in on-demand platforms like delivery and taxi. An important topic, as these platforms have only made the structures of how a tariff is built more complex over the years, so the worker often does not know what the earnings are before accepting a gig and the gap between what the worker receives and what the customer pays has grown. In a market where transparency was promised, information asymmetry has only grown to the disadvantage of workers.

Besides the issue of transparency (it is questionable to what extent dynamic tariffs for work are at all desirable), there is no debate on the level of compensation a worker should receive for the service provided. A topic that is ranked number one when you look at what platform workers are taking to the streets worldwide for, according to earlier research by Leeds University.

💡

Check also the blog and podcast I produced “How and why does the platform worker protest? Scientists provide overview and insights”, based on an interview with researchers Vera Trappmann and Simon Joyce.

That was therefore the topic of my contribution: ‘Facilitating workers and policy makers in the gig economy making better informed decisions: the case study of the Living Tariff’. The basis of this concept comes from the Living Wage methodology, where a minimum wage for a worker is calculated based on the costs a worker has to incur to live a decent life in country X in region X. A wage floor based on actual costs. Where the Living Wage is of value to employees, someone who is not an employee can do little with it. Whereas certain costs and risks of work in the case of an employee are borne by the employer, a non-employee (freelancers and informal market) has to bear these costs and risks themselves. These should be included in the hourly rate to arrive at a fair minimum threshold.

💡

Last year I interviewed Valeria Pulignano, who is in the lead of the Respect Me project, for The Gig Work Podcast by the WageIndicator on her research on unpaid labour. Check the podcast and blog.

With the Living Tariff methodology, these costs and risks are taken into account, allowing the freelancer to know what they need to earn per hour to eventually arrive at a Living Wage, after deducting risks and costs. This methodology fills an important gap in the calculation of minimum income (wages and tariffs).

A next step is for this calculation to be included in consultations with various stakeholders (clients/platforms, workers/trade unions and policymakers/politics) in a discussion on fair compensation for workers. If you want to know more about this, check out the Living Tariff page, the slides I used for my presentation, or the blog and podcast I produced on this topic.

To conclude

It was a delight to attend this conference, hear the insights of researchers and then have great conversations and discussions about this. What I would like to see is a sequel to a conference like this, but with a broader stakeholder perspective. At the opening, someone in the audience asked ‘are we also going to talk about opportunities?’. A fair question, but in order to arrive at exploiting opportunities (and reducing risks), an insight from different stakeholders is needed. Insights from policymakers and platforms themselves, for example. Now platforms have an image problem due to the sometimes questionable practices of some big players, but the market of platform companies is in fact mainly an SME market. According to the ‘Monitor online platforms 2023’ (CBS, 2023), 64.2 per cent of the 1,600 Dutch platform companies have 2 or fewer employees. Only 5 per cent of these companies have more than 100 employees. Those opportunities may not come directly from the big players, but there are plenty of small(er) platform companies that are a lot more approachable where positive change could come from. Whereas now there is sometimes a bit too much emphasis on struggle, which is explainable because of the union and collective action perspective, it would also not be wrong to apply experiments on a smaller scale with parties that do want to collaborate. And to take the lessons from there and scale them up. This also includes a discussion on the more fundamental questions about the value of work. The title of the conference was ‘reclaiming the value of work in the digital economy’. This is a step that, as far as I am concerned, is often skipped, because it also involves a piece of self-reflection on parts of the (vulnerable) labour market that are not included in the debate.

I am convinced that there are still plenty of opportunities, only it is up to those involved to take responsibility and take up, test, validate and scale up these opportunities. I am happy to contribute to that.