Let me point out the irony that’s sitting at the center of this week’s news, because I think it’s worth naming out loud.
Google issued a manual action this week penalizing a forum for AI-generated content under the label of “thin content.” At the same time, Sundar Pichai went on the record saying Google needs Gemini 4 to stay competitive at the AI frontier. So Google is cracking down on AI content with one hand while sprinting to produce more powerful AI with the other.
I’m not saying that’s hypocritical — the situations aren’t equivalent, and Google’s position on AI content has always been nuanced (generated content that adds value is fine; low-quality bulk generation isn’t). But the juxtaposition is striking. The industry is deep enough into the AI era now that we’re starting to see the contradictions stack up. This week had several of them.
Here’s what happened.
1. Google Issued a Manual Action for AI-Generated Forum Posts — And It Should Get Your Attention
This was the biggest SEO story of the week, and I don’t think it’s gotten enough airtime yet.
A forum called WindowsForum.com received a manual action for “thin content with little or no added value” covering over half a million posts — specifically targeting the /threads/ section of the site. The part that makes this interesting: the forum’s AI chatbot had been posting automated responses to user questions since 2023, logging over a hundred thousand replies. SEOs digging into it, including Australian practitioner Gagan Ghotra, flagged the chatbot posts as the likely trigger.
What makes this significant beyond the headline is the specificity of the targeting. The forum had other sections that were far more obvious candidates for a thin content finding — a syndicated headline feed, an archived knowledge base — and neither of those got touched. The AI chatbot section did. That’s not accidental.
Here’s my take after 16 years in this: Google’s definition of “value add” has always centered on original expertise and lived experience. That’s why Matt Cutts described the opposite of thin content as content written by the actual author, with unique insight, leaning into real expertise. AI has no experience. It processes patterns. On a forum, where the entire social contract is that you’re getting answers from people who have actually dealt with the problem, AI-generated responses are structurally hollow — even if they sound plausible.
If you’re using AI to auto-generate responses, comments, or user-facing content at scale on any platform, this case should make you pause. Google is paying attention, and they’re getting precise.
Sources: Search Engine Journal | Search Engine Roundtable
2. Another Weekend of Ranking Volatility. Unconfirmed. Again.
Starting late Thursday, July 23rd and running through the weekend, tracking tools lit up with significant ranking movement across verticals. Barry Schwartz flagged it at Search Engine Roundtable — the tools were showing clear volatility, Google gave us nothing officially.
I said it last week and I’ll say it again: unconfirmed ranking updates are the new normal. The days of Google dutifully posting a blog update saying “hey we rolled out a core update, here’s what we know” feel increasingly distant. What we get now is silence, occasional third-party data, and a lot of speculation in SEO communities.
If you saw drops over the weekend, don’t spiral. Give it a full week before concluding anything. And definitely don’t make sweeping changes to your content strategy based on 72 hours of Search Console data. I’ve watched clients make that mistake in real-time — it almost never helps, and often makes things worse by introducing additional variables right when you’re trying to diagnose the problem.
The one thing worth doing right now: document what you saw and when. Dates, affected pages, traffic delta. Whatever eventually gets confirmed about this update, having a clean timeline will help you understand whether you were hit and why.
Source: Search Engine Roundtable
3. Google Got Very Explicit About Fake Reviews — And Issued a Warning With It
Google updated its review snippets structured data guidelines this week with language that feels almost embarrassingly obvious: don’t include fake or undisclosed incentivized reviews in your structured data markup. Their exact words: “Don’t include fake or undisclosed incentivized reviews on your page or in your structured data markup.”
Separately, Search Engine Journal reported that Google expanded its review guidelines more broadly and warned of manual actions.
Here’s why I’m not laughing at the fact that Google felt the need to spell this out: they generally don’t update documentation to cover things that aren’t actually happening at scale. The fact that they added this explicit prohibition, paired with a manual action warning, tells me they’re seeing enough review manipulation in structured data that it’s become worth addressing directly.
If you manage e-commerce sites or local business listings with reviews, take five minutes to audit what’s actually showing up in your review markup. Specifically: are any of those reviews coming from discount programs, freebie offers, or review generation campaigns that weren’t explicitly disclosed? That’s the exposure now. Schema.org review markup is something crawlers read directly, and Google is clearly watching it more closely.
