Google’s August 2026 spam update did not penalize AI-generated content. It penalized low-quality content produced at scale, regardless of whether a human or an AI wrote it.
At TJ Digital, we create content for about 60 websites, and 100% of it is AI-generated. I just finished comparing our clients’ data from the 10 days after the spam update to the 10 days before. There was no correlation between AI-generated content and ranking losses.
Google has been clear about this. Their spam policy states that appropriate use of AI or automation is not against their guidelines. Using AI doesn’t give content any special gains or penalties.
The update rolled out from August 18 to August 21, 2026. It took about two days and 16 hours, applied globally, and Google announced it as a standard spam update. No new AI-specific policy accompanied it.
What Types of Sites Were Hit by This Spam Update?
Glenn Gabe published detailed case studies of four sites heavily impacted by the August update. Every one of them combined extreme scale with little original value.
These four sites illustrate what Google was actually targeting.
| Site Type | What They Were Doing | Result |
| YMYL site | Massive programmatic expansion with AI text inserted into template pages | Lost rankings for 200,000+ queries |
| Amazon affiliate | Scraped product data from Amazon, added generic AI text, linked out with affiliate links | Lost rankings for 14,000+ queries |
| Large programmatic site | 1.5M+ indexed URLs, ~85% programmatic, content available elsewhere | Declines across multiple sections |
| Aggressive scaled site | 250,000+ URLs, misleading functionality, redirected users to riskier sites | Dropped for nearly 25,000 queries |
In every case, the problem was the same. These sites were scraping data from other sources and creating pages that added nothing new. The AI just made it cheaper to do at scale.
Google’s spam policy defines this as “scaled content abuse,” meaning large amounts of unoriginal content that provides little to no value to users, no matter how it’s created. That last part is important. The policy doesn’t care whether a human or a machine produced it.
Does Google Have a Problem With 100% AI-Generated Content?
No. Google’s official position focuses entirely on purpose and quality. There’s no rule that says “more than 50% AI is bad” or “every article requires human-written paragraphs.”
Ahrefs ran a study analyzing about 150,000 pages from the top 10 positions across 100,000 search results in June 2026. Among pages ranking in positions 1 through 3, 5.3% were classified as 100% AI-generated. Another 9% were at least 80% AI.
Fully AI-generated pages showed up at every position in the top 10. At position 1, 8.4% of pages were classified as at least 80% AI content.
Ahrefs did find that higher AI content correlated with somewhat lower rankings on average. But they were careful to point out that this correlation likely reflects quality differences, not an AI detection penalty. Sites that rely heavily on AI tend to be younger, weaker, and more aggressively scaled, and those are the actual ranking signals at play.
The real risk with fully automated content is that without strong inputs and editorial controls, it becomes very easy to publish generic summaries that add nothing new. Those are exactly the pages Google is trying to filter out.
What Did We See Across Our Clients After the Update?
Most of our clients saw at least a 10% shift in organic traffic from Google after the update. Some shifted as much as 60%. But there was no correlation between the AI-generated content we’ve created and the pages that moved.
In fact, our pages tended to fare better since the update. I don’t think that’s because they’re AI-generated. I think it’s because they’re high-quality pages targeting the right keywords for each business.
I looked for patterns across different content types. Some listicle pages went up, others went down. Some location pages went up, others went down.
The only consistent pattern I found across our clients was about keyword specificity.
Pages targeting broad, competitive keywords tended to drop. Pages targeting keywords closely matched to what our clients actually specialize in tended to rise.
Why Do Broad Keywords Drop While Niche Keywords Rise After Google Updates?
This is a trend we’ve seen across essentially every Google update for the past five to ten years. One of the easiest levers Google has to decrease spam in search results is to increase the weight they put on authority. As a result, bigger, more authoritative brands continue to win on competitive terms, and it gets harder for smaller sites.
This doesn’t mean Google penalized smaller sites. Broad head terms like “CRM software” attract many credible candidate pages and cover multiple search intents. A specialist has to outperform established competitors on relevance, utility, links, and reputation all at once.
