Microwork in the global South, the motor of platform capitalism: interview with Phil Jones

In recent years, many books and articles have focused on microwork platforms and workers who train, annotate and tag data for artificial intelligence. Last week, another book was released on this issue, Work Without the Worker: Labor in the Age of Platform Capitalism, by Phil Jones, researcher at the think tank Autonomy. The book argues that understanding these low-paid jobs – which make our digital life possible – has a lot to say about the state of global capitalism.

The global South plays a leading role in the book. According to Jones, “Microwork truly represents not the phoenix of the South but a further twist in our planetary crisis of work. Microwork is teh sum of the same processes of sluggish growth, proletarianization and declining labour demand that have ballooned the informal sectors of countries such as India, Venezuela and Kenya”.

Read the interview with Phil Jones:

DIGILABOUR: In the last few years, there have been some books on microwork platforms, but very focused on the Global North. What is the place of the Global South in your book? And how to go beyond a vision of the Global South beyond “precarity” and “absence”? 

PHIL JONES: The global South is the motor of platform capitalism. It’s where the majority of the work to power machine learning is done. Facebook, Google, Amazon, TikTok and Microsoft all rely on large amounts of short data tasks to power their artificial intelligence capacities. Because they now represent the most powerful companies in the world, this means microworkers in countries such as India, the Philippines and Venezuela – the countries where much of this work is outsourced to – are in a contradictory position. On the one hand, they have immense potential power to cause serious disruption to these platforms by organising and going on strike. Yet, they are also doing a highly fragmented kind of work, where they are often alone working in their homes or in internet cafes, frequently hamstrung by non-disclosure agreements and the threat of having their accounts closed down. So it’s actually really difficult to organise.

DIGILABOUR: Noopur Raval criticizes notions such as “hidden labor” and “ghost workers”. How does your book deal with these conceptual issues? 

PHIL JONES: In the book, I maintain the position that the majority of work under capitalism is hidden. The point of the book is to define the concrete conditions of microwork. Microwork is purposefully opaque, in part so that the companies who use these platforms can do so without public knowledge, but mostly so that workers are unable to see each other, effectively organise and understand the kinds of project they are working on. Contractors on platforms are not obliged to tell workers about the kinds of project they are working on. One academic study revealed that when asked what project they thought they were working on, workers often did not know. This means that workers are very often in the position of helping to create technologies like facial recognition or automated drones, without having ever given proper consent. I hope this is ok. Please contact me with any further thoughts.

DIGILABOUR: How do you relate refugees to crowdsourcing platforms? 

PHIL JONES: Refugees are working for large tech companies via extended supply chains. These supply chains are mediated via impact sourcing companies and organisations such as Sama and Cloud Factory. These companies set up microwork programmes in refugee camps such as Dadaab in Kenya, and provide computers, broadband etc. and then train refugees to do basic data work, before helping to contract their labour to big tech companies. There is actually very little recent information on precisely which camps these companies are operating in, the conditions of labour they provide such as wage rates etc. in part because these companies have stopped publishing public reports on their activities and impact. From the data that is available, it appears that some of these companies have been providing refugee workers to sites like Amazon Mechanical Turk, where wage rates are very low and conditions very poor. 

DIGILABOUR: And how are the impact sourcing and PR efforts of these platforms? 

PHIL JONES: Unsurprisingly, the companies market these programmes as ‘opportunities’ for refugees, rather than opportunities for the tech companies that exploit them. These PR efforts have been swallowed uncritically by much of the business and tech press, where  ‘impact sourcing’ has mostly received boosterish coverage. There has been very little effort on the part of these publications to criticize their activities, report on what they are actually up to and hold them to account. What we currently need is a deeper journalistic investigation into the concrete conditions of those who participate in such programmes.

DIGILABOUR: What did you find most interesting about crowdworkers workspaces?

PHIL JONES: Well most crowdworkers don’t have a designated or centralized workspace. They work from home or internet cafes. Those who do have workspaces may be in prison or refugee camps, where its very hard to get a precise picture of their workspaces. In China, microwork actually takes place in workspaces which, somewhat ironically, are often old renovated factories or warehouses, potentially where much of the work has been automated or at least enhanced by AI. The conditions in these spaces are, by most reports, pretty terrible, often entailing excessive surveillance, long days and repetitive labour. 

DIGILABOUR: What futures of work should we fight for? What are the directions for imagining other futures?

PHIL JONES: We should be fighting for a future beyond the wage, where the value of work as well as leisure time is distributed equally. The world we should be fighting for is one where all have more free-time, a basic income and the opportunity to pursue the activities that interest them without exploitation or alienation. In a perverse way, microwork – and the crisis of work more generally – points the way to new horizons. The fact that these sites so often fail to provide a wage in a consistent manner shows the paucity of our current system of remuneration. The fact that workers on microwork sites spend more time hunting for paid work than actually doing it reveals that our time can be used for pursuits outside of remunerative work.  

BRAZILIAN RESEARCH ON THIS ISSUE:

The Brazilian Workers in Amazon Mechanical Turk: Dreams and realities of ghost workers (Bruno Moreschi, Gabriel Pereira, and Fabio Cozman)

Beyond Mechanical Turk: The Work of Brazilians on Global AI Platforms (Rafael Grohmann and Willian Fernandes Araújo)

Heteromation and Microwork in Brazil – In Portuguese (Matheus Viana Braz)