When it comes to organic search marketing, a loophole is not a strategy. Loopholes have expiration dates, and in many cases, the time spent chasing them would have been better spent on work that doesn’t backfire later on.
Any SEO who has worked in the field for a while has inevitably lived through a seemingly endless loop of fads, hype cycles, and shortcuts aimed at tricking search engines (and now, AI systems). Exact-match domains, article directories, parasite SEO, guest posting at scale, PBNs, programmatic pages, and now the AI-generated listicles, scaled AI content, and “inauthentic mentions” of the GEO era. Most of these things work for a while… until they don’t. And once Google, Bing, and more recently the AI companies themselves crack down, the sites that leaned on these approaches often end up worse off than if they’d never tried them at all, sometimes spending years cleaning up while traffic, visibility, and leads stay suppressed.
Plenty of people in this industry have battle scars from these experiences, myself included. My first real SEO wins came from buying links in 2010-2011, and Penguin erased them in a single morning in 2012. Six years later, the 2018 “Medic” update hit a client of ours that had been using AI and automation to scale medical content (long before this was popular), and it was the hardest hit from an algorithm update that I had experienced firsthand since the Penguin update. I think these scars are the badge of honor that inevitably comes with experience in SEO: thinking you know better than Google, and learning the hard way that you almost always don’t.
All of that said, there is one investment that has held up through every one of those cycles, and it’s the same approach I’ve recommended for many years: investing in real humans and placing them at the center of your marketing strategy. As AI makes generic content nearly free to produce, I believe that real, identifiable experts are becoming the clearest antidote to scaled, machine-generated slop. This article covers why, what the evidence shows so far, and how brands can consider “human-centric marketing” as a bigger part of their strategies.
Google Has Been Saying The Same Thing For 20+ Years
What I find interesting, looking back, is how consistent Google’s guidance has been the entire time, if you read between the lines (or read the guidelines at all). Google’s Webmaster Guidelines from 2003 (yes, 23 years ago) already told site owners to “make pages for users, not for search engines,” and warned against “cookie cutter” approaches with “little or no original content.” They even said that site owners following the spirit of the guidelines would do better than “those who spend their time looking for loopholes they can exploit.”

The list of spam violations Google laid out back then has carried on, in some form, for more than 20 years, right into the AI search era. I put together this infographic showing how Google’s original spam policies map to popular GEO tactics we see today with AI search:

Automation in particular has been in the crosshairs for well over a decade. In August 2010, Google’s John Mueller told a site owner that publishing Google Translate output directly “could be seen as creating auto-generated content, which would be against our Webmaster Guidelines,” and recommended having a human correct it first. By 2012, Google had published a dedicated help page on automatically generated content. Its examples included text “translated by an automated tool without human review or curation,” Markov chains, automated synonymizing, scraped feeds, and “stitching or combining content from different web pages without adding sufficient value.”

For what it’s worth, I still see this exact issue affecting sites today – perhaps more than ever – including companies who come to me after experimenting with scaled AI translations that get little human review. It’s interesting that John’s 15+ year old answer still holds up, although to be fair, most site owners have never read Google’s guidelines (and nobody can be expected to know about a forum reply from 2010). That’s what their SEO experts are (supposedly) there to help them with.
Funny enough, that 2012 list reads a lot like the sales pitch for the same AI content tools that have recently captured the attention of C-level marketers everywhere, many of which work in ways that come very close to (if not directly violate) Google’s long-standing anti-spam policies. It’s not a huge surprise that so many of the sites featured in their case studies have since rapidly lost organic traffic.
Google’s wording has evolved since then, but the thread running through it has stayed remarkably consistent. In February 2023, a few months after ChatGPT launched, Google said it has “many years of experience dealing with automation being used in an attempt to game search results.” Then in March 2024, Google launched the scaled content abuse spam policy, which applies whether content is made “through automation, human efforts, or some combination of human and automated processes.” Automated translation, synonymizing, and stitching are still on the list, 14 years later. Marketers are now using LLMs to experiment heavily with all three. The main difference now is that generative AI sits at the top of the list of examples.
Across more than 20 years of guidelines, the common thread is fairly simple: content made for search engines at scale, with no human reviewing it, adding to it, or standing behind it, eventually becomes a liability.
