I drew my own portrait in Blender – not AI

I did have hopes for AI when I first heard about it. It was exciting that we could create and train something for repetetive end goals, and enhance workflows. However the way AI is being used and developed is less about worksing smarter, and more about working cheaper. It is awful.

A handful of job listings I received were for seasoned professionals in video production, illustration, voice and animation, to train the company’s chosen AI bots. Essentially, it is the modern version of “we’ll hire you for one hour so you can teach our junior how to do your job inhouse”.

The defenders are a classic example of sunk-cost fallacy – this is when you invest so much time, effort and energy into something you NEED it to be true, even when all the evidence is stacked against you. The current argument is “AI is not taking your job, it is enhancing it”. However job postings tell me that they are intentionally wanting to take my job – and make me complicit in the process.

When advice is offered about how to work with AI, it is essentially a thinly-veiled “how not to get sued” guide. Take a look at this example from a self-described AI Animation & Generative AI & Children’s Entertainment IPs creator:

Describing the style still borrows from artists – just not the ones who have the finances to sue you. It is still unethical, because you have no idea where the content is being sourced from. The only way you would know that is if you entered it yourself – and than means actual work.

Rather than seeing jobs in which you learn and build systems, I am seeing jobs driven by the promise of fast, cheap content to flood our already stuffed feeds. Here’s an example of an AI generated post graphic found on LinkedIn:

Once, we were critical of details, now we’re letting things slide because our new toys make pretty things.

It’s most evident in services which entice general internet users to invest in training their systems. Want to look like a Studio Ghibli character? Or how about a water color sketch? Try our new AI generating app – but if you want to keep going, you’ll have to subscribe for a low monthly fee.

I am a proficiently skilled illustrator, so instead of getting AI to generate my art, I used my Wacom Cintiq, fired up the latest Beta version of Blender, and got to work. Why? Because I like the creative process. Because it is a rewarding challenge and because it is not lazy. Here is a quick recording of the process for your entertainment, which I posted to my Makertube Channel

You don’t have to be a good illustrator to draw your own, or even practice an artistic skill. As one friend put it “a stick figure on a napkin is more artistic”. There is merit in that, because it speaks to a time, place, moment, and a human interaction. If you wish to learn how to be creative, there are so many avenues and resources out there – I should know, I created a bunch of them for CG Cookie in the last four years. There are tools you can invest in, write off on your taxes, and they will serve you well for YEARS.

I really wish AI was being developed to assist with clear outcomes, but the truth is the feedback loop in retraining existing algorithms is getting increasingly filled with slop, and we’re dropping our standards and even our ethics in exchange for a quick dopamine hit.

NB: I have omitted some details because I still respect the privacy of the individuals and companies shown, and these screenshots have not been used with their prior knowledge or permission, however these details are publically available.