Hacker News Reader

Comment by wvos | Hacker News Reader

I believe AI uses publicly accessible content similarly to how humans do: we discuss what we read, use it, and create new content from it. This process is legal and acceptable. The fact that an artificial entity is doing it doesn't make it wrong. Moreover, I think intellectual property shouldn't be owned. In today's world, intellectual works should automatically benefit society.

View on HN

That’s fine, but then we need to stop privatizing the AI. If a coalition of companies (and maybe government?) were to train open model weights, and run those models on equal-access hardware (i.e. all residents of the country get N compute credits per year) then we’d be much closer to a system where “intellectual works should automatically benefit society” — however, that is not what we are seeing today.

Why? I don’t need to equally and freely share the expertise I develop by consuming publicly available information. In fact, I personally profit from it. Should I compensate every YouTube creator, author, journalist, for the money I’ve made in my career that their publicly available work contributed to in terms of my learning/education?

> AI uses publicly accessible content similarly to how humans do

It absolutely does not. Bunch of companies with pretty low ethical standards are scraping all they can find. People who wrote all this content never consented to this, and many of them explicitly opted-out.

Reply on news.ycombinator.com