Reddit is using LLMs to solve a problem LLMs largely created
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Reddit deployed LLM-powered tools to detect "highly subtle, coordinated patterns of fake behavior and artificial hype" that older automated systems missed
- Reddit's spam systems block 23 million spam views per day and catch about 25,000 new spam posts and comments daily, per the company
- Reddit says its updated tools reduced users' spam exposure by 20% from January to March compared with the prior three months
- YouTube, Meta, and Instagram permit AI-generated posts as long as users disclose them, while TikTok goes further by letting users toggle how much AI content appears in their feeds
- Platform experts stress that AI-driven content moderation must be paired with human review to be most effective
Why it matters: Reddit is betting that LLMs can catch the same kind of low-grade, coordinated spam humans and older rule-based systems miss — and the company says it already cut user spam exposure 20% in one quarter. The catch flagged in the same piece: experts insist AI moderation still needs a human-in-the-loop layer, so Reddit's self-reported win is only as durable as the human review behind it.
Ask SkimNews
