{"id":619951,"date":"2026-08-04T13:17:21","date_gmt":"2026-08-04T13:17:21","guid":{"rendered":"https:\/\/www.europesays.com\/ie\/619951\/"},"modified":"2026-08-04T13:17:21","modified_gmt":"2026-08-04T13:17:21","slug":"can-reddit-fend-off-a-new-wave-of-ai-seo-spam","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/ie\/619951\/","title":{"rendered":"Can Reddit fend off a new wave of AI SEO spam?"},"content":{"rendered":"

Earlier this year, a Reddit user had asked<\/a> members of a skincare-focused subreddit if anyone had tried a specific hypochlorous acid spray, a product often used for acne. There were dozens of responses; one from a user named Primary-Taro4254 seemed innocuous enough, at least at first.<\/p>\n

\u201cI haven\u2019t personally tried [that brand] so I can\u2019t fully compare, sorry,\u201d they commented. \u201cBut I have been using a similar hypochlorous acid spray from Honeydew Labs and I\u2019ve honestly really liked it.\u201d<\/p>\n

Primary-Taro4254 helpfully shared their experience using the spray to calm their skin\u2019s redness and irritation, and even threw in some of Honeydew Labs\u2019 credentials: \u201cOne thing I noticed when comparing products is that Honeydew Labs uses a 0.02% hypochlorous acid concentration and it\u2019s also accepted by the National Eczema Association, which made me feel a bit more confident trying it on sensitive skin.\u201d<\/p>\n

But a closer look at Primary-Taro4254\u2019s Reddit profile revealed that they apparently really, really liked the Honeydew Labs hypochlorous acid spray. It wasn\u2019t just the one thread where they were praising the product \u2014 they had posted similar messages on threads about unrelated products as well as in other skincare-related subreddits. Was Primary-Taro4254 working for Honeydew Labs? Was it a marketing firm\u2019s inartful promotional strategy? Was there a person behind it at all, or was it an automated spammer? Before long, a member of r\/SkincareAddiction \u2014 a popular subreddit with more than 1 million weekly visitors \u2014 had caught on<\/a>, telling others to \u201cbe aware of honeydew labs hypochlorous acid\/ marketing bots.\u201d Soon after, a moderator announced they had taken action on Primary-Taro4254 and would now automatically filter out Honeydew Labs posts for moderator review.<\/p>\n

(Honeydew Labs didn\u2019t respond to a request for comment. Primary-Taro4254 didn\u2019t respond to a message sent via Reddit.)<\/p>\n

It\u2019s hard to overstate how central Reddit has become to how billions of people experience the web over the last several years. It has gone from a constellation of niche communities with a sometimes cantankerous user base to becoming perhaps the most urgently important website on the internet. First, users began adding \u201creddit\u201d to the end of queries in order to wade through results littered with search engine optimization (SEO) junk, and in turn Google began linking to the site prominently<\/a>. Now, Reddit has become significant to brands and marketers for another reason: AI chatbots love to cite it.<\/p>\n

According to data compiled for The Verge by Semrush, an SEO company owned by Adobe, Reddit was the most-cited domain in May by AI tools ChatGPT, Perplexity, and Google\u2019s Gemini and AI Mode \u2014 more than any news publisher, repositories of scholarly articles, and even Wikipedia. AI responses are also encroaching into the online shopping journey: Semrush also found<\/a> that compared to a year ago, Google is inserting AI Overviews much more frequently into search queries with commercial intent, like when shoppers are doing research on products or comparing brands.<\/p>\n

When web traffic is funneled through LLMs instead of a traditional search results page, as Google has been doing<\/a>, being cited by a chatbot becomes ever more important. Now, a whole host of startups, brands, and agencies are coming to Reddit to try to promote their companies in hopes that they will get picked up by AI<\/a>. In an effort to try to manipulate LLM responses and stuff AI search results with brand-friendly answers, marketers have descended upon Reddit, a pseudo-anonymous platform that users have come to associate with authenticity and unfiltered, truthful opinions.<\/p>\n

They will have to get past the moderators first.<\/p>\n

Before Maya Adivi became one of r\/SkincareAddiction\u2019s moderators, she was an active user, immediately drawn to the community\u2019s attention to the science underpinning the beauty industry: Instead of general, simple questions like \u201cWhat should I do for acne?\u201d, users took a more rigorous approach, discussing and explaining new studies being published simply because they were passionate about the topic. Moderating for spam and promotional content has always been part of the job, Adivi says, but these days even a long, detailed breakdown of a study might be viewed with suspicion.<\/p>\n

