Posting referential tweets with subtext is a great way to expose AI bot replies
Kyrannio · x · 2026-11-07
A practical trick for outing AI bot accounts: post something highly referential with subtext and watch whether replies interpret it hyper-literally. Humans who get the reference engage with the subtext; bots just paraphrase it literally.
Related event: A Single Meme Tweet Is Enough to Expose AI Bot Accounts(2 posts)→
More from Fun
- Developer gets emotional finding pre-AI-era code he wrote by hand — prasenx · 2026-11-07
- Meme: Sam Altman doesn't want you to have RAM in your PC — penpcm · 2026-11-07
- Theo jokes "today is my last day at OpenAI" — see you Monday — Dimillian · 2026-11-07
- Dimillian announces today is his last day at OpenAI — then jokes 'See you Monday' — Dimillian · 2026-11-07
- First CHI review thanked the authors, prompting ICLR comparison — chhaviyadav_ · 2026-11-07
- Playing rock-paper-scissors with a robot next to a flying car: 'We live in the future we were promised' — beffjezos · 2026-11-07