These companies were already training their models on Wikipedia's wealth of information anyway. In this way Wikipedia is earning some revenue from the thing that was already happening, letting them put that money back into the non-profit.
Yeah I mean Wikipedia has regular dump files where you can just... download its entire content, or parts of it if you so wish. Getting money instead for that bandwidth is immediately an improvement
Wikipedia is also public knowledge. I personally think it's OK to use for training data.
However, there are other concerns about inaccuracies and some info on the site needs to be scrutinized and verified just because anyone can edit it, and the LLMs getting trained can't do that part.
What actually bothers me about this is that the companies training the LLMs are going to put them behind paywalls, removing the public knowledge part of this.
Non-profits also have owners. It being a non-profit just means that the company's main priority isn't turning a profit, so excess revenue generally gets pumped back into the non-profit.
11 Comments
Humanius@lemmy.world · 25 pts · 217d
Let me be the devil's advocate for this one.
These companies were already training their models on Wikipedia's wealth of information anyway. In this way Wikipedia is earning some revenue from the thing that was already happening, letting them put that money back into the non-profit.
Sunspear@piefed.social · 11 pts · 217d
Yeah I mean Wikipedia has regular dump files where you can just... download its entire content, or parts of it if you so wish. Getting money instead for that bandwidth is immediately an improvement
Nollij@sopuli.xyz · 2 pts · 217d
They weren't using those dumps. They were scraping the main site, at incredible expense to Wikipedia.
phaedrus@piefed.world · 2 pts · 217d
Wikipedia is also public knowledge. I personally think it's OK to use for training data.
However, there are other concerns about inaccuracies and some info on the site needs to be scrutinized and verified just because anyone can edit it, and the LLMs getting trained can't do that part.
What actually bothers me about this is that the companies training the LLMs are going to put them behind paywalls, removing the public knowledge part of this.
dumbass@piefed.social · 2 pts · 217d
Fuck, we really should have donated all those times.
deHaga@feddit.uk · 9 pts · 217d
I did donate and they still did this
dumbass@piefed.social · -2 pts · 217d
I'd request a refund, fuck em.
Skiluros@sh.itjust.works · 3 pts · 217d
I would argue this is a good thing. The data is being used anyways, at least they get some money to keep the project going.
hector@lemmy.today · 0 pts · 217d
Wikipedia has an owner? I thought it was all non profit ey or some shit.
Humanius@lemmy.world · 9 pts · 217d
Non-profits also have owners. It being a non-profit just means that the company's main priority isn't turning a profit, so excess revenue generally gets pumped back into the non-profit.
Skiluros@sh.itjust.works · 4 pts · 217d
The owner, Wikimedia Foundation, is a non-profit.