DiPLab’s report showing how Outlier tracks the workers that train Big Tech’s AI models was released yesterday, and the media have already started picking up on the story. This is how the story was covered in Spain, France, Norway, Poland and Denmark so far. Klassekampen (Norway) : “– American companies like Meta, Open Al and XAI are using Scale AI‘s platforms to exploit young, newly educated Europeans as a workforce to train their AI models” mediapart.fr (France) : “according to the conclusions of a report released on October 16, Outlier, a subsidiary of the American company Scale AI, itself a pioneer of the annotation since 2016, subjects its workers to extreme surveillance, harvesting and exploiting a very large number of their personal data.” Version2.dk (Denmark) : “In order to be able to work on the platform at all, the workers are forced to provide sensitive biometric data and facial scans through an invasive third-party app called Persona, which is owned by Peter Thiels Founders Fund.” Wyborcza.pl (Poland) : “a market has quietly emerged in which Outlier exploits highly skilled Europeans for not particularly high wages while simultaneously building a system of constant surveillance and data extraction that can be used for both AI training and commercial or military purposes” El Pais (Spain) concludes : “The authors of the report believe that Outlier is in breach of European regulation as it is today. They believe that the deregulation promoted by the Commission could weaken GDPR articles that function today as firewalls that stop the transfer of data and knowledge to the United States.” La Tribune (France) “Biometric facial scans, endless waits for tasks, intense pressure once work is assigned, and unexplained account suspensions… A report by DiPLab, (affiliated with Institut Polytechnique de Paris and Télécom Paris), and Reversing.Works describes the little-known working conditions of European workers recruited by Outlier.”

DiPLab is releasing a report on the surveillance of data workers today. Entitled “Surveillance at Scale : How AI training platforms gather personal data from European workers” this report is a research project in collaboration with Reversing.works and Intérêt à Agir, funded by the Digital Freedom Fund and the European Artificial Intelligence & Society Fund. International technology companies collect vast amounts of personal data from European citizens. Public debate and policymaking often focus on data taken from consumers through e-commerce, social media, or streaming platforms. Yet many Europeans also use digital platforms to work and train AI systems. Online freelance platforms are usually discussed in terms of low pay and insecure “gigs”. Much less attention is given to another resource these platforms extract from workers: their personal data. This report examines the case of Outlier, a key data work platform for artificial intelligence. Outlier is operated by Scale AI, a company that recruits workers to perform tasks such as evaluating AI-generated responses, creating training data, and improving AI models. This report shows how digital platforms rely on data workers worldwide, including in the EU, to label, check, and organize training data. These data are harvested online. But Outlier also collects workers’ personal data, financial records, biometric information, and even facial recognition scans. This data is used for surveillance purposes, and probably for further AI training purposes. Our interdisciplinary team analyzes the evidence collected from qualitative interviews, network logs, and guidelines, to provide evidence of a large-scale data collection process, through hidden trackers. We also provide a legal analysis in terms of likely European law (GDPR) data protection violations : the AI WORLd project reveals how the EU has become a periphery for data extractivism in the production process of AI models. Ethical AI requires the ethical treatment of the modern-day factory workers whose work makes it possible. Beyond the economic exploitation data workers through inadequate compensation, this report reveals the extraction of personal data for purposes ranging from covert AI training to potential military applications. These practices raise serious concerns about social justice, labor law, and European data security. Surveillance_at_scale_diplab_october_2026Download To cite this report : Thomas Le Bonniec , Antonio A. Casilli, Mathilde Abel, Simone Robutti, Gaetano Priori, Donatella D., Paul S. Surveillance at Scale : How AI Training Platforms Gather Personal Data from European Workers DiPLab / Reversing.Works report, 2026.

