I've been increasingly concerned about the corporate monopoly over frontier LLMs. While many ethically-minded people choose to boycott these models, I believe passive resistance alone cannot break the structural grip of big tech. To truly “liberate” these technologies and turn them into public goods, we need to look beyond moral high grounds and engage with the material basis of AI—specifically compute, data, and the relations of production.

I've written two posts exploring this through the lens of historical materialism. The first piece analyzes why current “open source” definitions struggle with LLMs, and the second discusses what it means to “act materialistically” in our imperfect world. My goal is to suggest a path forward that moves from mere boycotting to a more proactive, structural socialization of AI infrastructure.

If you've been feeling uneasy about the AI landscape but aren't sure if boycotting is the final answer, I'd love for you to give these a read:

• Histomat of F/OSS: We should reclaim LLMs, not reject them
• Acting materialistically in an imperfect world: LLMs as means of production and social relations

#LLM #AI #opensource #historicalmaterialism #histomat #materialism #digitalcommons

@hongminhee @mrmarquez Damn good piece @hongminhee. I agree with you in so far I can understand since I’m not a programmer. Very keen to find a way to do this for artists and creators.

Hi @hongminhee,
maybe #LLMs are the X-ray of IT.
In the early days used like candy. (Kids got their feet x-rayed in stores on open appliances, so the parents could see if the shoes fit. No kidding)
As experience grew, use was regulated and cut down increasingly. But it's still used to this day. For narrow usecases. Applied carefully.

Admittedly I doubt LLMs are as useful as #Xray. I think it's rather the #asbestos (which made wonderful things of concrete possible but mostly wasn't worth the downsides).

@hongminhee thanks for writing that down ! I think it resonates with what we're currently doing at @swheritage , we have the data but we care about how it's reused https://www.softwareheritage.org/2023/10/19/swh-statement-on-llm-for-code/

Current big players still prefer (brutally) scraping themselves... but a few (too few!) companies or research groups are considering reusing our data for actually open models. First attempt was StarCoder https://huggingface.co/blog/starcoder . Good things might happen, too!