E-E-A-T Gave Us A Way Out Of The Loop
When Google started using the acronym “E-A-T” in 2014 (now called “E-E-A-T”), it finally gave us a framework that shows how to even do SEO work without getting caught up in the loopholes. Since then, E-E-A-T has been misunderstood, misused, and treated as a checklist or a single ranking factor that works the same way for every site (it doesn’t). In 2021, I even wrote an article aimed specifically at debunking many of these misconceptions.
All of that being said, I still to this day believe E-E-A-T is the best framework we have for approaching our SEO/GEO strategies in ways that won’t be penalized or devalued later, and it gives us plenty of meaningful work to do once we focus on the right things. There’s a reason E-E-A-T is mentioned 120 times in the current version of Google’s Search Quality Rater Guidelines, and why Bing uses similar language, focusing on “quality and credibility,” in its own guidelines.
This type of work is people-first, meaning actual humans with lived knowledge sit at the center of the approach. It’s much harder than loophole-focused SEO, because it needs buy-in, time, and resources from the experts at the company, not just the SEO team operating behind the scenes. But the return is high, and it affects more channels than just organic. It tends to stand the test of time; it’s good for your users and your conversions; it appears to help with AI systems; and it’s unlikely to get you penalized (unless you apply a shortcut mindset to it, which isn’t really an E-E-A-T strategy at all). It’s also the opposite of what folks publicly say they hate about the impact SEO has had on Google’s results, from The Verge asking whether SEO experts ruined the internet to a 2024 study finding product review results increasingly full of SEO spam.
Why I’ve Made Humans The Focal Point Of My SEO/GEO Approach
I’ve been preaching these same ideas for years, and I probably sound like a broken record to anyone who has followed me for a while (my MozCon 2022 talk was all about how to focus on real expertise, with numerous examples of real experts representing the brands they worked for). I even named my holistic approach to SEO/GEO the E-E-A-T Method and made it the core philosophy of my new consultancy, Algorythmic, because that’s how deeply I believe in the efficacy of this approach.
But this wasn’t always a big focus in how I approached SEO; it’s something I learned the hard way.
That 2018 Medic client I referred to before had been scaling a fast-growing library of health articles with no visible credentials from experts, and no sign of a real human behind any of it. When the update hit, Google’s Danny Sullivan pointed site owners back to the Search Quality Rater Guidelines (his tweet has since been deleted, but Marie Haynes wrote about his guidance at the time). So I spent months reading and re-reading Google’s guidelines. Being the nerd that I am, I even printed and laminated a copy and kept it on my desk (which I admitted on stage in 2019 at BrightonSEO). When I read it, the answer to our client’s challenge became obvious, and it was repeated throughout the guidelines: the site was heavily YMYL (Your Money or Your Life) and showed nowhere near enough E-A-T for the health topics it covered and the medical advice it provided.
The client wasn’t convinced at first, as this required a dramatic overhaul of their approach – so much so that we decided to part ways while they tried to bring back the site’s performance using their own methods. But after almost two years without recovering, the client came back to work with our team again. We worked together on a renewed SEO strategy, and about six months later, we relaunched the entire site, rebuilt from top to bottom around E-A-T, with human experts at the center: real doctors, real medical reviewers, and their original insights. The site climbed back to its previous levels after the re-platforming was complete. From the day the update hit, the full recovery process took more than three years. (This long recovery timeline is why I always warn sites to avoid getting hit in the first place!)

Back in January 2019, the night before my first-ever conference talk at SMX West (which was – surprise! – focused on E-A-T), I sat at the hotel bar studying Mark Traphagen’s work on author authority. He, along with Jennifer Slegg, had been documenting the history of “authorship” and “author authority” on Google over time. (Funny side note: I ended up meeting Mark in person for the first time at the conference breakfast the next morning, a few hours before I went on stage, which felt auspicious.) After my presentation, when an attendee asked what a small site should do if its writers aren’t well known, my answer was to “cultivate your authors’ profiles, even if it means starting from scratch,” which ruffled some feathers in a room filled with technical SEOs who didn’t all buy into the idea that Google could figure these things out. Yet nearly eight years of research and client results later, I’m even more convinced that it can.
A few months later, Google’s official guidance on core updates linked to my BrightonSEO deck on E-A-T as a recommended resource, along with work from a few other SEOs. This solidified for me that E-A-T was becoming a big priority for them. In 2020, I had the great privilege of digging into the patents that might explain how E-A-T works with the late, great Bill Slawski, whose knowledge of Google’s patents was unmatched. This exercise helped me understand a more technical explanation for how Google’s systems might recognize individual experts. The following year, I took the topic to the MozCon stage with a talk on how the E-A-T ecosystem has transformed organic search.