The window for ambiguity here is closing.
Sources: Search Engine Roundtable | Search Engine Journal
4. ChatGPT Ads Is Growing Up Fast — And That Changes the Paid Media Map
OpenAI dropped a stack of ChatGPT Ads Manager updates this week: conversion-optimized bidding, geo-exclusions, bulk campaign management tools, API access, and new product ad formats. The daily budget pacing got updated too.
This is not a side project anymore. This is a paid media platform that is actively iterating at the pace you’d expect from a company that feels the urgency of burning through its revenue model quickly.
What this means for the SEO and content side: the ad-supported layer of AI search is now real and expanding. We’ve spent the last two years asking whether AI search would eat organic traffic. The more interesting question now is whether AI search becomes its own paid media channel — and whether that accelerates the commoditization of organic citations in these systems.
I don’t have a clean answer to that yet. But brands that have been writing off ChatGPT as a traffic source because “it doesn’t drive clicks” need to update their framing. It drives visibility. And visibility now has a price attached, which means organic visibility in AI answers just got a paid competitor for the same real estate.
Search Engine Land also reported that ChatGPT gained access to Yelp reviews, ratings, and photos this week — which expands its usefulness for local intent queries significantly. Local SEOs should note that ChatGPT is quietly becoming a more credible source for “where should I go” questions, and that Yelp data integration is a meaningful content signal feeding those answers.
Sources: Search Engine Land | Search Engine Roundtable
5. “Evergreen Content Is Dead” — And I Mostly Agree With the Uncomfortable Part
Search Engine Journal ran a piece this week with a title that’s going to make some content marketers flinch: “Evergreen Content Is Over — The Individual Is The Only Strategy Left.” The argument is that in an environment where AI can generate infinite technically-correct content on any topic, the only thing that remains scarce is a genuine human perspective.
I’ve been saying a version of this for a couple of years, but I want to be precise about what I actually believe versus the version that makes for a good headline.
Evergreen content isn’t dead. The concept of creating content that remains relevant over time still has value. What’s dying is the commodity version of evergreen content — the “10 best tools for X” articles, the “how to do Y” tutorials that say the same thing as 400 other pages, optimized to the same keywords, with no original insight to distinguish them. AI generates that content faster and cheaper than any human team. If that’s your strategy, you’re competing on a track where you will lose.
What survives is the stuff that requires the author to have actually done the thing they’re writing about. First-hand experience. Documented results. A perspective that comes from being in the room when decisions were made. That’s the kind of content AI can’t replicate, and it’s the kind Google has been trying to surface more reliably (with mixed success) since the helpful content era started.
If you’re evaluating your content strategy right now, the question isn’t “should we make less content?” It’s “what percentage of our content requires the author’s actual brain to exist?” That percentage needs to go up.
Source: Search Engine Journal
6. Google Search Console Now Has a Social Posts Report — Use It
Quick practical note to close things out: Google added a Social Posts report to Search Console. It tracks how your social posts perform in Google Search, which has been surfacing social content in results more aggressively over the last year.
If you’ve been publishing social content on LinkedIn, Facebook, or other platforms that get indexed by Google, this report gives you actual data on clicks, impressions, and CTR for that content directly inside Search Console. That’s a new signal chain worth tracking.
Honestly, the bigger implication here is that Google is treating social content as a legitimate part of your web presence in a way that’s measurable now. If you’ve been siloing your SEO and social teams, here’s a concrete reason to get them talking to each other.
Source: Search Engine Journal
The Week in Summary
The throughline this week is that Google is trying to hold two opposing things at once: a genuine commitment to surfacing human expertise and original value, while also pushing harder into AI-powered everything. The review guidelines update, the AI-generated thin content manual action, the “evergreen content is over” debate — they all touch the same underlying question about what authentic content looks like in an ecosystem increasingly flooded with generated output.
The ranking volatility and ChatGPT’s rapid ad product expansion are separate signals, but they point in the same direction: the ground is still moving, and it’s moving faster than the official announcements acknowledge.
Stay close to your data. Don’t chase headlines. See you next Monday.
Your Friendly SEO is a weekly roundup written from the perspective of someone who has been in this industry for 16+ years. If something here was useful, share it with someone who needs a level-headed take on a loud week.