{"id":437,"date":"2024-12-16T09:50:21","date_gmt":"2024-12-16T09:50:21","guid":{"rendered":"https:\/\/wpmu.mau.se\/msm24group5\/?p=437"},"modified":"2024-12-17T15:32:41","modified_gmt":"2024-12-17T15:32:41","slug":"ai-has-come-to-recruit-me","status":"publish","type":"post","link":"https:\/\/wpmu.mau.se\/msm24group5\/2024\/12\/16\/ai-has-come-to-recruit-me\/","title":{"rendered":"AI has come to recruit me"},"content":{"rendered":"
<\/div>\r\n

A mysterious job offer<\/strong> popped up in my Linkedin inbox about a month ago: \u201cHi Eva, I\u2019m a recruiter at <\/span>DataAnnotation<\/span><\/a>, where we pay smart folks to train AI. Earn $1K+ USD weekly paid via PayPal. Work as much or as little as you want \u2014 make your own schedule!\u201d (see the full message below). After weeks of intensively researching gig workers and labour conditions for this blog, the forces of AI<\/strong> seemed to be pulling me in.\u00a0<\/span><\/p>\n

\"Screenshot
A screenshot of the Linkedin message and the DataAnnotation recruiter website, both intended to attract talent using employee branding techniques. Images: Eva\u2019s Linkedin inbox and DataAnnotation.<\/em><\/figcaption><\/figure>\n

Following the recruiter\u2019s advice and a genuine interest in fieldwork, I applied for the offer and immediately received a one-hour test. The assessment was a prerequisite to access paid projects, including: <\/span>\u201cAI chatbot training,<\/strong> text classification (sentiment analysis), content moderation, image classification, search evaluation, creative writing, data entry, and more\u201d.<\/span><\/p>\n

Fact-checking, language skills, analytic techniques, critical thinking and creativity were some of the areas evaluated. However, a disclaimer stated: \u201cAll responses must be written by you from scratch. We are not looking for perfect English \u2014 it\u2019s fine if there are typos or grammar mistakes. <\/span>Do not use any translation tools to assist you\u201d.<\/strong> In other words, was AI banning me from using AI itself in order to prove my humanity?<\/span><\/p>\n

\"Exercise
The assessment included an exercise on poetry, evaluating the best response to the prompt \u201cWrite a tragic six line poem about music\u201d. Image: DataAnnotation.<\/em><\/figcaption><\/figure>\n

What seemed surprising \u2014 and even incoherent \u2014 reveals one of the most fundamental premises of this technology: <\/span>AI is neither artificial nor intelligent,<\/strong> and requires large amounts of human labour. However, the workforce needed to run AI systems remains deliberately hidden as \u201c<\/span>ghost work<\/span><\/a>\u201d (Gray and Suri, 2019).<\/span><\/p>\n

\u201cAll technologies \u2014 even the most seemingly advanced, autonomous, artificial intelligence systems \u2014 are ultimately products of people\u201d, states Jathan Sadowski, Senior Lecturer at Monash University (Australia), on <\/span>his website<\/span><\/a>. The efforts to hide the role of human labour is described by the author as the <\/span>\u201c<\/b>Potemkin AI<\/a><\/strong>\u201d,<\/b> a fa\u00e7ade of cultural hype made to attract venture capital. \u201cAI systems are, in large part,<\/span> technical illusions to fool the public<\/strong> (…) AI systems deploy technical buzzwords that sound like a magician\u2019s incantations: <\/span>Smart! Intelligent! Automated! Cognitive computing! Deep learning!<\/span><\/i> (…) If an AI application relies heavily on human labour rather than machine learning, that doesn\u2019t make for a good sales pitch to venture capitalists and customers\u201d, explains (Sadowski, 2022).<\/span><\/p>\n

AI winter in 2025?<\/b><\/h4>\n

While tech moguls such as Elon Musk announced that <\/span>AI will surpass human intelligence<\/span><\/a> by 2025, the AI hype grows exponentially as do the chances of falling into a new <\/span>AI winter<\/span><\/a>. AI companies\u2019 revenues have skyrocketed over the past two years, but their computational costs for running more sophisticated AI models have also multiplied. For example, improving an AI\u2019s recognition accuracy from 60% to 67.5% required quadrupling the training data, and some top tech corporations, including OpenAI, are currently<\/span> operating at a loss<\/b><\/a>. <\/b>The <\/span>fear of a looming AI winter<\/strong> freezing all funding and interest in AI \u2014 as already happened in the 1970s \u2014 is real.<\/span><\/p>\n

\"Two
Although the history of AI began in antiquity, the research into thinking machines became prevalent in the late 1930s. Image: ARL Technical Library.<\/em><\/figcaption><\/figure>\n

In the meantime, we are also witnessing the emergence of a planetary and fragmented <\/span>labour market for gig work<\/strong> (Graham, 2022) to fulfill the urgent need for \u201chumans in the loop\u201d to retrain AI (Schmidt, 2022). Contrary to the fundamental economic law of supply and demand, the current high demand for human power translates into a race to the bottom in terms of wages and labour rights, and a \u201cproletarianisation\u201d of data science jobs (Steinhoff, 2022).<\/span><\/p>\n

