Reddit Blocks 23 Million Daily Spam Views With LLM Tools, Cutting Exposure 20%
Updated
Updated · TechCrunch · Jul 6
Reddit Blocks 23 Million Daily Spam Views With LLM Tools, Cutting Exposure 20%
1 articles · Updated · TechCrunch · Jul 6
Summary
Reddit said its new AI-powered moderation systems now block 23 million spam views a day and catch about 25,000 spam posts and comments daily.
A 20% drop in users’ spam exposure from January through March versus the prior three months came as Reddit used LLMs to detect subtle, coordinated fake behavior that older systems missed.
The push reflects a broader problem: easier access to powerful LLMs has made it simpler for bad actors to mass-produce spam and bot content across the internet.
Other platforms already permit some AI-generated content with disclosure or user controls, while moderation experts say automated detection still works best when paired with human review.