Three spam updates since June: an audit for AI-written client content
By Alan Magliocca · Published November 2026
Quick answer: Google's Search Status Dashboard lists three spam updates between late June and early October 2026, two of them since August: August 18 and September 24, the second running until October 8. Line up each client's traffic against those dates, then audit AI-written page groups against Google's scaled content abuse and doorway policies before the November core update settles.
Key takeaways
- Google ran spam updates starting June 24, August 18 and September 24, 2026, plus a short one on March 24.
- The September update took 13 days and 16 hours and finished October 8, the longest spam update of 2026.
- Google named no target for either update, and the September update did not target link spam.
- Scaled content abuse is about pages made mainly to rank without helping users, and Google lists generative AI as one way to produce them.
- City and service pages built from one template are the AI-written content most worth checking first.
What happened
Google's Search Status Dashboard lists four spam updates in 2026. Three of them ran between late June and early October, and two of those started after August 1:
- June 2026 spam update: started June 24 and ran about two days.
- August 2026 spam update: started August 18 and ran 2 days and 16 hours. Search Engine Roundtable reported it finished on August 21.
- September 2026 spam update: started September 24 and ran 13 days and 16 hours. Search Engine Roundtable's completion report says Google confirmed it was complete on October 8.
The fourth, the March 2026 spam update, started March 24 and lasted under a day.
Google described the September update as a normal spam update for all languages and locations. It didn't name a target for it or for the August update, and Search Engine Roundtable notes the September update was not a link spam update. On November 10, Google started a different kind of update, the November 2026 core update, which may take up to two weeks.
What it means for agency owners
If you write content for clients with AI, Google's spam policies are the rulebook these updates enforce. The one that matters most is scaled content abuse: many pages generated mainly to manipulate rankings rather than help people. Google's examples include:
- using generative AI or similar tools to produce many pages that add no value for users;
- scraping feeds or search results, including by swapping synonyms or translating, with little added value;
- stitching together content from other pages without adding anything;
- spreading content over several sites to hide how much of it there is.
The test is the purpose and the value of the pages, not the tool that wrote them. A batch of AI drafts that a human checked and filled with the client's own facts is a different thing from 200 lightly varied pages published in a week.
Two other policies catch common agency work. Doorway abuse covers many pages or domains aimed at different cities or regions that all funnel people to one page. That describes plenty of "plumber in [town]" page sets. Thin affiliation covers product pages with descriptions and reviews copied straight from the merchant.
Timing also matters for client conversations. A drop that began on August 18 or September 24 points to a spam update. A drop that begins after November 10 is likely to be blurred by the core update, which we cover in what to tell clients during the November core update.
What to do this week
- Pull Search Console data for each client from August 1 to November 9, by day. Mark August 18 to 21 and September 24 to October 8. Stopping at November 9 keeps the core update out of the picture.
- Open the Manual actions report. A manual action is a person at Google flagging the site. It is a separate problem from an algorithmic drop and has its own fix, a reconsideration request.
- List every AI-assisted page group: city pages, service pages, blog batches, product descriptions. Note how many pages each has, when they were published and whether they share a template.
- Score each group with the checklist below and decide whether to improve, merge, noindex or remove it.
- Hold large changes until the core update ends. Google said it may take up to two weeks from November 10, so plan to ship after about November 24.
- Write a one-page content rule for every client: what a human checks before publishing and which first-hand details each page needs.
How AIVX Labs helps
AIVX Labs doesn't read Search Console or tell you whether Google classed a page as spam, so the traffic check above happens in each client's account. The Website Analyzer scores a site out of 100 and lists fixes in order, and the Fix Package turns that list into something you can hand the client. The website audit and fix package demo shows RICO running both from one plain-language request. The AI Content Studio drafts pages, and those drafts still need the client's own facts and a human edit before they go live.
Which AI-written pages are most at risk?
Use these questions on each page group. The more "yes" answers, the higher the priority.
- Same template, swapped words? If the only difference between pages is the town or service name, the pages look like doorways.
- No first-hand detail? No photos of real jobs, no named staff, no prices, no local specifics such as permits or response times.
- Published in a burst? Dozens of pages in a few days, with no internal links pointing to them.
- Answers already on page one? If the page restates what every competitor says, it adds nothing.
- No one owns it? No author, no review date, and nobody at the client could explain the claims on it.
- Traffic pattern matches the dates? Clicks started falling on or just after August 18 or September 24 and didn't recover.
Pages that pass are worth keeping. Pages that fail on three or more points are the ones to merge or rewrite first.
A worked example: a plumber with 60 city pages
Picture a plumbing client with 60 pages titled "Emergency plumber in [town]," all written from one AI template last spring. Search Console shows clicks to that folder dropping in the days after September 24.
A sensible clean-up looks like this:
- Keep the towns the client actually serves often. Ask which 10 to 15 towns produce most jobs.
- Rewrite those pages with real detail: typical response time, recent job photos, local permit notes, reviews from customers in that town.
- Merge the rest into one service-area page that lists every town, and redirect the old URLs to it.
- Link the kept pages from the main service page so they sit in a clear structure.
- Record what changed and when, so you can compare traffic once the core update has finished.
This is slower than generating 60 pages, but it produces pages a customer would find useful, which is what the policy asks for. The same thinking applies to AI search: pages with specific, verifiable facts are the ones assistants can quote, as BrightStage's guide to generative engine optimization explains.
If a client also hosts content written by other companies, check it against the separate site reputation abuse policy, which now works differently in the EU.
Sources
- Google Search Status Dashboard: Ranking incident history, 2026-11-10
- Google Search Central: Spam policies for Google web search, 2026-08-28
- Search Engine Roundtable: Google September 2026 spam update done rolling out, 2026-10-08
- Search Engine Roundtable: Google August 2026 spam update done rolling out, 2026-08-21
Related
Written by Alan Magliocca, founder of AIVX Labs and BrightStage AI (Podavinci LLC). Written with AI assistance and checked against the linked sources.
Read our featured article on LinkedIn
Get AIVX Labs verified for $1: RICO, your AI operator, with 7 days of full access.
Get AIVX Labs Verified for $1Follow along:LinkedInYouTubeTikTokXFacebookInstagramSkool community
Frequently asked questions
Does Google penalize content because AI wrote it?
Google's scaled content abuse policy is about pages made mainly to manipulate rankings without helping users. Generative AI is listed as one way to produce such pages, but the policy targets the purpose and value, not the tool.
How do I tell a spam update drop from a core update drop?
Compare the date the decline started with the dashboard. The 2026 spam updates started June 24, August 18 and September 24, and the November core update started November 10.
Is a spam update the same as a manual action?
No. A spam update is an automated change, while a manual action is applied by a person at Google and appears in the Manual actions report in Search Console, where you can request reconsideration after fixing it.
