Judge Amit Mehta dismissed two antitrust suits from Penske Media and Chegg that blamed Google's AI Overviews for draining their web traffic. Lost search clicks are now a problem for the people who run a site's search settings and its licensing talks.
Perspective Coverage
6 publishers
- Builder
- Builder 15%
- Operator
- Operator 43%
- Investor
- Investor 42%
Reality
- Evidence82
- Adoption
- Insufficient
- Hype gap+5
- Incentives45
- Confidence78
Google Search Console now counts page impressions in AI Overviews and AI Mode, in reports that reached all websites worldwide by August 31, 2026. Teams can confirm their pages show up in AI answers, but linking those appearances to visits takes other data.
Reality
- Evidence55
- Adoption
- Insufficient
- Hype gap−5
- Incentives55
- Confidence50
Google is testing Flipkart purchases inside Gemini in India, six months after its shopping tools there still sent buyers to outside sites to pay. Flipkart stays seller of record, so Google gains the screen where shoppers choose while the sale remains Flipkart's.
Perspective Coverage
3 publishers
- Builder
- Builder 33%
- Operator
- Operator 32%
- Investor
- Investor 35%
Reality
- Evidence58
- Adoption12
- Hype gap+25
- Incentives60
- Confidence55
Google is testing a Gemini Buy button that sends Indian shoppers to checkout at Flipkart, where it holds a stake bought for about $350 million in 2024. Retailers selling through AI assistants now have to choose whose checkout screen they will accept inside the answer.
Reality
- Evidence55
- Adoption10
- Hype gap+5
- Incentives65
- Confidence55
A preregistered trial pushed one arm's entire seven-day query stream into AI Mode. The 18.8 point drop in outbound clicks is therefore the effect at full substitution, and a publisher's own exposure share scales it.
Reality
- Evidence52
- Adoption
- Insufficient
- Hype gap+14
- Incentives68
- Confidence46
Semrush's 2026 index of 126 million prompts puts cross-platform brand overlap at 67% and cited-source overlap near 30%. Those figures compare two platforms against each other, and Semrush gives no rate for the mention-citation mismatch.
Reality
- Evidence30
- Adoption22
- Hype gap+25
- Incentives80
- Confidence40
Robby Stein said the Discover experiment starts on video cards and opens a short topic overview with links to related stories. Google has published no referral data, so publishers cannot size the effect yet.
Reality
- Evidence40
- Adoption10
- Hype gap+15
- Incentives70
- Confidence42
A query-length dataset running to August 2026 has 1-2 word searches down to 24% of impressions while still holding 52% of conversions, and its two shortest buckets already account for 98% of the conversion total.
Reality
- Evidence25
- Adoption15
- Hype gap+30
- Incentives70
- Confidence35
The number comes from a Forbes Tech Council post by a marketing agency director, and it is the only measurement in the record behind the claim that rank tracking and AI citation tracking now count different populations.
Reality
- Evidence24
- Adoption
- Insufficient
- Hype gap+46
- Incentives84
- Confidence28
Google is testing monthly payments for content that grounds Gemini, AI Overviews and AI Mode, with the figure already visible to selected publishers in Search Console and the payment formula still undisclosed.
Reality
- Evidence32
- Adoption12
- Hype gap+25
- Incentives68
- Confidence38
Productrise measured Google's two shopping surfaces side by side for three weeks in August and found they barely show the same products. Merchants who rank in one have little claim on the other, and AI Mode's prices run higher.
Reality
- Evidence32
- Adoption
- Insufficient
- Hype gap+40
- Incentives60
- Confidence30
Google's AI Mode began completing hotel bookings in the US in late August, with Booking.com, Expedia, Hilton, Marriott and IHG supplying rooms. The published evidence about what goes wrong is one chatbot case from 2022.
Reality
- Evidence48
- Adoption58
- Hype gap+14
- Incentives42
- Confidence55
The engines lean on Reddit, YouTube and LinkedIn, which means the pages deciding whether an assistant recommends your product sit on properties where your only interface is a post and somebody else's moderation queue.
Reality
- Evidence22
- Adoption30
- Hype gap+42
- Incentives78
- Confidence52
Google's June switch lets publishers leave AI summaries without touching their rankings, and the Commission's July questionnaire, due August 28, asks whether a control nobody can measure counts as a choice.
Reality
- Evidence42
- Adoption
- Insufficient
- Hype gap+12
- Incentives72
- Confidence48
Aggarwal et al. report visibility gains of up to roughly 40% inside generated answers, but the benchmark supplies the source documents, so the number scores selection and leaves crawl and index eligibility upstream of anything you can format.
Reality
- Evidence52
- Adoption28
- Hype gap−12
- Incentives40
- Confidence48
A Search Engine Land analysis published in February 2026 found that repeating the same ChatGPT prompt returns different brands, which makes visibility a proportion you sample. The arithmetic of that proportion sets how many runs your report needs.
Reality
- Evidence30
- Adoption
- Insufficient
- Hype gap+15
- Incentives78
- Confidence38
Google's generative AI performance reports went worldwide on August 31, 2026, so web teams get a first-party count of URL appearances in AI Overviews, AI Mode and Discover, sliced by page, country, device and date.
Reality
- Evidence36
- Adoption44
- Hype gap+14
- Incentives78
- Confidence45
US hotel checkout now completes inside AI Mode, with the hotel or travel site left as merchant of record for refunds, disputes and the support desk, while Google takes the conversion event and the partner keeps the cost of what follows.
Reality
- Evidence44
- Adoption28
- Hype gap+32
- Incentives66
- Confidence48
Promptwatch ties the collapse to an August 8 jump in domain-scoped fan-out queries. The data is provisional. The dependency it exposes is not.
Reality
- Evidence32
- Adoption38
- Hype gap+34
- Incentives72
- Confidence44
A SparkToro estimate puts clickless US Google searches at 68.01% for the first four months of 2026. The counters that survive that are citation and what a visitor does after the answer.
Reality
- Evidence26
- Adoption
- Insufficient
- Hype gap+18
- Incentives
- Insufficient
- Confidence28