Who is working for Amazon Mechanical Turk?

These experiments delve into the lives of Amazon Mechanical Turk workers. Following the logic of the platform, workers have been employed for a small fee and asked to give a glimpse of their private lifes. Unlike the typical use of the platform, the submitted data prompt reflection on hidden labor, material quality, and ways of rebelling against a system on the verge of ethics.

The ever-growing use of platforms such as Amazon Mechanical Turk, allow large groups of individuals to participate in creating, tagging and describing data.
Introduced by Amazon in 2005, Mechanical Turk is named after an 18th century “automatic” chess-playing machine, which was handily beating humans in chess games. Of course, the robot was not using any artificial intelligence algorithms back then. The secret of the Mechanical Turk machine was a human operator, hidden inside the machine,who was the real intelligence. ¹

Today, Amazon Mechanical Turk is a marketplace for tasks that cannot be easily automated, which are then fulfilled by human users. Such tasks, mainly involve the creation of datasets: collections of data that act as pedagogical learning material when dealing with Artificial intelligences, most specifically, machine learning systems. Current Artificial intelligences are in fact able to "learn” to execute such tasks by simply repeatedly analysing large numbers of labelled examples and counterexamples.
Such data is analysed and processed by workers on platforms like Mechanical Turks. But little is known about these individuals that complete these tasks.

On July 02, 2022 I uploaded 200 images on Amazon Mechanical Turkand I asked the workers on the platform to describe what they saw. The description above is one of the hundreds that I received in less than an hour.

The images that have been described are a result of the same process. On June 16, 2022, I started this experiment by asking the workers at Amazon Mechanical Turk to send me a photo of their view out of the window and the city they live in for $0.20.

As organisations use Mechanical Turk to create datasets, in these experiments more private dataseta have been created. Following the logic of the platform, for a small fee workers have been asked to show a glimpse of their private life, their view close to their desk, their working station. At the limit of ethics and labour exploitation I asked myself whether this is an acceptable approach to use in research and in general, if employing random workers around the world is the ideal practice to shape future artificial intelligences.

While both tasks were performed, the material collected varied considerably: from blurred photos to details of curtains, from landscapes to computer screenshots.

The same consideration can be made with the texts collected. Some descriptions are lengthy, others are short, some are tags, others instead are just typos. Following the idea that when talking about a technology one has to place oneself within it, becoming a 'Requester' on Amazon Mechanical Turk allowed me to find out more about how the system works, the review process and the quality of the data collected.

Although much is documented about the demographics of the platform and the fact that different payments do not affect the quality of the results, these 'results' are rarely visible. Most of the data collected did not reflect the assigned task. Many results were gibberish, or when asked to upload a photo, these were often desktop screenshots, stock images, or photos on the computer. These rebellious approaches to the platform allow workers to collect their rewards easily and can be ignored by requesters. For this reason, such datasets are shown in their entirety on this site.

Datasets are no exception. The information they carry is a reflection of time, politics and geographical contexts.
Although, in the past years, there has been a growing attention of bias studies in the field of machine learning, there is still a general disinterest for the ones carried by the hidden geographies of datasets. Consequentyl, a central focus is on the geographies of Amazon Mechanical Turk. Not necessarly on the demographics of its workers, but rather on the different places where data is annotated and created.