Look at the thumbnail above I generated for this specific article.

Yes, these two separate images put on top of another for a 16:9 website article thumbnail are completely generated by AI, with ChatGPT’s and OpenAI’s DALL-E. And they couldn’t do a basic job right. First image of the bubble was asked to be a 1920x1080 image, presumably 16:9, which didn’t accurately nail because the final image dimensions are vaguely resembling a 16:9 image except the exact dimensions are a complete miss (the final image resolution is 1792x1024). The second image of the hand holding the needle not only has an obvious hallucination, common in most GenAI image generators, with the needle tip coming out of one of the fingers, but I ordered a “transparent PNG” and all I got was a webp format JPG of what I presume is one of those fake PNG thumbnails you usually get on Google Image search results. Atrocious results for what’s supposed to be “scary” technology supposedly able to replace real artists and creatives.

You might remember a couple videos we did back in the day, one about a certain Playstation startup sound, which ended up being a rabbithole about the use of SEO with AI, and another one about the dangers of AI and deepfakes. Both these videos seemingly have halfway aged like milk it seems, because what it seems apparent is that the “scary” AI technology that’s supposed to replace us creatives, bloggers, artists et al is far from competent enough to properly fool anyone else other than your typical Texas Christian boomer grandma looking at pictures of Shrimp Jesus.

A contemporary of our generation, someone who seemed to break into the scene like a galloping rhino breaking through and dismantling the typical GenAI business, Ed Zitron, has claimed even that Sora, the GenAI video generation tool from OpenAI, seems to be dead on arrival, because of its technical inability to use the training data properly to generate minimally convincing results. He noticed these patterns not just in his newsletter blog, but also in his podcast series Better Offline, and seems to recently confirm my suspicions: this AI fad is going to be a DotCom styled bubble that’s about to burst, seemingly because major tech corporations like OpenAI, Microsoft and so on have been burning money and resources into generating record-breaking losses, as a result of the AI technology’s inability to not only make a solid profit, but also captivate convinced customers outside of the usual AI evangelists, those who will willingly eat a piece of AI-generated feces sandwich and actually swallow unlike yours truly did back in 2019 thanks to MaximilianMus. That’s so low even for me… yikes.

But this not only proves I prove myself smarter than those AI tech zealots; after all, all they’re gonna use GenAI for is to either fool Boomers with pictures of Shrimp Jesus or to either AI generate gooner material of questionable ethics; this also proves that Big Tech ain’t that particularly clever. No, they don’t come from a background of actually getting involved with technology; unlike the actual engineers who devised the original Google, Facebook, etc… these tech firms are consultancy firms in disguise, as Ed Zitron again put it, people who are only interested in making the most profit possible, pursuit yet again the next big thing in Tech, that next iPhone moment, that would make their corporations’ stock prices soar up infinitely, just so their CEOs can afford extra yachts and private jets, which further destroy our planet with toxic chemicals. In the meantime, the GenAI technology requires so much electricity use with its massive use of Nvidia GPUs that it further accelerates the melting of the icecaps on Earth, contributing to climate change even worse and the rising sea levels that will swallow us whole before Ben Shapiro dares to sell his house to the Innsmouth People.

So, for what? So we can further make pictures of Mickey Mouse holding a gun, driving a plane into the Twin Towers and enacting 9/11 once again in such a comical copyright infringement that resembles an act of terrorism? More pictures of Shrimp Jesus? The other glaring problem is that these AI generated slop trashbits are being either way reposted on the same Big Tech social media AI is scraping from for training data, so the end result is shit in, shit out; AI regurgitating redefecated defecations of the feces it just ate after shitting out. If you thought me about to bite a self-made shit sandwich was bad, think again. Yikes.

And the final glaring problem this will create in the future, as soon as this AI bubble bursts and creates an ongoing economic crisis similar to that of the 2000’s DotCom bubble, is that no big tech executive will pay for it. Oh no, they will land on golden parachutes. Everyone else is going to pay for it; the small businesses, the worker class, the average Joe and Jane, the commonground folk… I, myself, included. We’re headed into a disaster that’s going to have catastrophic consequences for the wellbeing of our ground society.

But, what can we do about it? I know I sound angry about this, rightfully so, and you’re in the right to be angry, but all I suggest is to build positive technology and creative media that doesn’t rely itself on being a con like GenAI is. Think of tech and media that can build a positive impact on the world around you. The thing is, we’re on it. But you… you might need to consider it, at least, because with that, no one at Big Tech will be happy about it.