Years ago, moderators would look for certain telltale signs of spam or covert promotional content: a review that was a bit too glowing, product links with tracking tags, or before and after pictures that were the same as those included in product listings for the brand. AI search has changed how marketers attempt to game systems such that some SEO experts say that brands don\u2019t even need a backlink<\/a> anymore to effectively get their websites cited by a chatbot \u2014 merely a mention is enough. A suspected spammer might pose a question like \u201cHow does everyone keep track of their skincare routine?\u201d that at first glance appears to be authentic engagement, but then a wave of comments will mention AI skincare apps, for example. Other posts might ask, \u201cI keep seeing ads for this brand. What\u2019s your experience with it?\u201d Dubious spam responses might then follow a slightly self-deprecating format: \u201cI thought this product would be too aggressive, but it was actually very gentle. Slow and steady wins the race!\u201d Posts are often open-ended questions, mimicking how a user might prompt an AI chatbot.<\/p>\n

\u201cIt\u2019s a lot more subtle, but we\u2019ve become more sophisticated as well,\u201d Adivi says.<\/p>\n

A brand secretly trying to promote itself becomes \u2018persona non grata\u2019<\/p>\n

r\/SkincareAddiction prohibits accounts affiliated with brands from posting in the subreddit \u2014 Adivi and other moderators will sometimes give a warning to accounts that violate the rule, but more often they will automatically ban brands and sock puppet accounts violating the rule. A brand secretly trying to promote itself becomes \u201cpersona non grata,\u201d Adivi says. She was recently alerted to a marketing agency that claimed it had targeted and \u201chijacked\u201d r\/SkincareAddiction to promote an unknown Korean beauty brand and to rank on Google; Adivi removed the Reddit thread the agency used and added the brand to the keyword filtering list. Moderators are also strict about who gets to do sanctioned Ask Me Anything-style threads: It has to benefit the community, not serve as a platform to shill for your company. In the old SEO days, everyone knew that backlinks were important \u2014 in the age of AEO (\u201canswer engine optimization\u201d) or GEO (\u201cgenerative engine optimization\u201d), marketers are throwing spaghetti at the wall to see what sticks. So they keep coming back to Reddit.<\/p>\n

\u201cWith AEO right now, no one knows exactly how to do it, and everyone tells you how important Reddit is,\u201d Adivi, who has a marketing background, says.<\/p>\n

Google has its own rules<\/a> around spam and manipulating its systems. In an email, Google spokesperson Jennifer Kutz acknowledged there is high- and low-quality content on forums like Reddit, just as in other formats; Kutz says Google\u2019s systems are designed to detect when user-generated content is helpful or low quality. Kutz also said that Google\u2019s AI search features don\u2019t aim to display content from any specific site or platform, and that Reddit \u201cgets no special preference\u201d in Google\u2019s ranking system.<\/p>\n

\u201cThere will always be bad actors trying to game the system \u2014 in fact, spamming predates the internet \u2014 and fighting spam is a core expertise for us,\u201d Kutz told The Verge in an email. \u201cWe have strong protections against manipulation across Search, including our AI features, and we\u2019ve kept results 99% spam-free for years.\u201d<\/p>\n

Reddit becoming the main character of search is amusing in some ways \u2014 only in 2024 did the company introduce native tools<\/a> to help businesses grow their presence on the platform. Google has long used a website\u2019s \u201cauthority\u201d in its soup of search result ranking signals; Reddit, with its anonymous users and fiefdom moderation ecosystem, turns the idea of trustworthiness on its head. Reddit\u2019s user-generated content \u2014 mountains of which LLMs have been trained on<\/a> \u2014 is trusted precisely because it seems to come from nowhere in particular.<\/p>\n

As a result of Reddit\u2019s growing influence, the website\u2019s legions of unpaid volunteer moderators have suddenly found themselves in a unique position: The communities they oversee and nurture are now potentially very valuable, not just to the people who love skincare and indie music but also to the brands and companies that profit from those fandoms. As advertisers and the SEO industry tweak their tactics for showing up in AI search<\/a>, new segments of the internet must begin to reconcile being hyper-visible online and the weirdness that attention elicits.<\/p>\n

