It has been know for a long time that, when it comes to AI training, quality is much more important than quantity. Yet Big Tech goes for quantity because it is cheaper and takes less time to train a model on non-curated garbage scrapped from the internet than taking the time and the (human) effort of curating the data. That is one of the problems with their models: Garbage in, garbage out.

bloomberg.com/news/articles/20

Thread 1/3

0

If you have a fediverse account, you can quote this note from your own instance. Search https://mstdn.social/users/pretergeek/statuses/113499466115257447 on your instance and quote it. (Note that quoting is not supported in Mastodon.)