More specific queries reduce the candidate set. A search like “CRM for independent UK mortgage brokers” gives a specialist site a real opportunity to prove exact topical relevance and unique expertise. There are fewer pages competing, and the searcher’s intent is much clearer.
Google’s own 2026 optimization guide actually warns against responding to this by creating a separate page for every possible keyword variation. Doing that primarily to capture search traffic can violate the scaled content policy. Google’s semantic systems can understand relevance even when a page doesn’t exactly match every query phrase.
Can Fully AI-Generated Content Rank Long-Term?
The answer depends entirely on how you use AI.
SE Ranking ran an experiment that illustrates this perfectly. They published six AI-assisted articles on an established site, and humans revised the drafts for clarity and accuracy.
Between June 2024 and July 2025, those six articles produced about 555,000 impressions and over 2,300 clicks. Three ranked in Google’s top 10.
Then they ran a second experiment. They created 20 new domains and published 2,000 AI-generated articles across them, 100 per site, with no manual edits, no internal linking, and no images.
Google indexed 71% of them within the first 36 days and early results looked promising. But around the three-month mark, the share of pages ranking in Google’s top 100 fell from 28% down to 3%.
That contrast is the clearest practical illustration of what works and what doesn’t.
AI content can rank when it’s part of a quality process. The speed AI provides should be reinvested into research, fact-checking, unique data, and editorial judgment.
Semrush’s 2026 survey of 224 SEO professionals found that 70% of teams identified speed as AI’s top benefit, but only 19% said it improved content quality. That gap is where the risk lives.
What’s the Best SEO Strategy After Google’s Spam Updates?
The best move is to make each page on your site harder to replace.
Google’s July 2026 guidance says that unique, compelling, and useful content is likely to influence long-term search visibility more than any other single factor. They specifically warn that simply increasing the number of pages does not make a site more relevant or higher quality.
For small and medium-sized businesses, that means a few things.
Focus on what you’re actually best at. Don’t try to compete with companies that are 10 or 100 times your size on broad industry terms. Focus on long-tail, bottom-of-funnel keywords that closely match what your ideal customer would be searching for.
Those terms have clearer intent, less competition, and higher conversion rates.
Build pages with an information advantage. Google’s quality criteria repeatedly ask whether your content provides original research, first-hand experience, or information that can’t be found elsewhere. Proprietary data, real case studies, actual product tests, original photography, and expert conclusions are all things a generic AI summary can’t reproduce.
Ask yourself this. Could an ordinary language model generate this same page without access to my business, my expertise, or my evidence? If the answer is yes, the page is vulnerable.
Use AI to increase editorial quality, not page count. The strongest evidence favors using AI as an input into a quality process. Let AI handle research, organization, and drafting, then invest the time you save into subject-matter expertise, fact-checking, unique data, and editorial judgment.
At TJ Digital, we’ve built systems that do exactly this. Our content starts with each client’s actual words and expertise through their Brand Ambassador, then goes through multiple layers of review. The human inputs are what make the content worth ranking.
Audit your indexed pages. If you have hundreds or thousands of thin pages sitting in Google’s index just because they were cheap to produce, they could be dragging down your site. Would you still want this page available to customers if Google sent it zero traffic? Pages whose only justification is an exact-match keyword deserve a hard look.
How Can You Use AI Content Without Getting Penalized?
We use AI in every workflow. We’re transparent about it. And we’ve never been penalized for it.
We don’t use AI to create commodity content at scale. We use it to deliver more high-quality work within each client’s budget. Every piece of content starts with the client’s actual expertise, goes through our Brand Ambassador system, and gets reviewed by our team before it goes live.
Google’s spam updates go after automation used to flood the web with pages that exist only to capture search traffic. The only thing that determines whether your content is at risk is whether the page adds genuine value.
Talk to TJ Digital about building content that ranks in both traditional and AI search. Request a free digital marketing audit and we’ll show you where the opportunities are.