What I Mean By “Real Humans”
“Humanize your brand” and “keep a human in the loop” have become marketing clichés of their own, so I want to be specific. By “real humans,” I mean practitioners with genuine expertise or experience: the doctor who has treated the condition, the engineer who built the system, the plumber who owns a local plumbing business, or the patient who lived through the illness. They earn their audience naturally, and across platforms, by sharing insights that are honest, original, often opinionated, and based on real experience. You can usually tell them apart from someone who showed up to sell you something, and if a sale happens at all, it tends to follow from the trust they’ve built.
My approach, for SEO, AI search, and broader marketing, is to put these real people at the center of the strategy whenever possible. It’s also what I’ve employed with my own personal brand (more on that later).
I think this is the strategy most likely to survive whatever comes next, for a simple reason. Nearly every popular shortcut in the vicious cycle of SEO could be scaled. A real expert’s reputation can’t. It takes years of showing up, being right (and sometimes wrong, and open to admitting that) in public, and having other people vouch for you without needing to rely on inauthentic mentions or listicles calling yourself the best expert. As AI makes everything else cheaper to produce, I’d expect the value of that kind of earned reputation to keep climbing.
In April 2024, on the Authority Hacker podcast, I summed it up with one of the sentences I’ve said most often in my career:
“I’ve been talking about this for years: People like following people.”
(Jump to 53:26):
Sidenote: Don’t mind the ghost opening the door behind me. That was my dog Marcy, who can never decide where she wants to lie down when I’m doing podcasts.
What Real People Bring That AI Can’t
It’s no secret that folks on social platforms engage with other people far more than with brands. On LinkedIn, personal profiles outperform company pages by roughly 2x to 8x on impressions, depending on whose data you believe.
Here’s what real people bring to the table that AI can’t:
- Expertise earned over years in the trenches. Someone who has spent 10 or 20 years doing the work has seen what happens after the advice gets followed: the edge cases, the things that broke, or why certain products are better than others. That pattern recognition only comes from lived experience, and it’s often exactly the level of helpful insight that folks are looking for when they go searching for advice. An LLM can summarize what others have written about a job, but it has never had to do the job itself.
- Feelings and lived experience. AI doesn’t have feelings (even though it “feels” to us like it sometimes might). It has never sat with a scared patient, lost a big client, or been nervous before a first conference talk. It can imitate empathy and humor, but it can’t draw on experiences it never had, which is a big part of why AI writing so often feels stale (and sometimes a little alien). Writing from someone who has actually been through the thing reads differently, and it matters to readers that a real person is behind it: across nine experiments with more than 6,000 people, the same supportive message was rated as more empathic and more genuine when folks believed a human wrote it.
- Something new to the conversation. AI tools are getting better at research and analysis, and plenty of folks (myself included) use them to help work through data. But someone still has to decide which question is worth asking, bring data or experience nobody else has, and put their name behind the conclusion. Google’s helpful content guidance asks whether content provides “original information, reporting, research, or analysis,” and in my experience, that still starts with a person (even if that person is using AI in the process).
- Left on its own, AI tends to pull everyone toward the same answer: a 2026 study in PNAS Nexus found that leading AI models score about as well as people on creativity tests, but their answers are far more similar to each other than people’s are.
- Humor and personality. Inside jokes, a client call that became awkward, a baby or dog wandering into a podcast recording (looking at you, Marcy). That stuff seems to be a big part of why folks feel like they know the people they follow. There’s a word for that feeling: Cambridge Dictionary named “parasocial” its 2025 Word of the Year, describing a connection someone feels with a famous person or online personality that they don’t actually know.
- Flaws and vulnerability. Psychologists call part of this the pratfall effect: in a 1966 experiment, a highly competent person became more likable after spilling a cup of coffee. In my experience, my posts where I admit I don’t know something, or share what I learned from my own mistakes, tend to get strong engagement from others who have felt the same way. People seem to engage with that far more than with confident-sounding AI-splaining.
Real Humans Leave A Trail
Across platforms, signs of a real human may be the clearest way to separate spam and automated content from the content that provides the most value.
A real human expert leaves evidence everywhere, and that footprint is easily detectable: photos from the conference they spoke at, video of the talk, a personal website (even a slightly neglected one), podcast interviews, years of social posts and replies, or journalists quoting them by name. I believe Google has systems in place to at least identify these signals, and likely reward them, from Knowledge Panels to the new Search profiles. The leaked Google documentation from May 2024 points in the same direction: it showed that Google stores the author of a document, and has an attribute called isAuthor that marks whether an entity mentioned on a page is also the author of that page.
