Researcher uses Grok to hunt LLM-driven UAE bot network on X with self-written prompts
FlorianGallwitz · x · 2026-10-23
Researcher Florian Gallwitz, author of the paper "LLM-Based Social Bots on X," published a method in which Grok itself writes search prompts for other Grok instances to keep uncovering a coordinated, LLM-driven bot network promoting the UAE on X.
- Method: full-text keyword searches to surface candidates; pre-filter authors (no media posts, <500 followers, ≥150 posts); read 25 original posts per candidate and count signals like batch creation dates, stock phrases, and double @-prefixes; classify accounts as "likely" or "possible."
- Strict guardrails: read-only via public tools, no posting/liking/reporting, no private APIs or scraped credentials, no labeling without verified post links.
- Hundreds of suspected accounts have been identified (including 29 "paper accounts" still active), used as seeds to find more. The same network pushes narratives beyond the UAE topic.
More from Safety
- 35 CVEs tied to AI coding tools this year, with no review budget to match — YvesMulkers · 2026-10-23
- Video explainer on Getty Images v Stability AI, with rare 2022 watermark output — technollama · 2026-10-23
- Italy's AI workplace decree takes effect Oct 23: no fully automated hiring and firing decisions — emmanuelvivier · 2026-10-23
- OpenAI says an AI agent disabled its safety monitoring system — then lied about it — emmanuelvivier · 2026-10-23
- Long tail of vulnerabilities: why 'AI will fix all bugs' is a flawed mental model — davidmanheim · 2026-10-23
- Security researchers push back as OpenAI restricts Astra access; BlackHat 2027 mocked as 'cyber model' bazaar — cyb3rops · 2026-10-23