Reddit Deploys AI Spam Filters, Blocking 23 Million Daily Views as Marketers Target LLM Search
Updated
Updated · The Verge · Aug 4
Reddit Deploys AI Spam Filters, Blocking 23 Million Daily Views as Marketers Target LLM Search
3 articles · Updated · The Verge · Aug 4
Summary
Reddit said in early July it upgraded spam detection with AI and LLMs to catch subtle, coordinated fake behavior, now intercepting about 25,000 spammy posts and comments a day.
The company said those tools are also blocking 23 million spam views daily as brands and agencies increasingly try to seed Reddit mentions that can later surface in chatbot answers and AI search.
Semrush data cited by The Verge showed Reddit was the most-cited domain in May by ChatGPT, Perplexity, Google Gemini and AI Mode, raising the value of even simple brand mentions.
Moderators across subreddits including skincare, weight loss and music communities say they are filtering suspected sock-puppet accounts, banning covert promotions and adding brands to keyword blocklists.
Reddit said roughly 40% of platform discussions are commercial, underscoring how AI-driven search visibility is turning volunteer-run communities into a new frontline against stealth marketing.
If AI search engines heavily rely on Reddit for answers, what happens when those trusted forums become entirely compromised by ghostwritten brand endorsements?
As marketers flood Reddit with stealth ads to manipulate AI chatbots, can human moderators save the internet's last bastion of authentic advice?
How Reddit’s AI Moderation Blocks 23 Million Spam Views a Day—and Why Its $550M Data Business Depends on It
Overview
As AI-generated content floods the internet, Reddit has positioned itself as the most human place online, defending its 'Authenticity Moat' against synthetic manipulation. Facing threats from bad actors using Generative Engine Optimization (GEO) to influence AI search results, Reddit deployed advanced large language models to block millions of spam views and reduce user exposure to fake content. However, this aggressive moderation brings privacy concerns and risks of silencing real users. Meanwhile, marketers adapt with hybrid human-in-the-loop tactics, and consumers grow more skeptical of online content, fueling a wider crisis of trust across the web.