AI and the Rise of the Luddites

For as long as we’ve had technology and technological growth, we’ve had detractors. Every technological paradigm shift came with its own spectrum of critics, from those who use it sparingly, to those who’ve decided to opt out completely. For example, we’ve seen similar conversations play out with genetically engineered crops and trains to smartphones and social media. Cue the AI movement. (Note I’m using AI here as shorthand for LLM text Transformers + multimodal and Image Diffusion models)

With the inevitable march of technological progress, each technological step forward has come faster and faster, being applied globally at record rates. Personally, it feels like this current wave of AI technology has kind of exploded all at once like a tidal wave crashing onto the global populace. Since roughly the release of ChatGPT 3.5 in November 2022, the current iteration of AI has slithered its way into everything – from AI-generated images in advertisements, to AI slop becoming so prevalent on Youtube that they’ve been forced to sweep the field, and to the Dead Internet Theory finally fulfilling its prophecy in social media forums like Reddit and Twitter. This has, of course, led to its own Anti-AI Luddite movement.

Currently, most people are still relatively untuned to the current state of the AI landscape. They’re stuck in this liminal state where they don’t personally use it, but more and more of their life is being occupied by the outputs of AI – as mentioned previously – AI use in the media they consume, corporate layoffs (purportedly) due to AI, and as most PC builders are aware, increasing prices in major components (RAM, GPUs, SSDs, potentially more in the future) due to AI companies gobbling up production bandwidth of everything. This is leading to a growing backlash.

There’s an increasing Anti-AI sentiment growing among the public. It’s not limited to the political left wing, but they do make up a major component of it. The political left is increasingly becoming anti-AI, which makes sense to me if you consider they make up a lot of the arts and education fields. These AI models took in their work as inputs to produce these models that are then used to put them out of a job. The effect isn’t felt uniformly; fields that map better to current AI models, like translation, writing, image and video generation, and coding are more impacted than, say, plumbing. I see many comments online from these type of AI luddites where they attempt to boycott any companies that openly use AI. or for example, PC gamers who openly wish for the AI bubble to pop so they can get affordable PCs again. Personally, I find the prospect very unlikely.

While I do think it’s possible many of the AI companies that exist today may go out of business in the future due to untenable economics, I do think the field of applied AI as we have it today will keep growing into the future, and likely rapidly. I’d say if there is a bubble, it’s more akin to the dot com bubble – overpriced individual companies, but the thesis of everyone moving to the web proved to be true eventually. Worst case scenario, that’s how I see the AI field playing out in the future. There are simply too many applications both good and bad for us to simply press the shutdown button. We would basically need to rise up and enact a Butlerian Jihad in order to wipe out all traces of AI in our current society, I don’t see people coming together to actually eradicate AI technology – many people may hate and fear it, but few are willing to risk life and limb to purge it from existence. From my point of view, once ChatGPT 3.5 came out, there was no putting that genie back into its proverbial bottle. And so I’m at a point where I either adapt or die. Also, to be honest, I’ve always been a technophile, so playing with this new emerging technology has been really novel and interesting. It’s such a step change from anything that has existed before it. In particular, it’s always been incredibly funny that for the AI field, we held up the Turing Test as the gold standard for when we would hit a point where we could consider machines sentient. These AI models trivially broke past this 70~ish year standard and everyone collectively shrugged. I do think we’re going to continue to see people polarized either for or against AI in the near future. But, at this point…