The human costs of the AI boom

If you use apps from world-leading technology companies such as OpenAI, Amazon, Microsoft or Google, there is a big chance you have already consumed services produced by online remote work — also known as cloudwork. Big and small organizations across the economy increasingly rely on outsourced labor available to them via platforms like Scale AI, Freelancer.com, Amazon Mechanical Turk, Fiverr and Upwork.

Recently, these platforms have become crucial for artificial intelligence (AI) companies to train their AI systems and ensure they operate correctly. OpenAI is a client of Scale AI and Remotasks, labeling data for their apps ChatGPT and DALL-E. Social networks hire platforms for content moderation. Beyond the tech world, universities, businesses and NGOs (nongovernmental organizations) regularly use these platforms to hire translators, graphic designers or IT experts.

Cloudwork platforms have become an essential earning opportunity for a rising number of people. A breakout study by the University of Oxford scholars Otto Kässi, Vili Lehdonvirta and Fabian Stephany estimated that more than 163 million people have registered on those websites.

Freelancer.com alone has more than 67 million workers registered, the equivalent of the population of the United Kingdom. These workers, many of whom are based in low- and middle-income countries, work for clients who can be in the other corner of the world.

AI companies are using online labor to outsource the large amounts of hidden manual work needed to build their applications.

From the client perspective, the offer is hard to resist. Online labor platforms present them with a cheap, often skilled, workforce available 24/7. And the fierce competition for jobs on the platforms ensures the hiring company will have the upper hand in negotiating the conditions and pay. It is not surprising that AI companies are jumping on this opportunity to outsource the large amounts of hidden manual work needed to build their applications.

But for the workers on the other side, the situation can be dire. They can spend large parts of their day searching and applying for jobs, time that is unpaid. For most platforms, there is no guarantee that the tasks offered will not fall below their minimum wage.

And if they have any problems with the client, there is not always a clear appeal process, putting them at risk of not getting paid at all.

For the third year, the Fairwork project, also based at the University of Oxford, evaluated 15 of these cloudwork platforms, such as Freelancer.com, Amazon Mechanical Turk, Fiverr and Upwork. Companies were scored on a scale of 0 to 10, according to the five principles of fair work — pay, conditions, contracts, management, and representation.

The scoring criteria included, among other things, whether the platform paid workers on time, whether it mitigated health and safety risks, and whether there were clear communication channels and due process for decisions affecting workers. The results were based on a survey of 752 workers from 94 countries, as well as information gathered from platform managers.

None of the platforms met more than half of these basic standards of decent work. Popular freelance platforms like Fiverr and Freelancer.com scored 2 and 1 points, respectively.

Platforms focused on small tasks, like data labeling or content moderation, had among the worst scores: Appen (3), Clickworker (1), Scale/Remotasks (1), Microworkers (0), and Amazon’s Mechanical Turk (0). On average, workers surveyed were earning a dismal $2.15 USD for their working time on these platforms.

The results point to how many of these platforms are falling short of meeting basic workers’ rights and protections. Because of the international nature of this industry, cloudwork platforms can often circumvent national labor regulations, creating a sort of virtual Wild West where workers stand to lose.

Concerningly, this model is spreading fast — supercharged by demand from AI companies and the larger trend toward remote and flexible work following the pandemic.

Notably, some platforms in the study agreed to make some improvements, such as Terawork and Comeup, which implemented a minimum wage floor. However, the larger challenges in the industry can only be solved through regulations that ensure these companies are meeting the same labor standards as everyone else.

At the international level, the International Labour Organization is discussing the development of standards to ensure workers’ rights in the digital economy. This is a golden opportunity to address the problems faced by workers in cloudwork platforms. However, this will not be effective if national governments around the world do not get involved too.

Unfortunately, so far, discussions around regulating the platform economy have focused primarily on sectors like food delivery and ride-hailing. It is urgent that policymakers and regulators consider also the unique challenges faced by online remote workers, and put the same energy in finding solutions for this hidden, but very real, workforce.