Google SEO October 2026: Stricter AI Rules, No Core Update

A quiet dashboard over a restless search market

In October 2026, Google's organic search changed mostly through policy rather than ranking algorithms. Nothing new appeared on Google's Search Status Dashboard. The September 2026 spam update began on September 24 and was marked complete on October 8, after 13 days and 16 hours. Google did not announce a core update in the weeks that followed.27 By October 19, the spam update was still the newest ranking entry on the dashboard, even as forum posters described falling Discover traffic, Google News results stuck on old stories, and Search Console data running days late.22 Coverage from trade outlets, agencies and independent blogs agreed on the dashboard facts. As of mid-October, Google's only confirmed broad core updates for 2026 were in March and May.26

The disagreement is about what the volatility means. Some prominent practitioners, including Lily Ray, Glenn Gabe and Marie Haynes, expect a major update soon. That is a prediction, though, and Google has made no such statement.27 One of the more careful readings argues that the complaints don't fit a single ranking event. Discover and Google News use systems separate from core web ranking, and many site owners date their problems to late August, a month before the spam update started.22 The same analysis notes that rising visibility scores alongside flat clicks look more like a changing results page than lost rankings. More AI Overviews, ads or local packs can take clicks without moving a site's position.22

That reading holds up best. October's traffic losses are mainly about how the page around the blue links is changing, not the blue links themselves.

The real October news was written in documentation

Google's most important moves this month were edits to its guidance. On October 1, Google revised its guidance on generative AI content, adding language from the Search Quality Rater Guidelines. The revised text calls it "critical" to manually fact-check and review all AI-generated content before publishing.27 Several outlets stressed that this is not a ban on AI writing. Google said the change brings its documentation in line with the rater guidelines and what it has said at events, so it is not a new ranking update aimed at AI text.2126

Google's guide to helpful, people-first content changed the same week. It now defines main content as the part of a page that directly serves its purpose, whether that's an article, a calculator or a tool, and lists four qualities raters assess.21 Those qualities are effort, originality, skill and accuracy.27 The guidance on authorship got noticeably stricter. Made-up names, AI-generated headshots and false credentials are now described as deception and called a signal of a low-quality page.28

Read together, these edits show Google putting its long-standing quality ideas into writing at a time when scaled, AI-generated content is common. At an event in Barcelona on October 5, Google said more spam updates are coming, that AI is helping it catch spam, and that scaled content is a bigger problem than link spam.27 Marie Haynes reported a jump in requests for help with manual actions, mostly over thin and scaled content. Google has not confirmed any wave of enforcement, and the evidence so far is anecdotal.27

Technical SEO: timelines, crawlers and broken trackers

Several technical details this month are directly useful. Google analyst Gary Illyes shared rough timings: about 20 hours to discover a new URL, about 1.5 hours to index a page when indexing succeeds, and anywhere from months to never when quality is the obstacle.21 John Mueller said AI crawlers generally give site owners no way to submit sitemaps. They find them the usual way, through robots.txt or standard filenames such as sitemap.xml, so non-standard naming can hide a site from those bots.28

Measurement went wrong in places, too. Morningscore reported that from September 24, Google returned shortened results pages to its automated rank tracker, affecting up to about 60% of its queries at the peak. Missing results were logged as "not ranking."27 Anyone who saw tracker losses during the spam-update window should check them against Search Console before acting. Search Console had problems of its own: the performance report fell more than 60 hours behind around October 11.21 Google also added a filter that combines data from several countries or devices in one view.28

AI Overviews: the numbers disagree, the direction doesn't

The estimates of AI Overviews' impact vary widely. Estimates of how often AI Overviews appear range from more than 20% of searches23 to about 48%18. Ahrefs' September crawl found them on 73% to 78% of US desktop searches.19 DemandSphere and Ahrefs data showed them on 82.91% of branded queries on September 29, which suggests that even brand-name searches, long considered safe, now often start with an AI summary.27 Much of this spread comes from differences in method, such as desktop versus mobile, query mix and sampling, so no single figure should be treated as definitive.

