Mastodon Feed
cwebber@social.coop ("Christine Lemmer-Webber") wrote:
The irony of the book-shredding is AI companies are looking for as much pre-LLM tainted output as they can get to train their LLMs.
Non-LLM affected content is the most valuable for feeding the LLMs themselves. Does this mean that Codeberg, by being a code hosting system opting out of LLM-generated content, becomes the most desirable source of content to feed the LLMs themselves? :\