Built Among Us Sim to Test LLM Deception: Claude vs. Kimi
panickssery · x · 2026-08-22
Inspired by the Generative Agents Simulacra paper, the author built simulations to observe how LLMs play games against each other.
- Testbed: Chose "Among Us" (Mafia style game) and mapped every pixel to build a custom sim (pinned on profile).
- Matchup: Had Claude play against Kimi to gather data on LLM deception in game scenarios.
- Duration: Working on variations of this idea since October 2025 (10 months).
Related event: Researcher Builds Among Us Testbed to Study LLM Deception and Game Play(6 posts)→
More from coding & agent
- Are the fastest GPU kernels now written by AI? Kernels are hyper-verifiable, argues one dev — fleetwood___ · 2026-10-16
- Anthropic lead engineer's 4-question prompt makes agents more ambitious, now a free skill — daniel_mac8 · 2026-10-16
- ChainForge dev shows 'feature map' for orchestrating subagents on large PRs — IanArawjo · 2026-10-16
- aiDotEngineer benchmarks agent orchestration pipelines, staging humans as Foreman and workers — vikvang1 · 2026-10-16
- Developer mass-produces 200 videos in hours using Claude Code + Hyperframes — mcraddock · 2026-10-16
- Penny University hosts software factory discussion; dev shares $40/mo vendor-agnostic AI stack — JnBrymn · 2026-10-16