The typical fake author, on the other hand, has a stock-looking AI-generated headshot, a two-sentence bio (if any), and no other evidence of their real existence anywhere on the web. This approach isn’t limited to small sites, either: in 2023, Sports Illustrated was found publishing product reviews under authors who didn’t appear to exist, including one whose headshot was for sale on a site that sells AI-generated portraits. In that same year, fake authors were one of the most common patterns I found among sites hit by the Helpful Content Update, and in October 2026, Google added this exact tactic to its helpful content documentation: “Fabricating creator profiles (such as by using AI-generated headshots, made-up names, or false credentials to make content appear as if it was written by human experts) is a form of deception.”
I predict that this trail of being a “real human” will matter even more as the web fills up with homogeneous vibe-coded sites. Anyone can now spin up a polished website in an afternoon, complete with a logo, a hundred blog posts, and a team page full of friendly AI-generated headshots. If you’ve looked at many of these sites, you’ve probably noticed they all start to look the same: the same purple gradient, the same three feature cards with emoji icons, the same “Trusted by” logo strip.

When a convincing-looking brand costs next to nothing and takes little time to build, polish alone doesn’t tell anyone very much anymore. What’s left are the signals that still take years to build: a real person or real people with real histories, who other real people have chosen to follow, quote, and recommend.
Search Is Rebuilding Itself Around Real Voices
Since the 2018 Medic update, Google appears to have spent the past 8 years building systems that reward exactly that kind of trail. In 2019, I analyzed 64 sites that had clearly won or lost across the 2018 and 2019 core updates, checking each one for 30 on-page E-A-T signals and researching the credentials of the people writing their content. The winning sites were 258% more likely than the losing sites to use real experts on their content (with the usual caveat that this was a small sample, and correlation isn’t causation). Since then, Google has put many of those same observations into writing:
- December 2022: the E for Experience, making E-A-T into E-E-A-T. When Google added “Experience” to E-A-T, I wrote up the changes for Search Engine Land, including a new section on the reputation of individual creators and a line asking raters to weigh the “extent to which a human being actively worked to create satisfying content.”
- Google had already been moving in this direction. Starting in 2021, its Product Reviews updates rewarded reviews that showed evidence the reviewer had actually used the product, like original photos, videos, or measurements from testing it. That’s something a real person can do relatively easily, and scaled, machine-generated content generally can’t. Garage Gym Reviews is a good example: its equipment is tested by certified personal trainers, coaches, and competitive athletes who actually lift, run, and row on it. Travel + Leisure does something similar, sending products to its testing lab in New York and out on real trips with its travel editors. In both cases, real people with lived experience are doing the testing and the writing. I think these updates were an early sign of how much weight Google planned to put on real human experience.
- 2023: Reddit takes over in the “Hidden Gems” era of Search. In May 2023, Google said it wanted to surface more “hidden gems,” like a forum comment or a post on a small blog. Later that year, it confirmed the changes had been rolling out for months as part of core ranking, built to show more firsthand knowledge and personal experience from forums, social media, and blogs.
- By November 2023, I was posting on LinkedIn that Google seemed to be “way overindexing on ‘authentic content’ and ‘real experience'” in its search results. Social media, UGC, and forums were by far the biggest winning category in my 2023 winners and losers analysis, and my former intern Silvia Gituto’s deep dive on Reddit found it went from the 68th to the 5th most visible domain in Google’s US search results (per Sistrix) in a single year. At one point, a five-year-old Reddit post from a since-deleted account outranked the Mayo Clinic for “how to potty train a boy.” The era of real experience-focused UGC had arrived.

- 2025 and 2026: a renewed focus on individual creators. In September 2025, Google redesigned Google Discover so users can follow individual creators, then launched Search profiles: claimable, verified pages for creators that users can follow directly from Search.
- In October 2026, Damien Andell of 1492.vision flagged a Google job posting about turning Discover into a “proactive, hyper-personalized feed,” and noted that it explicitly calls for keeping human voices and sources at the center. His team has also documented Instagram and TikTok posts being tested in Discover, alongside X and YouTube. With these changes, individual creators get to play a bigger part in Google’s version of a personalized social feed.