Complaints about potential LLM-related spam and promotional content pop up in a diverse array of subreddits: A user on r\/CleaningTips questioned<\/a> whether the subreddit needs a ban on product names; another user warned<\/a> the popular r\/BuyItForLife community that it was being \u201cbrigaded by ad bots to skew Google and AI search results.\u201d In its rules against self-promotion, r\/FoodNYC specifically calls out brands targeting and commenting on old posts. A moderator of r\/salestechniques posted<\/a> a \u201cname and shame\u201d thread filled with brands they said were astroturfing the subreddit. Reddit itself has acknowledged the new kinds of spam. In early July, the company announced<\/a> it was beefing up its spam detection tools with AI and LLMs to better catch \u201cthe highly subtle, coordinated patterns of fake behavior and artificial hype.\u201d The company said it is catching 25,000 spammy posts and comments and blocking 23 million spam views a day as a result.<\/p>\n

\u201cIn Reddit\u2019s 21 years, one constant has been that spam tactics continually evolve as new technologies emerge,\u201d Jared Nelson, senior director of global corporate reputation communications at the company, told The Verge in an email. \u201cEvery time bad actors change their playbook, we strengthen ours.\u201d<\/p>\n

Reddit says that roughly 40 percent of discussions on the platform are commercial in nature<\/p>\n

Nelson declined to name specific signals its systems look for, saying Reddit didn\u2019t want to give bad actors a playbook for evading detection. Nelson said the company evaluates accounts when they\u2019re created, looks for coordinated manipulation across accounts and conversations, and uses platform activity and user reports to improve detection.<\/p>\n

Some brands have taken to spinning up their own subreddits where the company itself moderates the community in an attempt to boost a brand\u2019s presence on the platform. Reddit and Wikipedia \u2014 places where brands can\u2019t simply dump their press releases \u2014 have become top of mind, says Mike Moschella, director of analytics at DKC, a marketing and PR firm.<\/p>\n

\u201cReddit is tricky, right? You can\u2019t bullshit Reddit,\u201d Moschella says. In a world where consumers are looking for answers to their highly specific, personalized questions, brands should be able to make their pitch more clearly to consumers instead of repeating the same lofty, imprecise promises of offering the best product in their category, he says. (Shopping is a huge part of Reddit: Nelson told The Verge that roughly 40 percent of discussions on the platform are commercial in nature.)<\/p>\n

\u201cThe starting point of all business analytics is this idea that you can have a better price, you have better service, or you can have better quality, but you can\u2019t have all three,\u201d Moschella says. \u201cIf you say [your company\u2019s stock] is the best stock ever in the universe, you can issue that press release a million times, but WallStreetBets [the subreddit] isn\u2019t going to buy it if the fundamentals don\u2019t agree.\u201d Reddit should force brands to be more \u201chonest and authentic,\u201d Moschella says \u2014 at least in an ideal world.<\/p>\n

Whether that\u2019s actually happening is another question. It\u2019s not just skincare subreddits feeling the effects of the brave new world of search. 404 Media reported<\/a> in June that biohacking subreddits were dealing with an onslaught of promotional spam about peptides<\/a>. On the weightloss subreddit r\/loseit, Dan, a moderator, has gone all in on trying to spot and clean up advertiser content disguised as authentic activity. (Dan asked The Verge to use his first name only to protect his privacy as a moderator of a large subreddit.)<\/p>\n

On r\/loseit, Dan has come across covert advertiser material for everything from food scanning apps and body scanning scales to GLP-1 providers. (Moderators have compiled a \u201cshit list of brands\u201d that they believe are trying to secretly advertise, Dan says. Those brand mentions are automatically filtered from the subreddit.) A normal person isn\u2019t precious about how they talk about a new smartwatch they got \u2014 they might not even mention the exact brand, Dan says. Brands, on the other hand, are cramming in keywords, using exact names, and discussing products in a positive light.<\/p>\n

\u201cThe main advantage that we have as people who are trying to stop [brands] from advertising is that they\u2019ve got an agenda that a normal poster doesn\u2019t have, and that is usually fundamentally what shines through and that\u2019s how we spot them,\u201d he says.<\/p>\n

Dan has also seen more and more posts he believes are AI-generated, based on certain tells: The final paragraph of a post beginning with \u201cI\u2019m curious to know,\u201d for example. Between auto-moderation tools and community members proactively reporting suspected ads, r\/loseit moderators are able to filter out much of the marketers attempting to promote their products.<\/p>\n

\u201cWe want it to be human to human because that\u2019s why I liked that subreddit anyway,\u201d Dan says. \u201cI found it very good for people to speak to one another outside of somewhere that\u2019s constantly advertising to you.\u201d<\/p>\n

