Alternately, will you start using royalties in perpetuity whenever you talk about some event, because you read an article or a book about that topic once and included something you learned in that article?
Basically everything you know, that is even somewhat recent is based on others’ content, do you track and cite every single thing you’ve ever read and send them royalties with every conversation?
I’m not trying to defend these big corporations, but for me this is a fundamental question we need to be asking.
As consequential as it will be, for me, the answer is that as long as you paid the cost of accessing the content (be it free or a subscription price) while collecting the information that is used to fundamentally transform the information in ways that seem to fall under fair use, then you cannot expect rights, short of full copy/paste plagiarism.
Dumb nit, but why not put your own press release through your model to prevent basic things like missing quote marks? Reminds me of that time an OAI released wildly inaccurate copy/pasted bar charts.