It is now less than a month to go until the start of the 9th annual conference of the International Network on Digital Labor (INDL-9, 9-11 September 2026): AI Supply Chains: Building an interdisciplinary research agenda for AI and labour. DiPLab, along with Yale University, ACM SIGGCAS, and the Geneva Graduate Institute, is truly delighted to be hosting such a vibrant international community of researchers, practitioners and activists at the International Labour Organization (ILO). DiPLab has allocated 10 small scholarships (EUR 400 each) to support travel, accommodation, and meal expenses of promising INDL-9 presenters. Scholarship will be awarded to recipients during the welcome address of the conference. Obtention is conditional on actual participation. After a careful selection process, the following ten applicants have been awarded a scholarship: * Kinchan Chakma * Prachi Sharma * Fitahiana Razafimahenina * Sonam Rai * Shikha Shalini * Amanda Biazzi * Prachi Tayade * Saadeddine Igamane * José Ángel Cerón Hernández * Fasica Gebrekidan We would like to extend our sincere thanks to the selection committee (Matheus Viana Braz, Stella Lopez, Marco Marrone, Myriam Raymond, and Iraklis Vogiatsis) for their time, care, and thoughtful consideration in a highly competitive selection process with a large number of outstanding applications.

Paola Tubaro was invited to give a talk at the ASOR Institute in Santiago in early July, and shared some results of her research in order to deepen dialogue on these important topics. She discussed how algorithmic management comes with a promise of higher productivity, but also increases surveillance in companies, while the apparent autonomy of self-employed workers on platforms—from Rappi delivery drivers to Workana designers—proves to be illusory. Her presentation, largely though not exclusively based on this open-access book chapter, analyzed the consequences of these transformations on job quality at two key stages of the AI life cycle: its production through human labor (so-called “data work”) and its deployment in workplaces across various sectors, from logistics to healthcare. Taken together, they reveal how AI accelerates preexisting trends in the global economy: the outsourcing of business functions while maintaining control over production—a phenomenon studied by D. Weil as the “fissured workplace”—and a “regime of dispersion,” first theorized by C. Datchary, in which workers manage multiple disconnected tasks. To protect job quality, she argued, corporate strategies and labor regulations must address the cognitive overload associated with fissuring, develop worker-centered organizational models, and implement safeguards for all users. The Millennium Institute for Authority and Social Regulation (ASOR), recently created in Chile, aims to conduct cutting-edge interdisciplinary research on the decline and reconfiguration of authority relationships in contemporary society and their impacts on social regulation. Authority, claim the founders of the Institute, is a pervasive social mechanism that is in place whenever there are power asymmetries. How people manage authority – in the family, school, workplace etc. – affects social cohesion and the extent to which (un)healthy relations of domination may arise. The rise of algorithmic management, especially AI-supported, is widely thought to affect labour relationships, endowing our bosses with new tools or even – as in the exemplary book by Aloisi and De Stefano) that algorithms actually are our new bosses. At the same time, the rise of platform labour – also closely linked to the AI boom – has promoted the model of independent workers who have no boss. Authority in the workplace changes gets more fragmented, disrupting traditional relational models, not only vertically (the hierarchical chain of command), but also horizontally (team composition, collegial work). What are the effects?

For years, Julian Posada (Yale Assistant Professor and DiPLab research associate) has been one of the architects of the field of digital labor studies. From helping shepherd the transition from the European Network on Digital Labor (ENDL, 2017–2018) to the International Network on Digital Labor (INDL, 2019–present), to co-organizing three consecutive INDL conferences, he has been present at every stage of a research community finding its shape and its ambitions. Today, INDL has created several regional chapters, in Middle East and Africa, in Latin America, and now (thanks to Julian Posada) in North America, too. This foundation has its first public moment: a Symposium on Labor and Artificial Intelligence, hosted at Yale University on April 28 and 29, 2026. The event brings together scholars and practitioners to examine how workers are building, using, and pushing back against AI systems. The programme opens with a keynote by Julia Ticona and closes with a roundtable featuring organizers from across the platform economy. Researchers and doctoral students are especially encouraged to attend the doctoral colloquium on April 28 (1:00–5:30 PM, KT 401), where emerging scholars can workshop their research alongside peers ahead of the main symposium on April 29 (9:00 AM–5:00 PM, HQ 134).

