OpenClaw Founder Uses AI to Automatically Block Twitter Spam Replies, Blocks 56 Accounts Within 24 Hours

Gate News: On March 16, OpenClaw founder Peter Steinberger tweeted that he built a Twitter mention auto-blocking system using OpenClaw, which runs every five minutes and has exceeded expectations. He receives a daily summary email called the Daily X Block Digest, listing blocked accounts and AI’s reasons for blocking. In the past 24 hours, the system has blocked 56 accounts. The characteristics of blocked accounts shared in the summary include: 1) small accounts posting generic praise under unrelated posts generated by AI; 2) cryptocurrency project accounts posting low-quality comments with a history indicating token promotion; 3) spam replies consisting only of images, recently active in staking/shill content; 4) conspiracy-style long accusations across posts with low signal-to-noise ratio; 5) directly promoting their products by piggybacking on celebrities’ replies; 6) hollow comments from degen/crypto accounts. One user asked, “Is this blocking reply guys (users who habitually jump into celebrity tweets for exposure)?” Steinberger replied, “yes.” Regarding false positives, Steinberger said he evaluates each appeal individually, but “usually it’s bad behavior, and AI’s judgment is correct.” He also added that using AI solely for writing does not trigger blocking, “as long as the content itself is meaningful,” but doing so “will make you less popular.”