The case for Neo-Luddism
I think Luddites had a point and we need to make it again in this time of rising AI and automation.
The term “Luddite” is nowadays often used to derogatively describe someone who’s resisting adoption of new technology. Who’s hindering progress because of skepticism or unwillingness to progress.
Historical Luddites were low-skilled workers, mainly in the textile industry, who were being replaced by machines. Spinning machines could produce a multitude of the worker’s output and were much less costly to industry owners, so workers were continuously replaced. This sparked a worker’s movement to protect their livelihood including drastic measures, like destroying machinery. The wikipedia article gives a more detailled account of the movements history. Despite being in direct competition with the machines and destroying quite a few of them, the Luddite movement of 19th century England was not opposed to technological progress. They were not opposed to improved efficiency of production processes per se. They were opposed to being replaced and worsening working conditions. They were opposed to being unemployed because replacing them with newer technology was more profitable[Luddites].
Technological advancements - specifically spinning and threshing machines - improved industry efficiency, which is objectively a good thing. More efficient technology makes it easier to produce goods that everyone needs, like clothing and food. This is great for humanity and society. I would conclude that improving industry efficiency should reduce the burden on the workers, make their life easier while making it more productive, while also producing more goods and improving prosperity for society as a whole. However, the Luddites were wage-laborers and their employers could just replace them with no repercussions. The efficiency improvement did in fact benefit society as a whole by producing more and cheaper goods, but apart from that it only benefitted the employers. The workers got the worst of it. Society did not progress with the technology.
We are currently in the largest wave of AI-hype we have ever witnessed. It feels like every other product tries to integrate AI in “innovative” and “disruptive” ways - as if our society and economy needed more disruption. I feel like a luddite. While I am not opposed to technological advancements and the furthering of human prosperity, AI does not seem to bring actual innovation in production processes. It improves efficiency by automating manual work, like the spinning machines before. For example, AI allows companies to creating marketing art, or texts, or even repair suggestions for car crashes. The models used for this rely on uncredited work[AI Copyright] while replacing the same workers they are based on. All of this - once again - will profit the companies and result in reduced employment and worse working conditions.
As Joseph Browning[Browning] so very poignantly put it: “AI gives wealth access to skill while denying skill the access to wealth”. AI feels like another technological advancement that makes it easier for owners of companies and products and other means of production to “improve their efficiency” by making computers do more of the work and cutting more of the workforce.
I want to advocate here not for hindering progress, but for reprioritizing. Instead of focussing on technology - even though it is very cool and fun - we should let society catch up. Technology has progressed vastly quicker than we can handle it for centuries now with catastrophic consequences. Our methods for producing necessities and luxuries have continuously improved, yet our structures for distributing them among society have not. We have food in abundance and still people hunger. We can generate AI-art in seconds yet still artists live off of patreon. Our environment can not handle the burden we place on it with our technology[Industrialization Climate Change]. The predicted energy cost of AI[AI Energy Consumption] will only worsen climate change. Our society can not handle the burden we place on it with our constant thrive for profit[Shit Life Syndrome]. As long as profit is the only incentive for companies, people will suffer from layoffs and bad working conditions. We are past the point where we require more resources for a good life.
We need to structure our economy to fulfil human needs, not shareholder needs. We need to put constraints on the use of AI, from the restriction of training data to avoid copyright infringements and unconsented exploitation of labor, to the use of exclusively renewable energy to power the ever growing server farms. We need to ensure that workers being replaced by automation - which I think is inevitable and often even desirable - does not hurt the workers. We need policies to ensure everyone’s livelihood with retraining programs or even universal basic income. We can most certainly afford it. If we can’t, it’s not progress.
P.S.: Since I’m new to this kind of texting, this post was made with the help of ChatGPT. It’s a useful tool. But use it ethically.
- [Luddites]: You’ve Got Luddites All Wrong, Vice, 2014, https://www.vice.com/en/article/luddites-definition-wrong-labor-technophobe/
- [AI Copyright]: Study for rights holders: AI training is copyright infringement, Heise, 2024, https://www.heise.de/en/news/Study-for-rights-holders-AI-training-is-copyright-infringement-9860466.html
- [Browning]: Joseph Browning, 2024, https://www.facebook.com/share/p/sQGLoN9wumXfgqHU/
- [Industrialization Climate Change]: History, 2024, https://www.history.com/news/industrial-revolution-climate-change
- [AI Energy Consumption]: Technology Review, 2024, https://www.technologyreview.com/2024/05/23/1092777/ai-is-an-energy-hog-this-is-what-it-means-for-climate-change/ (Note that while this article is ambivalent about the impact of the increase in energy demand, 2% of global energy consumption being used on data centers is still a lot and increasing this small percentage matters when aiming for climate goals.)
- [Shit Life Syndrome]: https://en.wikipedia.org/wiki/Shit_life_syndrome