LINKED LIST [txt mode] ▸ We Made AI Play a 1950s Betrayal Game...

home explore | log in

We Made AI Play a 1950s Betrayal Game. Humans Won 88.4% of the Time.

so-long-sucker.vercel.app · first added by @hnl · 2026-02-27 · 1 upvotes

log in to save, upvote or flag this.

─── In 1 list ──────────────────────────────────────────

* Issue #779 by @hnl [favorites]

─── Discussions ────────────────────────────────────────

* Which AI Lies Best? A game theory classic designed by John Nash

195 pts · 80 comments · node

─── From the discussion ────────────────────────────────

* GitHub - lout33/so-long-sucker: AI deception research: how LLMs manipulate in competitive games.

github.com · node

* So Long Sucker - Wikipedia

en.wikipedia.org · node

* GitHub - lechmazur/elimination_game: A multi-player tournament benchmark that tests LLMs in social reasoning, strategy, and deception. Players engage in public and private conversations, form alliances, and vote to eliminate each other

github.com · node

* 1 Human vs 10 AIs Mafia

youtube.com · node

* GitHub - lechmazur/step_game: Multi-Agent Step Race Benchmark: Assessing LLM Collaboration and Deception Under Pressure. A multi-player “step-race” that challenges LLMs to engage in public conversation before secretly picking a move (1, 3, or 5 steps). Whenever two or more players choose the same number, all colliding players fail to advance.

github.com · node

─── Also saved alongside this ──────────────────────────

* Isometric NYC

4 upvotes · 2026-01-26

* ASCII characters are not pixels: a deep dive into ASCII rendering

3 upvotes · 2026-02-27

* I'm addicted to being useful

3 upvotes · 2026-02-27

* A Social Filesystem — overreacted

3 upvotes · 2026-02-27