I was wondering... Common Crawl [0] exists, why do "AI companies" do this on their own? Is Common Crawl data not fresh enough? Is there some other deficiency?

I was wondering... Common Crawl [0] exists, why do "AI companies" do this on their own? Is Common Crawl data not fresh enough? Is there some other deficiency? Whatever the problem is, a single crawler which every AI company can reference seems like a compromise to solve this issue, doesn't it? [0] https://commoncrawl.org/ https://commoncrawl.org/