Alex’s Reviews > Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence > Status Update

"some of those refugees are looking way too healthy not to be terrorists" (paraphrased)
Other than that, AI-bros definitely are human.

Starting with a faulty theory: that emotions are universally expressed in the same way, and thus can be read with a degree of certainty from the faces of people. The experiment is also flawed, but who cares? This set the base for AI's facial recognition, and mainly the same woo as lie detectors. Even the database is built mainly on white American people, which gives AI-agents even a racist streak.

In order for the AI to predict anything, it must be able to recognise what it's working with. Classification of objects is a relatively simple thing. However, when it comes to humans, it gets complicated. Not only are several identities vague at best, but complete ethnic categories are laden with racist (or colonial) subtext. There is also a whole point to be made about slurs being also used as categories.

" [IBM's researchers state]"aspects of our heritage (...) are reflected in our faces." This claim goes against decades of research that has challenged the idea that race, gender and identity are biological categories at all but are better understood as politically, culturally and socially constructed."
So, AI lobbed us some +/- 50 years in the past, racism-wise.

Data, in the world of AI, is presented as something neutral, unbiased. This however is not the case: already in the 1980's datasets were created using sets of data, without questioning whether or not the data 'mining' (personal data now being seen as a natural resource) was ethical or not, whether or not it harmed a specific group in the way it was presented.

This chapter was about labour, and time-keeping. We've seen how AI is used in a Fordian sort of way closely watching the time expended by the workers, in order to make production more effective. A Panopticon of sorts, if you will.

Or an Amazon warehouse.

An introduction to the concept of the Potemkin-village and the Panopticon. Both have a certain relation to Gen-AI. The Panopticon because both that and AI are ways to keep an eye on a multitude of workers and/or prisoners by extending the least amount of labour.
The Potemkin-village (colloquially known as an "image d'Épinal" in French) because everything is a facade and covering the labour of actual people.

"When a company introduces a new AI platform, workers are rarely allowed to opt out. This is less of a collaboration than a forced engagement, where workers are expected to re-skill, keep up, and unquestioningly accept each new technical development."

I think that's important to keep in mind.

I'm starting to see a pattern. The kind of pattern that is so obvious that we confused it for "how things are"

Oh fuck.

"The tech sector heavily publicizes its environmental policies, sustainability initiatives, and plans to address climate-related problems using AI as a problem-solving tool. It is all part of a highly produced public image of a sustainable tech industry with no carbon emissions. In reality, it takes a gargantuan amount of enery(...)"
Oh, so it's not even an original idea.