The podcast Penser les Luttes (Radio Parleur / Le Média TV) has just released a new episode titled Artificial Intelligence and Its Hidden Slaves (in French). We’re glad to share it here, as it features our colleague and DiPLab research associate Myriam Raymond. In the episode, Myriam Raymond speaks about who actually trains AI systems: the data workers behind tools like ChatGPT, Claude, and Gemini. Paid per task, often invisible, these workers label images, correct outputs, and perform the micro-tasks that make large language models function. As she explains: identifying objects in images, creating small visual assets, validating model outputs. Essential work, systematically undercounted. Myriam draws on her recent fieldwork for DiPLab, including her study on AI data workers in Egypt (part of a growing body of investigations into this workforce in the Global South). The episode also features sociologist Juan Sebastian Carbonell, who contextualises the rise of data work within broader questions of offshoring, automation rhetoric, and labour restructuring.

DiPLab’s Antonio A. Casilli was invited to speak on Pillole di Eta Beta, the technology programme broadcast on Italy’s RAI Radio 3, in an episode that opens with a striking new phenomenon: in Los Angeles, people are being paid to simply live their lives on camera. Wearing body-mounted cameras and sensor bracelets, workers film themselves doing household chores. Thousands of US workers have already been recruited for this work, paid a few dozen dollars for hours of first-person footage that becomes raw material for the next generation of autonomous machines. For Casilli, what is unfolding in Los Angeles is the latest iteration of a phenomenon that has involved millions of workers across Asia, Africa, and Latin America for over a decade: training algorithms, labeling images, moderating content. A digital proletariat that the technology industry systematically erases from its triumphant narrative. And yet without it, none of its products would function. The episode also raises a harder question about users. Niantic, the company behind Pokémon Go, recently sold 30 billion video sequences, captured from players navigating the real world through augmented reality, to a robotics delivery company. Millions of people filmed streets, parks, and shops without knowing their footage would end up training autonomous delivery systems.

DiPLab’s Antonio A. Casilli was invited to launch the brand new Cyber Revolution podcast as the guest of its very first episode. In conversation with Hugo Pompougnac, president of the association Espaces Marx, he discusses DiPLab’s research on data workers, their working conditions, their forms of organisation, and their emerging strategies of resistance. The conversation situates this new workforce within a longer history: that of working-class antagonism. Because however novel the algorithms, the platforms, and the micro-task interfaces may seem, the dynamics at play (exploitation, invisibilization, collective struggle) are ones the labor movement has always known.

On 9 April 2026, DiPLab co-founder Antonio A. Casilli spoke at the Dénumérisons ! symposium at the French Assemblée Nationale, at the invitation of MP Lisa Belluco. The question brought to the floor: behind the apparent “freedom” and “intelligence” of digital platforms lies a workforce whose rights are systematically violated. Click workers, data labelers, content moderators, and micro-task workers who keep AI systems running are exposed to violent content, subject to constant algorithmic evaluation, and denied basic labor protections. Any serious effort to regulate AI must begin by putting the human beings who build it back at the center, making their labor visible, their rights enforceable, and their dignity non-negotiable.

Public debate about work tends to fixate on numbers: jobs created, jobs lost, jobs threatened by automation. Job Quality in a Turbulent Era, a new volume edited by Agnieszka Piasna and Janine Leschke (Edward Elgar Publishing), asks a harder question: what is actually happening to the quality of work? The full book is available in open access via Edward Elgar Publishing. DiPLab’s Antonio A. Casilli and Paola Tubaro contribute a chapter examining how AI and platformization are “fissuring” contemporary workplaces: fragmenting tasks, eroding protections, and reshaping job quality across sectors in ways that aggregate employment statistics simply cannot capture. To cite this chapter: > Antonio A. Casilli, Paola Tubaro. 2026. What is AI Doing to Job Quality? > Platformization, Fissured Workplaces and Dispersion. Agnieszka Piasna; Janine > Leschke. Job Quality in a Turbulent Era, Edward Elgar Publishing, pp.45-60. Casilli_Tubaro_JobQuality_Chap3Download