AI Search Cites The Places Where People Share What They Know
Real experts and real lived experiences appear to drive a big share of AI search citations, and you can see it in which sites get cited most. YouTube and Reddit are consistently the two most cited websites in Google’s AI Overviews, at 24.5% and 18.1% of citations among the top 50 cited domains in Ahrefs’ November 2026 analysis.
Other studies point to LinkedIn, too. In a Semrush study of 325,000 prompts earlier this year, LinkedIn was the second most cited domain in ChatGPT Search, Google AI Mode, and Perplexity, and 59% of the LinkedIn content cited in ChatGPT Search and AI Mode came from individual members rather than company pages.
This doesn’t surprise me – I’ve seen my LinkedIn posts and Pulse articles show up in Google’s AI search results immediately after publishing, like this test I did with a LinkedIn Pulse article about Google Discover. It earned the top ranking in AI Overviews and an AI Mode citation within seven hours of publishing.


In other words, a lot of the information AI search systems draw on to build responses comes from the places where real humans go to share what they actually know, like a creator on YouTube walking through how they fixed something or which product they like best, a Reddit thread full of firsthand experience, or a practitioner’s thoughtful LinkedIn post or Pulse article. Unfortunately, these same platforms are also being flooded with AI slop, and it’s not a secret.
YouTube And Reddit Are Cracking Down On Inauthentic Content
YouTube and Reddit have been fairly explicit about what they want, and it also points to authentic human experiences. In July 2025, YouTube renamed its “repetitious content” monetization rule to “inauthentic content,” covering AI-generated videos made from generic templates “without adding the creator’s original, authentic insights or perspective,” and CEO Neal Mohan’s 2026 letter called out “AI slop” by name. Reddit CEO Steve Huffman wrote in February that “if you can’t easily distinguish a real person’s thoughts or recommendations from a bot, that trust erodes,” and a month later Reddit started labeling automated accounts. On both platforms, the most reliable path to visibility appears to be the one this whole post is about: be a real, identifiable person – sharing authentic and honest experiences – whom other people choose to trust.
One Person Can Change How A Company Is Perceived
I’ve seen firsthand what sharing publicly can do for a business. When I joined Amsive, I had never given an SEO talk on a stage or done anything public-facing in the industry. Since then, in almost nine years on the agency’s SEO team, I’ve published research on its blog under my own name, alongside podcast interviews, conference talks, webinars, and other marketing efforts on the agency’s behalf. That work has brought the agency links, traffic, and name recognition that keep growing long after each piece goes live. When I recently looked at the agency’s backlinks in Ahrefs, a few things stood out:
- Interviews and collaborations I’ve done with journalists have earned Amsive links and brand mentions on high-authority publications like the New York Times, BBC, Washington Post, CNET, and many more.
- Nine of the 10 most-linked articles on amsive.com are original data studies from the SEO team, including seven I wrote or contributed to.
- About 1 in 13 of the roughly 7,800 sites linking to amsive.com mention my name in or right next to the link, usually as a quote from an article or interview.
- My author page on amsive.com is linked from 120 sites, more than all but seven of the company’s articles.
I’m sharing this because of where those links point: every one of them goes to the company. One person (or a team of people) sharing what they know under their own names can grow a company’s authority in a way that brand content on its own rarely does. I’ve seen the same pattern at countless other brands that put trusted experts at the front of their marketing, and I try to employ this same approach in my client strategies.
Of course, I’m not the only example of this, and others in the industry are seeing success with the same approach. For years, Wix had what the Ahrefs blog called “a bit of a bad reputation for SEO.” Plenty of SEOs would have steered clients away from it. Then Wix did something smart: it hired respected SEOs like Crystal Carter and Mordy Oberstein to represent the brand within the SEO community.
Crystal joined in 2022, and since then she has spoken at many major search and marketing conferences, co-hosted Wix’s SERP’s Up podcast, and published her own research on the Wix site. Today, when SEO publications write about whether Wix is good for SEO, they often point to her by name. That same Ahrefs article credits her and Mordy with pushing for “more SEO features and better communication.” About 1 in 20 sites linking to Wix’s SEO Hub mention Crystal by name, the podcast shows up in “best SEO podcasts” roundups from Backlinko and Ahrefs, and her expert page is the most-linked expert profile on the hub.
For a more famous example outside our industry, look at travel expert Rick Steves. He self-published his first travel guide in 1980, and he has spent about four months a year in Europe for decades, walking the routes himself and updating his guidebooks every year. That one person’s lived experience grew into a company with more than 100 employees, a PBS series that has run since 2000, a thriving forum and online community, a public radio show, and a tour program that takes more than 30,000 travelers to Europe each year.
