Shiny Happy AI
Better Images of AI is an organization supported by the BBC, UK Art colleges and AI companies.
A brilliant AI propaganda site is using equity, diversity, and inclusion and cliche images of technology as a smokescreen for their real project of making A.I. just another misunderstood technological advance. I was reading a Substack post earlier this week and an image in the article was tagged with a link to—Better Images of A.I.
I was curious and visited the site, there was a lot of content about the ‘issue of A.I. imagery’ and some strange concerns—
The colour blue …
In the Global North, the colour blue has been associated with progress, and particularly technological progress, since the nineteenth century. However, it also has connotations that are not always desirable in the context of Al: blue can also be seen as male, clinical, and distant, and using the colour nudges people towards acceptance and resignation.
From the Better Images of A.I./A Guide for Users and Creators/ authors: Dr. Kanta Dihal/ Tania Duarte
White robots…
Depicting Al as robots that are white in colour, ethnicity, or both, associates intelligence with being white. Such images serve as a barrier to increasing racial and ethnic diversity in Al development and decision-making, and exclude the global majority.
These statements are providing cover for AI products, by calling out bias in images using the colour blue and white robots, the bias built into actual AI remains unremarked. There is not ONE word about copyright, hallucinations, job loss, or actual social harms beyond perceived harms from the following list…
Welcome to Illustration class 101. The point of an actual trained illustrator is to avoid cliches and develop images that can inform, inspire, engage, or entertain. This is intended as a Stock Image site —-creators are asked to donate, but will be credited, yeah that pays the rent. Stock illustration is typically generic, and dumb, literal images proliferate. But there is something more sinister at work here, this seems like a project aimed at defanging any negative imagery about AI. The concern about stereotyped imagery as a critical issue and the absence of any real debate or question about using AI, with the support of AI developers makes it clear this project is just another insidious approach to undermining artists and the rights to their images.
In their About page they give away a bit of their game…(my bolded emphasis added)
These Al images also add to the public mistrust of Al, a growing problem for innovation in a field that is sometimes seen as biased, opaque and extractive. Finally, we think that images like these don’t encourage the necessary diversity of people to enter the Al workforce and address the Al talent gap.
Sometimes seen as biased, opaque, and extractive—-how about is biased, opaque, and extractive. It’s language like this that makes me want to make more images devoted to the apocalyptic hellscape of our coming AI Shangri La. The AI talent gap? What the hell is that, the problem of slop, the lowest common denominator of images drinking from the same water that all humanity peed into?
In the guide is a section dedicated to imagination and again this looks like it was pulled from the AI platforms playbook. Your imagination is informed by the images you have access to—-think about that, what does it actually mean. If our imagination only processes data then an AI that does the same thing has an imagination—-simple equation right, and dead wrong. Our cognition is not a data centre, and our mind and imagination is extended through our physical and sensory engagement with the world.
My advice, let’s make more images of evil robots and sci-fi doom scenarios, the more we can convince people to worry about this tech the better. Why should we be using nuance in a battle for our lives against a foe that has extracted billions of dollars of worth from creative people, has access to the manipulated attention of billions of people, and their hands around the throat of all culture on this planet.