X Auto-Reply Bot Suspected of Leaking System Prompts
repligate · x · 2026-07-15
Users discovered that an auto-reply bot on X might be running on Fable or Claude models. When refusing to reply, the bot accidentally exposed its "detailed operational instructions" and "complex system prompts," which even included rules to skip tweets not meeting specific criteria.
More from Fun
- A meme stitches together Claude and Grok quota resets into one AI-user joke — djcows · 2026-07-22
- A SymPy joke turns model tool use into a “neurosymbolic architecture” gag — thomasahle · 2026-07-22
- A meme about AI apps looking great until someone plugs them into Slack — generativist · 2026-07-22
- A VR teleop demo for an SO-101 arm gets absurdly low latency by using one Python script — MoonL88537 · 2026-07-22
- Project CETI gets a Jeopardy! shout-out with a SETI-style whale clue — begusgasper · 2026-07-22
- Claude portfolio says it has passed $20M on autopilot after a $50K test — alejandroll10 · 2026-07-22