Technocolonialism in the gig economy<\/b><\/h4>\n

India and the Philippines have become the main market for content moderation outsourcing, but Indian and <\/span>Filipino <\/span>content cleaners<\/span><\/i><\/a> were left out of the <\/span>$52 million legal settlement<\/span><\/a> to provide relief to Facebook content moderators exposed to distressing and violent content. Venezuela has become the<\/span> source of cheap and highly qualified labour<\/strong> in the field of data annotation due to the socioeconomic collapse of the country, aggravated by hyperinflation and mass starvation after 2017 (Schmidt, 2022). Other examples of the emergence of modern slavery across the globe have been described by my colleague Pree in <\/span>this blog post<\/span><\/a>.<\/span><\/p>\n

Thus, the <\/span>AI industry profits from catastrophe<\/span><\/a>, and its dynamics of data capital and digital labour follow the colonial political and economic relations where the margins serve the core (Graham, 2019). In other words, tech companies exacerbate <\/span>AI colonialism<\/strong> or <\/span>technocolonialism<\/strong> (Madianou, 2024).<\/span><\/p>\n

The AI industry race is dominated by the <\/span>United States and China,<\/strong> redefining the geographies of the core and the periphery of this neocolonial extractivism<\/span>. <\/span><\/i>The fact that many citizens in the Global North also earn precarious wages through piecemeal digital work \u2014 producing datasets, training AI systems, serving platforms \u2014 reveals how the periphery is also nestled unevenly inside the imperial core of innovation (Sadowski, 2022).<\/span><\/p>\n

Data annotators wanted<\/b><\/h4>\n

As I wait to hear back from my potential employer, I navigated Linkedin to search for <\/span>current DataAnnotation employees<\/strong> in my area (Spain). These are some of my findings:<\/span><\/p>\n

    \n
  • 17 profiles are currently identified as \u201cAI data trainer\u201d, \u201cdata annotator\u201d, or \u201cAI content writer\u201d for DataAnnotation in Spain (the global workforce is estimated to be between 5,000 and 10,000 employees worldwide, according to the company on Linkedin).<\/span><\/li>\n
  • 100% of them have been hired in the past nine months.<\/span><\/li>\n
  • 100% of them have higher education qualifications<\/strong>:<\/strong> BA (a total of 6 employees), MA (9), and PhD (2).<\/span><\/li>\n
  • Their educational backgrounds vary across the following fields: Language, Literature and Translation (6 employees); Communication (3); Education (3), and other disciplines (5).<\/span><\/li>\n
  • 100% of them have an advanced or proficient level of English.<\/strong><\/li>\n
  • 47% of them have between 5 and 10 years of work experience<\/strong>,<\/strong> followed by 24% with more than 20 years of work experience.<\/span><\/li>\n
  • 53% of all profiles displayed the <\/span>\u201cOpen to work\u201d frame<\/strong> in their picture.<\/span><\/li>\n
  • 70% of them are women.<\/strong><\/li>\n<\/ul>\n

    To sum up, the average profile of a<\/span> Spanish AI data trainer<\/strong> for DataAnnotation is: a woman with an MA in Languages or Communication, and between 5 and 10 years of relevant work experience. In other words, me.<\/span><\/p>\n

    However, what I found particularly striking is that half of the Spanish AI trainers at DataAnnotation have publicly indicated that they are seeking alternative employment opportunities, despite having started this role less than a year ago and being assured of earning over $1,000 USD per week. How come the allegedly <\/span>$4,000 monthly income is not enough<\/strong> to keep Spanish workers in the loop when the average monthly salary<\/a> in Spain is \u20ac1,500 net?<\/span><\/p>\n

    AI and care<\/b><\/h4>\n

    The price for annotations depends on the complexity and the level of accuracy of the task, starting from $0.05 for a box (Graham, 2022). An individual human would need up to two hours to process a complex segmentation of an entire image, with 99% of accuracy, which costs $6.40 retail. Tech companies would not be profitable if they paid their workers the minimum wage at the standard of Western nations. Thus, the tasks flow to the planetary workers willing to accept the<\/span> lowest remuneration possible<\/strong> (Schmidt, 2022).<\/span><\/p>\n

    DataAnnotation is no exception. On this <\/span>Reddit thread<\/span><\/a>, employees share their experience on being paid $25 per hour for high-quality labeling \u2014 such as, legal classification, driverless car data, or medical imagery \u2014 or only $35 in three months. Some regular employees have also shared their huge <\/span>downturn in projects<\/strong> during the last months. \u201cI haven\u2019t had anything to work on, except for the OG imaging projects which are paid $0.03 a pop, and I\u2019m making $20 maximum\u201d, explains a <\/span>data annotator<\/span><\/a>.<\/span><\/p>\n

    I continued exploring the <\/span>DataAnnotation\u2019s website<\/span><\/a>, which is exclusively designed for employee branding and recruitment. There is a curated effort to make the position attractive by highlighting a <\/span>section of testimonials \u2014 some of them with real people, others which appear to be generated by AI<\/span><\/strong>. This is an example that caught my attention:<\/span><\/p>\n