A forum owner's Google Search Console screenshot has become one of the most argued-over SEO cases of the week. The notice, message code WNC-651700 and dated around 17 July 2026, reported a manual action for "Thin content with little or no added value" against a single URL pattern on windowsforum.com. Google gave one reason. A second theory about what really triggered it is circulating among SEOs, and that theory is speculation, not anything Google has confirmed.
The case was written up by Roger Montti for Search Engine Journal on 27 July 2026, after the site owner posted the notice to Reddit. Note the hedge in that outlet's own headline: Google "may be" penalizing AI content. Outside Google, nobody knows.
What the manual action actually did
This was a partial manual action rather than a sitewide one. A human reviewer at Google looked at part of the site, judged that part unworthy of ranking, and suppressed it pending a reconsideration request. The rest of the domain was left alone. Suppressed pending review is the accurate description here, and it is a different thing from being removed from the index.
The Affects field named exactly one URL pattern: windowsforum.com/threads/. Behind that pattern sit 168,290 visible threads and over half a million posts, twenty years deep, all suppressed together. Google supplied no example URLs. That is what makes these notices hard to act on: you are told a directory is thin, not which pages the reviewer opened.
The part that confused the site owner
Two sections of the same site look far more obviously thin than the discussion threads, and neither was actioned. One is a syndicated headline feed. The other is an archived knowledge base. Together they account for 11,927 threads with a mean age of about 12.5 years, and both have been inactive for three years.
As the owner put it, that material was crawled, indexed and ranked continuously through Panda, through Penguin, through the September 2023 helpful content update, and through a migration from the original domain, without ever drawing a manual action. The live discussion area is the part that got flagged.
The AI bot theory is speculation
On 26 July, Australian-based SEO Gagan Ghotra suggested publicly that the forum's AI bot, which auto-replies to most posts, probably triggered the action. That is one practitioner's reading of the available evidence. It is not Google's stated reason. Google's stated reason is thin content.
The detail behind the guess comes from a screenshot discussed in the article: the forum characterises its AI chatbot as a staff member, the chatbot has been active since 2023, and it has posted over a hundred thousand responses to questions. Whether a reviewer reacted to exactly that is unknown, and the speculation should be read as speculation.
What Google counts as thin content
Google has several definitions of thin content and they do not all point the same way. The list includes thin affiliate content, syndicated content and article-marketing content. Former Googler Matt Cutts described thin affiliate content as material lacking "original insight or research or analysis" or other value-adding content such as original video. The opposite, in his framing, is content the owners wrote themselves that is unique and leans on the author's actual expertise.
Google has also said before that AI-generated content is not automatically bad. Both positions can hold at once. The question a reviewer asks is whether a page adds something, not which tool typed it.
Why an AI reply can read as thin on a forum specifically
The reasoning the article offers is about the category rather than this one site, and it is worth keeping separate from the speculation. The expected value-add of a forum is answers grounded in actual human experience: someone who hit the same error, on the same hardware, and can say what fixed it. An AI deals in received knowledge, knowledge that came from somewhere else. On a page whose whole reason to exist is first-hand experience, an answer with none of it sits closer to a summary of other pages than to a new contribution.
That argument does not prove what happened at windowsforum.com. It does explain why a forum is a harsher environment for automated answers than, say, a product specification page.
What this means for marketers in Thailand
Read the story for what it is: one site, one manual action, plus a practitioner's guess. It is not an algorithm update, not a policy change, and not evidence that Google penalizes AI content as a class. Anyone selling it this week as a new rule is running ahead of the facts.
The useful takeaway is narrower and worth acting on now. If you run a forum, a Q&A section, reviews or any other user-generated content, and you have layered an AI auto-responder on top of it, audit that stack before a notice arrives rather than after. Two practical moves:
- Look at what a reviewer would see. Open twenty threads at random. If most of them end in a bot reply with no human follow-up, that is your sample too.
- Label AI replies clearly. Presenting a chatbot as staff takes away the reader's ability to weigh the answer. Labelling costs nothing and is defensible in a reconsideration request.
For Thai sites carrying bilingual user-generated content the exposure doubles, because a thin Thai section and a thin English section are two separate sets of URLs under one domain. Regular SEO audit work on the UGC directories, not only on the money pages, is what surfaces this early. Where AI-assisted content is already part of the workflow, the same discipline belongs in how it is planned and reviewed under AI SEO.
FAQ
Does this mean Google now penalizes AI content?
No. Google cited thin content. The AI link is speculation by SEOs reading the case, and Google has previously said AI-generated content is not automatically bad.
Were the pages removed from Google's index?
The description given is suppression pending a reconsideration request, applied to one URL pattern. The rest of the site was unaffected.
Should we stop using AI on our site?
Nothing in this case supports that as a rule. The question worth asking is whether each page adds something a reader cannot get elsewhere, and whether readers can tell which replies came from a bot.
If your site carries user-generated content with automated replies layered on top, reviewing those directories is cheap insurance. Relevant Audience can look at how that content reads to search engines and to readers. Start with an AI SEO review.
