Google shipped its biggest-ever Search model upgrade during the November core update, OpenAI's ChatGPT Business Hub began reporting real click data, and both platforms kept packing ads deeper into AI answers.
This week Google stacked major changes on top of each other: it launched the November 2026 core update and, days later, pushed Gemini 4 Hydrogen — which it calls the biggest model upgrade to Search yet — while a Search Console AI-report data bug muddied everyone's measurement at the worst possible time. On the AI-search side, OpenAI's new ChatGPT Business Hub started showing brands their mentions, profile views and clicks, Yahoo Scout left beta with a plan to pay publishers, and ads kept spreading — ChatGPT visual ads during image generation, Target CPA/CPC bidding, and Google's own AI Mode multiturn ads. For site owners the takeaway is to ride out the core update without knee-jerk changes, start treating AI assistants as measurable referral channels, and watch how aggressively both Google and OpenAI are monetizing AI results.
Google rolled a new model, Gemini 4 Hydrogen, into Search and called it the largest model upgrade Search has ever had, citing faster responses, better query fan-out, stronger linking and more visual, multilingual answers. It landed in the middle of the November core update, making ranking movements especially hard to attribute this week.
Google began its November 2026 core update on November 11 and said it should take about two weeks to finish, wrapping up just before Black Friday. Sites seeing volatility are advised to wait for the rollout to complete before diagnosing or reacting.
Just before Hydrogen, Google made Gemini 4 Carbon — pitched as its strongest model for cybersecurity and long-horizon agentic work — available inside Search's AI Mode for subscribers. It marks another step in Google swapping faster, more capable models directly into the search experience.
Google confirmed a logging bug in the Search Console Generative AI performance report between November 9 and 11 that understated impressions. The timing — right as the core update began — means that window's AI-search data should not be trusted for analysis.
Google is trialing placing the citation and source icons at the very bottom of AI Overviews instead of the top. Burying the links further could make it even harder for users to notice and click through to the sites that supplied the answer.
Google is testing a stripped-down local pack inside AI Overviews that removes action buttons, photos and reviews. That gives searchers fewer ways to call, visit or get directions to a business directly from the results.
Google's John Mueller reiterated that pages stuck at 'Crawled/Discovered — currently not indexed' in Search Console often mean Google doesn't consider them high-quality enough to index. The fix is improving the content and site value, not forcing indexation.
In the final Non-Canonical Cabinet video, Gary Illyes and Chery Prommawin explained why moving to a CDN can spike Googlebot crawling and overload your origin server. They outline how to migrate without accidentally blocking Google's crawlers.
A court handed SerpApi another win in its legal fight with Google over search scraping, giving Google 21 days to amend its complaint. The case looks like it may be heading toward a close, with implications for how SERP data can be collected.
OpenAI launched the ChatGPT Business Hub, giving brands analytics on mentions, profile views and actual clicks from ChatGPT to their websites. It's the clearest signal yet that AI assistants can be treated as a measurable referral channel rather than an invisible one.
Yahoo moved its Scout AI search out of beta, launched Pivot Search, and said Scout has processed over 3 billion answers. Separately, Scout became the first commercial partner of the new Internet Licensing Collective, which will pay publishers when their content is used in AI answers.
OpenAI's new visual ad format appears to be live: asking ChatGPT to create an image can now surface an ad while the image renders. It's part of a steady expansion of advertising inside ChatGPT's core interactions.
Similarweb data cited by Business Insider shows Reddit's traffic falling about 13%, with Google's AI Overviews pointed to as a factor. It's a notable swing for a site that many SEOs treat as a key source of AI and search visibility.
A test of Gemini, ChatGPT and Perplexity shopping features found they sometimes recommend or link to risky and fake websites. For brands, it's a reminder to monitor how AI assistants represent your products and whether they're being confused with lookalike sites.
Ahrefs published research on which brands get named most often in ChatGPT responses, offering a data-driven look at AI 'share of voice.' It's useful input for benchmarking your own brand's visibility in generative answers.
Google Ads introduced an Invalid Traffic overview card that shows advertisers how it detects and credits back fake or invalid clicks. It's meant to give more front-end transparency into click-fraud protection inside the interface.
OpenAI's weekly ChatGPT Ads update brought intelligent bidding with Target CPA and Target CPC options, new cost targets in Ads Manager, more geographic flexibility and expanded product-set filters. The platform is quickly maturing toward parity with established ad tooling.
Google began rolling out multiturn ads inside AI Mode, placing sponsored results across the back-and-forth of a conversational search session. It extends paid placements deeper into Google's AI-driven results.
Google released Google Ads API version 26.2, touching assets, campaigns and reporting, alongside Ads Editor 2.14, which adds 10 new features including AI Max defaults and new retention controls. Advertisers and tool builders should plan upgrades accordingly.
Google clarified that using AI to impersonate public figures in ads is prohibited, even when the ad discloses that it was AI-generated or edited. The update tightens its unacceptable business practices policy around synthetic likenesses.
Google Ads introduced a new 'Ads with pay-per-lead goals terms' option that advertisers must explicitly opt into to activate. It signals continued expansion of lead-based goals and billing models within Google Ads.
Ahrefs detailed DR4, a rebuilt version of its Domain Rating metric that scores fewer sites, rates subdomains on their own, and correlates more strongly with ranking. Expect scores for many domains to shift when it lands, so recalibrate any link targets or reporting that lean on DR.
Analysis of French Search Console data shows AI Overviews driving significant click and click-through-rate losses, though the impact varies widely by site and query type. It's some of the clearest first-party evidence yet of AI Overviews' effect on organic clicks.
New SparkToro research estimates that traffic to the open web has dropped roughly 24% over the past five years as platforms and AI answers keep more clicks for themselves. It frames the long-term squeeze publishers and SEOs are navigating.
A new study finds that AI-driven search activity barely shows up in Google Ads search terms reports, leaving advertisers with a large blind spot. Measuring AI's influence on demand will require looking beyond standard Ads reporting.
In its Q3 results, publisher Ziff Davis disclosed that organic Google search referrals fell about 37% year over year, and management signaled it isn't expecting a recovery. It's another data point that AI-era traffic declines for content sites may be structural rather than temporary.
Turn the week's news into action. Ticks are saved in this browser only.
New writing, experiments, and the occasional story — straight to your inbox. Two to four emails a month.
Compiled from credible SEO/search publications and primary sources; each item links to its source. A new digest publishes every week — see all issues.