Self-promotional spam on Reddit had been happening for years, of course, to varying degrees of success. But this era of AI search manipulation comes at a time when audiences are looking for a reason to be suspicious of what they see online.<\/p>\n

And who can blame them? Practices that have become commonplace in marketing have slowly trickled into public consciousness: paid \u201cclipping\u201d campaigns<\/a> meant to maximize reach on short-form video feeds; bot farms spun up for PR damage control; brands pulling marketing stunts<\/a> that are later revealed as ad campaigns. \u201cThe feed is fake,\u201d Vulture declared<\/a> of Instagram, TikTok, and YouTube. Reddit was supposed to be different. Adding \u201creddit\u201d to the end of search terms was a way for users to signal they wanted an answer unburdened by the SEO slop that had come to fill the web<\/a>; so trusted was Reddit content that it has become shorthand for \u201cactually useful.\u201d<\/p>\n

This spring, reports<\/a> circulated<\/a> that a marketing agency was behind some of the social media frenzy surrounding the indie rock band Geese. The behind-the-scenes involvement of the agency prompted some feelings of betrayal from the band\u2019s fans, and the dustup became its own news cycle in music and entertainment communities \u2014 was Geese a \u201cpsyop\u201d? Was all the fandom actually just fake engagement? Several years ago, moderators of the popular subreddit r\/indieheads<\/a> had caught a major record label spamming and promoting its own artists in threads, says Matty Monroe, one of the moderators. The commotion around Geese prompted Monroe and others to take another look. Moderators uncovered four accounts within a week that they believe secretly belonged<\/a> to indie record labels.<\/p>\n

\u201cAll their history was just posting and commenting about artists on their own labels and nothing else,\u201d Monroe says. The accounts had throwaway random usernames and would swoop into threads mentioning artists on their labels. One especially suspicious comment went something like, \u201cI can\u2019t wait to preorder this [record] on Coral Green vinyl.\u201d It sounded, Monroe says, like something a social media manager would post on their brand\u2019s account. (He would know. Monroe also works as a marketing coordinator for a music distribution company.)<\/p>\n

\u201cWe ended up getting into contact with one of those labels being like, \u2018Hey, we know that you work here, so can we not do this please?\u2019\u201d As a compromise, r\/indieheads moderators launched a label verification process<\/a> that allowed record labels to participate in the subreddit while clearly identifying themselves to the public. Dozens of labels have signed up<\/a> so far.<\/p>\n

It\u2019s not clear whether labels are jumping into r\/indieheads because LLMs are citing Reddit, or if the activity is for general marketing and fan engagement purposes. But even without AI search, music subreddits like r\/indieheads have become influential, precisely because people trust Reddit as a neutral platform not yet filled purely with ads. r\/indieheads has become one of the \u201clast games in town\u201d when it comes to indie music marketing, Monroe says, as traditional publications have shuttered or pulled back their coverage.<\/p>\n

Part of what makes Reddit feel useful is that readers feel like they can find actual advice, product recommendations, or a new band to listen to without someone\u2019s thumb on the scale. After the Geese incident, some music fans seemed to feel uncomfortable with the idea that their new favorite band had been artificially placed in front of them on their feeds \u2014 \u201cWas my liking this music also artificial?\u201d some worried. It\u2019s an understandable reaction, Monroe emphasizes. But all the bots and user-generated advertiser content in the world can\u2019t place bodies in a stadium for a sold-out tour.<\/p>\n

\u201cWhat I do hope comes out of this all is just people realizing there\u2019s certain things you can\u2019t fake in the industry,\u201d Monroe says. You also can\u2019t astroturf your way into making listeners connect emotionally with music.<\/p>\n

On the flip side, the prevalence of stealth marketing campaigns could end up backfiring for everyone involved, not just the brands deploying anonymous accounts. Adivi, the moderator of r\/SkincareAddiction, says that often it\u2019s the lesser-known brands behaving \u201cegregiously\u201d in their marketing. Skincare subreddits have helped to propel smaller brands<\/a> into cult favorites, but the market has changed so much \u2014 and the subreddit has had to clamp down so much on moderation \u2014 that Adivi wonders if a new viral Reddit-approved upstart could ever take off the way they used to be able to.<\/p>\n

\u201cThe unfortunate thing is it does mean I end up being more suspicious of smaller brands, even though I would love to see more small brands out in the world,\u201d she says.<\/p>\n

Follow topics and authors<\/strong> from this story to see more like this in your personalized homepage feed and to receive email updates.<\/p>\n