Even if OpenAI is reproducing pieces of NYT articles, they still have a difficult argument because in no way is is a practical means of accessing paywalled NYT

Even if OpenAI is reproducing pieces of NYT articles, they still have a difficult argument because in no way is is a practical means of accessing paywalled NYT content, especially compared to alternatives. The entire value proposition of the NYT is news coverage, and probably 99.9% of their page views are from stories posted so recently that they aren't even in the training set of LLMs yet. If I want to reproduce a NYT story from LLM it's a prompt engineering mess, and I can only get old ones. On the other hand I can read any NYT story from today by archiving it: https://archive.is/5iVIE https://archive.is/5iVIE. So why is the NYT suing OpenAI and not the Internet Archive?