SEO Recap covering October 1, 2026: Google's AI fact-check rule and Penske lawsuit dismissal
Daily summary of what matters in SEO, GEO, AEO, and AI search generated with Claude Code (beware of hallucinations)
Generative Engine Optimization (GEO) & Answer Engine Optimization (AEO)
AI changed the buyer faster than businesses can adapt
Source: Search Engine Journal
- Pew surveyed 5,119 US adults in February and found ~50% now use AI chatbots, up from ~33% in 2024, with ~25% using them daily and information search the most common use. ChatGPT use rose from 18% of adults in 2023 to 44%.
- Buyers arrive mid-decision after research, comparison, and shortlisting already happened in a chat window, so pages built to win 10 SERP slots no longer match how models pick 3 names.
- Catalog pages are now read and summarized rather than visited, making their structure, consistency, and agreement with other sources matter more than page count.
- Referrer-based tools (GA4's AI Assistant channel added in May, Bing's AI Performance report in February) only capture the tail of a decision and can't measure influence where no click occurs.
- Reverse the usual order: learn what AI is saying about you first, decide what belongs in the answer, and choose metrics last.
AI in Search / AI Overviews
Google's AI contribution pilot pays ~100 publishers, often under 0.1% of ad revenue
Source: Search Engine Journal
- The Information reports Google is paying roughly 100 digital publishers through its AI contribution pilot, with payouts ranging from under $1,000 over several months to more than $1 million a year.
- For several small and midsize sites the payments amount to less than 0.1% of advertising revenue, and some participants say amounts change monthly without explanation.
- Content earns when it "contributes significantly" to responses in the Gemini app, AI Overviews, and AI Mode, but Google still hasn't defined "significantly" or shown how payments are calculated.
- Topics with strong interest but thinner coverage, such as anime and gaming, appear to earn more, and the low figures have kept some larger publishers out.
Google Gemini adds UTM parameters to outgoing links for referral attribution
Source: Search Engine Journal
- Google quietly added UTM parameters to Gemini's outgoing links so site owners can attribute referral traffic that otherwise lands in GA4's "direct" bucket.
- A Reddit user noted reported referrer pass-through runs ~9% from the Gemini mobile app and is stripped entirely by Android assistant invocations, while a UTM survives the in-app webview and shows in server logs.
- John Mueller confirmed seeing the change and asked for reproducible details on referrer behavior to forward to the team, wanting referrers retained alongside UTM tagging.
- The implementation is undocumented, so it's unclear exactly when Gemini triggers the UTMs, mirroring how ChatGPT only tags web-grounded answers.
Technical SEO
Google adds manual fact-check requirement to generative AI content guidance
Sources: Search Engine Journal · Search Engine Roundtable
- Google revised its "guidance on using generative AI content" page on Oct. 1, calling it "critical to manually factcheck and review all AI-generated content for accuracy and trustworthiness before publishing."
- The new wording explains the reasoning, noting generative models "don't retrieve facts, but predict a likely sequence of words," so outputs may contain hallucinations.
- The manual review now explicitly extends to metadata, covering title elements, meta descriptions, structured data, and image alt text that can appear in search results.
- Google says the edit syncs the doc with its developer-event talks and pulls in language from the Search Quality Raters guidelines, arriving during the September 2026 spam update.
Common structured data mistakes that hurt AI visibility
Source: Search Engine Journal
- Treating schema as a checklist rather than an entity strategy misses what LLMs want, which is reassurance of entities and relationships. Link articles to author and organization schema to build context across the site.
- Don't rely on schema alone to clarify entities. Marked-up data must align with what LLMs ingest elsewhere, so correcting outdated brand names and pricing across the web is part of the strategy.
- Use consistent @id entity identifiers so a single organization definition carries across pages and templates, preventing conflicting information.
- Avoid schema that doesn't match visible content, such as review ratings or stock and price values that contradict the page, which risks a Google manual penalty and confuses AI systems.
Organic Search & Algorithm Updates
Judge dismisses Penske and Chegg antitrust claims over AI Overviews
Sources: Search Engine Journal · Search Engine Roundtable
- Judge Amit Mehta dismissed the amended complaints from Chegg and Penske Media without prejudice, ruling the plaintiffs failed to meet federal antitrust requirements.
- Mehta rejected all 5 claim categories: reciprocal dealing, tying, unlawful monopoly maintenance, attempted monopolization and leveraging, and unjust enrichment.
- On reciprocal dealing he wrote "an expectation is not an agreement," finding no negotiated bargain to trade content for referral traffic
- Publishers lacked antitrust standing in the general search market, and did not define the publishing markets Google allegedly sought to dominate
- The judge said he was "not unsympathetic" to publishers and journalists "whose content Google takes and repurposes without compensation."
- A separate Sept. 30 ruling let Gannett, Daily Mail and a publisher class proceed on Google ad-tech claims, which are unrelated to AI Overviews.
Google September 2026 spam update rolls out in multiple waves
Source: Search Engine Roundtable
- The September 2026 spam update was announced Sept. 24, with the first impact wave Sept. 25-27 and a second wave flaring up Sept. 30 into Oct. 1.
- Unlike prior spam updates that rolled out in ~2 days, this one is expected to take about 2 weeks with multiple cycles of volatility.
- Glenn Gabe reported Reddit's AI translations surging back in Italy, Germany, Brazil and France after crashing during the May and June updates.
- Site owners reported sharp drops on Sept. 30, with Google Discover and News traffic repeatedly vanishing during the rollout.
Cloudflare moves Monetization Gateway and Pay Per Use into beta
Source: Search Engine Journal
- Cloudflare put 2 AI payment products into closed beta on Sept. 30 for US sellers and buyers.
- The Monetization Gateway lets sellers charge AI agents per request to an API, MCP tool, dataset, or site, with seller-set prices.
- It uses the x402 protocol and HTTP 402 status code, settling in USDC on the Base blockchain
- Settlements run from $0.001 up to $100, with fixed or variable (capped) pricing
- Pay Per Use compensates publishers after an AI company reports using their content, with the AI company defining the use and price and publishers choosing whether to accept.
- Usage under Pay Per Use is self-reported by buyers via timestamp, source URL, and event ID, and Cloudflare bills buyers and pays publishers monthly. Neither product reports transaction counts or payouts yet.
Why Google may be moving like Microsoft in the AI race
Source: Search Engine Journal
- Google's promised Gemini 3.5 Pro slipped from a June target and still hadn't shipped by late September, with Bloomberg reporting its coding fell short of targets.
- Comscore's US panel shows Gemini's share of AI assistant prompts rising from 17% in January to 30% in June while ChatGPT fell from 70% to 50%.
- Menlo Ventures' 2025 enterprise report put Anthropic at 40% of enterprise LLM API spend versus OpenAI's 27% and Google's 21%, and 54% to 11% in coding, though Menlo backs Anthropic.
- The argument is that Google lacks the motivation to build an assistant that makes its results page optional while Search revenue grew 17% to $63.3 billion, echoing how Microsoft's Bing distribution never translated into search leadership.
Published on