Judge Raises Critical Doubts About Google AI and Spam Update—What This Means for Your SEO Future

Ever wondered if Google’s been playing the role of both referee and player in the AI content game? This week’s SEO Pulse throws us right into the thick of it—with a savvy federal judge poking holes in Google’s AI content deals, making us question who really holds the cards when it comes to using web content for AI overviews. Meanwhile, Google’s latest spam update has folks buzzing about the fate of scaled AI publishing—is it the mass-produced content that’s drawing heat, or just the way it’s being churned out? Oh, and just when you thought the GEO debate had settled, Google’s own John Mueller jumps back in, saying AI search doesn’t require special GEO tweaks. Not to mention, there’s some juicy data showing how AI tools can recommend a brand but point the citation finger elsewhere—and Reddit’s tangled dance with ChatGPT citations adds another layer to this ever-evolving saga. So, how do visibility and value truly stack up in this new AI-driven search landscape? Buckle up; it’s a wild ride with plenty of surprises ahead. LEARN MORE

Welcome to the week’s Pulse: a judge questioned Google’s AI content arrangement, the fallout from the spam update has brought scaled AI publishing into question., and the GEO debate picked up again.

AI citation data also revealed how much recommendations can differ from links, while new ChatGPT testing added some complexity to last week’s Reddit story.

Here’s what matters for you and your work.

Judge Questions Google’s AI Overview Content Deal

A federal judge questioned whether Google’s search power gives it an advantage when using web content for AI Overviews.

Key facts: During Tuesday’s hearing in Penske Media’s antitrust case, Judge Amit Mehta questioned Google’s labeling of AI Overviews as a “product improvement.” Jason Kint, CEO of Digital Content Next, noted that Mehta described the situation as “really unfair,” adding that the improvement was being developed “on the backs of the publishers.”

Mehta hasn’t ruled on Google’s motion to dismiss.

Why This Matters

This case challenges the traffic-for-content arrangement, placing it directly into the middle of an antitrust dispute. Websites feel they can’t comfortably restrict Google Search from accessing their content, especially since Google relies on that content for AI responses.

What Publishing Professionals Are Saying

Kint, who attended the hearing, wrote:

“Mehta noted publishers lack control over how Google uses their content.”

He later argued through Digital Content Next that opting out of Google isn’t a realistic choice for publishers that depend on search traffic.

Read our full coverage: Judge Says Google/Publisher AI Situation ‘Seems Really Unfair’

Spam Update Reports Put Scaled AI Content Under Scrutiny

Google hasn’t said that its August spam update targeted AI-generated content, but reports from affected site owners have put automated publishing in the spotlight.

Key facts: Roger Montti collected examples of fully automated sites losing visibility while some sites using AI with human review held up better. These are anecdotal reports, not confirmation of what Google changed.

Google researchers recently released S-CTS, a system designed to detect coordinated synthetic abuse on video platforms. The paper does not relate to Google Search and does not suggest that the system was involved in any update.

Why This Matters

AI authorship by itself does not account for the ranking loss. Google’s spam policy addresses large-scale content designed mainly to manipulate rankings, whether it’s created by a person or AI.

What SEO Professionals Are Saying

SEO consultant Takuma Oka cautioned that his sample was small and that he wasn’t presenting the pattern as definitive.

Other practitioners focused less on AI itself and more on the method and purpose behind mass production.

Read our full coverage: Reports Indicate Google’s Spam Update Focused On SEO AI Content

Google Says Its AI Search Doesn’t Need Special GEO Work

Google’s John Mueller added another entry to the SEO-versus-GEO debate this week.

Key facts: Asked whether there are industries where GEO doesn’t matter yet, Mueller wrote on Bluesky:

“From our POV there’s nothing really special you need to do for generative AI responses in search.”

The important qualifier is “from our POV.” Mueller was describing Google Search, not ChatGPT, Perplexity, or every other AI system.

Why This Matters

If you’re already practicing SEO for Google, Mueller isn’t suggesting another technical checklist for AI Overviews or AI Mode. However, keep in mind that not all AI engines retrieve or choose sources in the same way.

What SEO Professionals Are Saying

Dan Akeju summed up the distinction on LinkedIn:

“Same fundamentals. Different retrieval layer.”

That may be where much of the debate eventually lands. The basics are similar, but measurement and retrieval might vary depending on the platform.

Read our full coverage: Google Answers If Some Sites Can Ignore GEO And Just Focus On SEO

AI Tools Recommend Brands But Cite Other Sites

New ecommerce data shows that getting recommended in an AI answer doesn’t mean your website gets the link.

Key facts: Shero Commerce collected 1,851 citations across Google AI Mode, ChatGPT, and Perplexity. Brand-owned pages received 2.8% of citations. When a brand was recommended by name, its own page received the citation 31% of the time.

Third-party sites frequently received the citation instead.

Why This Matters

Brand visibility and referral opportunities can sometimes operate independently. If your reporting only tracks mentions or recommendations, you might miss out on seeing who actually receives the citation and the potential clicks that come with it.

What SEO Professionals Are Saying

Aleyda Solís highlighted that split on LinkedIn:

“Your brand can get recommended in the answer… while the citation (and the click) goes to a magazine or a marketplace.”

The analysis looked at where citations appeared. It didn’t explore why AI systems selected those sources or if changing brand copy would influence the results.

Read our full coverage: AI Tools Recommend Brands But Cite Other Sites, Data Shows

ChatGPT’s Reddit Behavior Gets More Complicated

Promptwatch data showed Reddit’s ChatGPT Search citation share falling sharply in mid-August. New testing shows Reddit is still entering ChatGPT’s retrieval process and can still receive most of the final citations.

Key facts: In one query, ChatGPT specifically searched r/whatnotapp. Of 71 pages retrieved, 48 were Reddit threads, and six of eight final citations went to the subreddit.

Four days earlier, Suganthan Mohanadasan found ChatGPT retrieving 84 Reddit threads for another query and citing none of them.

Why This Matters

Citation share doesn’t show everything an AI system considered before generating an answer.

The two tests also show why it’s risky to describe ChatGPT as broadly favoring or avoiding a domain based on one set of queries.

What AI Search Professionals Are Saying

Mohanadasan summed up his follow-up testing on LinkedIn:

“Reddit citations haven’t simply been ‘switched off.’ They appear to be query dependent.”

Read the full breakdown: ChatGPT Searches Individual Subreddits By Name

Theme Of The Week: Visibility And Value Are Separating

This week’s stories highlight how presence and payoff can sometimes feel disconnected. Publishers can share valuable information that supports an AI’s answers without necessarily getting a visit themselves. Brands can earn recommendations even if another site is cited. Reddit, for instance, can contribute significantly to ChatGPT’s retrieval pool without showing up directly in the final sources.

Even the GEO debate touches on this same theme. Being crawled or retrieved is just one part of the process. Truly making an impact involves being mentioned, cited, clicked, and ultimately converted.

“AI visibility” is starting to cover too many of them at once.

More resources:

Featured Image: Cast Of Thousands/Shutterstock