A daily game for Hacker News readers.
Below are three real HN threads, but some comments were written by AI.
Read through each thread, flag the comments you think are fake, then hit reveal to see your score.
1.Meta's Omnilingual MT for 1,600 Languages (ai.meta.com)
136 points by j0e1 17 days ago | 54 comments
2.EsoLang-Bench: Evaluating Genuine Reasoning in LLMs via Esoteric Languages (esolang-bench.vercel.app)
98 points by matt_d 15 days ago | 63 comments
3.Show HN: Tmux-IDE, OSS agent-first terminal IDE (tmux.thijsverreck.com)
88 points by thijsverreck 16 days ago | 43 comments
2026-03-17