The impact on clicks is more consistent. Ahrefs measured a 58% lower click-through rate for the top result when an AI Overview appears, and Seer Interactive found organic CTR down 61% on queries with AI Overviews.16 A First Page Sage study covering 3.4 billion sessions found organic clicks per thousand impressions fell 58.6% when an Overview was present, with media and publishing losing 47.3% of organic sessions.11 The most rigorous evidence gives a smaller but still large number. A difference-in-differences study of English Wikipedia found AI Overview exposure cut daily traffic by about 15%.13 An experiment that hid the Overview from the same users found outbound organic clicks rose from 0.37 to 0.62 per search.12

There are some signs of partial recovery. Seer's 2026 data showed CTR on AI Overview queries rising from 1.3% in December 2025 to 2.4% in February 2026, though still well below where it was before AI Overviews.12 Brands cited inside an Overview earned roughly 120% more organic clicks per impression than brands that weren't cited.19 For SEO teams, the practical goal is now being cited inside the AI answer, alongside ranking below it.

The effects are also uneven. Chartbeat data showed small publishers losing about 60% of referral traffic over two years, compared with 22% for large publishers.14 Google, for its part, says organic clicks have been relatively stable year over year but hasn't published data to support that.19 Its Search Console AI report shows impressions but not clicks or queries.21 One site's nine months of tracking found 22.4% of its AI Overview traffic recorded as Direct rather than Organic Search.12 In practice, publishers can't check Google's claims against their own data, and part of the AI-driven traffic they do get is being credited to the wrong channel.

Google's counter-moves

Google appears aware of the pressure. On October 15, Hema Budaraju announced that inline links will appear higher in AI Mode and AI Overviews more often. The same post described tests of a block for official websites, wider news carousels, a "Dive Deeper" section, and clickable creator names in video results.23 Google said more than 400 publications are now in its News AI pilot licensing program, and that use of Preferred Sources has grown sevenfold this year.23 One roundup reported that the payments from Google's AI contribution pilot are very small.26 On October 21, Google combined its existing site-wide AI controls and page-level preview rules into its guide on sharing content with Search.22

Other changes point toward more AI on the results page. Google brought Gemini 4 Argon into AI Mode. Some users also saw a test of an AI Mode-style sidebar on the standard results page, though it couldn't be reproduced.26 If that sidebar rolls out widely, AI Mode will be one click away from every ordinary search.

Is GEO dead, or just SEO with a new name?

Meanwhile, the debate over what to call this work got heated. On May 15, 2026, Google said in official documentation that optimizing for generative AI features is "still SEO." It dismissed tactics such as llms.txt files and artificially chunked content.8 Hobo Web built on that in October and called generative engine optimization a rebrand of SEO, AEO and digital PR.8 A response from Novel Cognition checked the studies Hobo relied on. It found much of the critique of the original 2023 GEO paper holds up, but that a quote attributed to C-SEO Bench isn't in the paper.7 The response also highlighted a University of Toronto study finding that 97% of pages ChatGPT retrieved weren't in Google's top 50 for the same queries.7

Both sides have a point. For Google's own AI features, which pull from Google's index, the claim that this is "still SEO" is basically right. For ChatGPT and other assistants that use their own indexes, ranking well on Google clearly doesn't guarantee being cited. Bing's documentation does use the term "generative engine optimization."1

Investors and buyers aren't waiting for the debate to settle. Profound raised a $180 million Series D at a $1.8 billion valuation in September. Adobe completed its $1.9 billion purchase of Semrush in April, and Sitecore reportedly paid $225 million for Scrunch in June.10 GEO has also spread beyond marketing: Reuters reported that political consultants are now selling it to campaigns that want chatbots to describe their candidates favorably.2

The fundamentals Google keeps stressing still decide whether a page can be found at all: indexability, original work, real authors and verified facts. What has changed is the payoff. Ranking first pays off less than it used to, and being cited in AI answers matters more. Right now that is the outcome SEO teams have the most trouble measuring.