Paris Marx makes pretty much this point in episode 3 of Data Vampires;
https://techwontsave.us/episode/245_data_vampires_sacrificing_for_ai_episode_3
@tante
> the OSI's new "Open Source AI" definition is not open source (and that "open source AI" basically doesn't exist for larger systems)
Good to know I'm in good company. I wrote about this a couple of months back, as part of a larger critique of the current "AI" hype cycle;
https://disintermedia.substack.com/p/invasion-of-the-mole-trainers
Honestly @nextcloud, what are you doing endorsing this snake oil?
(1/2)
@mensrea
> one more reason to use for "why we shouldn't allow 'AI' in our work environment?"
Especially in a work environment that produces software. It just occurred to me that this means source code to any software written by a Trained MOLE would be in the public domain. Proprietary licenses and ToS wouldn't apply. Free licenses, whether copyleft or otherwise, wouldn't apply. Neither would Source Available licenses.
Cripes.
"The US Copyright Office has issued a series of rulings, upheld by the courts, asserting that nothing made by an AI can be copyrighted. By statute and international treaty, copyright is a right reserved for works of human creativity."
#CoryDoctorow, 2024
https://pluralistic.net/2024/10/19/gander-sauce/
Well that is good news.
"Requiring that AI models add digital watermarks to their output disclosing their provenance."
https://www.platformer.news/thursday-newsletter-3/
Oooh. That's clever. Because it can apply to your home and garden (not actually) "Open Source" versions too.
Strypey
