A 6-model board for decisions: hide opinions, let Claude chair and adopt the sharpest argument
aryanXmahajan · x · 2026-11-20
A practical workflow for getting true second opinions from AI: query 6 models from 6 companies simultaneously, telling each it won't see the others and putting today's date at the top to prevent stale arguments. Then have Claude chair the review — not counting votes but comparing arguments, under the rule "follow the majority only if its reasoning holds; overrule it if a smaller group makes the sharper argument." The chair must show which objection changed its mind. In the author's first real call (picking an opening line for a giveaway), 4 models picked the safe option and 2 the bold one; Claude kept the strongest line and incorporated the one correct point from the losing side — one decision with the best losing argument built in, not 6 watered-down opinions.
More from coding & agent
- zlib: 2.4K-star open-source CLI turns Z-Library into a skill for AI agents — udmrzn · 2026-11-20
- Andriy Burkov's True Positive Weekly #186: what to know when agents do the work — burkov · 2026-11-20
- Raycast launches MCP: let Claude, Codex and any agent drive your Mac launcher — dean_rie · 2026-11-20
- Zed Delta agents can now run background terminals and get notified on completion — dean_rie · 2026-11-20
- Gemini app for Mac ships computer use, automating tasks across folders, apps and the web — GeminiApp · 2026-11-20
- Linear data: 53% of merged code now ships with AI review only, no second engineer — rseroter · 2026-11-20