I wrote about his site on LinkedIn this summer, calling it the ultimate embodiment of E-E-A-T and the perfect example of an expert-led marketing approach. He tells you – from firsthand experience – what’s overrated, overpriced, or skippable, and because he makes money from books and tours, his content is free from affiliate influence. His honesty is the product. Google’s AI Overviews often cite his website, and his YouTube and TikTok pages too.
How To Build A Human-First Strategy With Brand Evangelists
Brands that can should start by identifying the real experts inside the company and investing in making them visible. In practice:
- Find the boots on the ground – the people who actually know things. Your product lead, your head of support, the engineer who answers every hard question in Slack, or the founder who still takes customer calls.
- Build up their presence on your site. Focus on building robust About pages, author and expert pages, and any other opportunity to showcase the real people and real experts at the company. Use their author profiles to highlight the great work they’ve done, connect to their social properties, and treat the page as a personal portfolio for the expert. Give them a voice, and a way of contributing to company communications, whether it’s contributing to the blog or building a platform on LinkedIn to support their efforts at the company.
- Encourage them to build an expert footprint outside your site. Experts can give conference talks, do podcast interviews, or help with posts on Reddit (or even help manage the official branded subreddit), LinkedIn, or other online communities where your buyers spend time. Help them earn and claim their Knowledge Panel and, when they qualify, their Search profile. Think of it as a way for them to build their personal brand, while simultaneously growing visibility for the company.
- Keep the pitch out of it. Folks follow people who are genuinely useful to them, and effective marketing is increasingly becoming authentic, not salesy. Share the data, the mistakes, the “I don’t know yet,” and credit other people’s work generously.
- Make your people easy to verify. In October, I shared that the most cited page on an enterprise client’s site in AI Overviews and AI Mode was its About page: a detailed, human-readable history of the company. I still believe About pages are typically underinvested in.

Two fair counterpoints, since this approach isn’t possible for every company: Some industries have compliance teams that make personal posting genuinely hard. And people may leave the company and take their personal brands with them. My answer is to try to build several voices instead of relying on one (my friend and colleague Viola Eva recently wrote about how they do exactly this at her company, Flow Agency). Furthermore, remember that any links, mentions, and coverage earned while someone represented the company can remain as valuable assets for the company, even after they move on.
Why This Stands The Test Of Time
In July 2023, I posted on LinkedIn that even if Google couldn’t tell AI content from human content, someone else could: “guess who else doesn’t like reading AI content? USERS.” The consumer research has since caught up. A Gartner survey this summer found that a majority of U.S. consumers say AI content has made them less trusting of brand messaging, and Emplifi’s 2026 research found that only 35% of Americans trust AI-generated content at all (not that everyone can necessarily tell the difference). I’d call it slop fatigue, and I think the platforms are finally responding to it (or at least trying to).
Every tactic that has blown up in this industry, from the links I bought in 2011 to the listicles and paid mentions all over the internet in 2026, was scaled for the sole purpose of driving visibility in search. That’s also, almost exactly, the list of things consumers say they dislike most about SEO. Real experts and experienced contributors have the opposite effect: they contribute the things that make the internet human, and the most popular, widely cited platforms are working hard to preserve their humanness in an internet flooded with AI slop.
There’s another reason this holds up: your reputation is ultimately decided by what other people say about you, not by what you say about yourself. One of the rater guideline passages I put on a slide back in 2019 reads: “When there is disagreement between what the website says about itself and what reputable independent sources say about the website, we’ll trust the independent sources.”
AI search is carrying that torch forward, as LLMs are getting better and better at determining someone’s reputation based on what others say about the brand or the person. That appears to increasingly include spotting self-promotion. Gemini now shows its reasoning steps, and in this example (h/t Gagan Ghotra), it directly evaluated whether “Best AI SEO agency” lists were self-promotional, with agencies ranking themselves highly, to judge how objective those rankings were. Being genuinely recommended by other people is the most sustainable way forward.

People like following people. That was true before Google, and it’ll still be true after whatever comes next. The brands that figure that out won’t need the next loophole.
More Resources:
- The New Organic Growth Stack: Search, Social, Communities & AI
- What Search Engines Trust Now: Authority, Freshness & First-Party Signals
- LinkedIn Shares What Works For AI Search Visibility
This post was originally published on Lily Ray NYC Substack.
Featured Image: Spiximcreation/Shutterstock