Spencer Greenberg had AI fact-check his essays — it caught a real reversed inequality
SpencrGreenberg · x · 2026-11-02
Spencer Greenberg (Clearer Thinking founder) shares a practical self-audit workflow: ask the most powerful AI available to find as many factual errors as it can in your published work, stating what you claimed, what's actually true, and linking sources. In his own run, "Error 1" was real — the AI correctly spotted a reversed inequality in his 2017 essay on estimating confidence intervals (should be n ≥ (Q/CI₀)², not n ≤ (Q/CI)²). He fixed it and is publishing further errors found in a thread.
Related event: Spencer Greenberg Lets AI Fact-Check His Own Writing(4 posts)→
More from Apps
- Reader proposes PID-controller feed aggregator to replace ad-driven social algorithms — xiaosun86 · 2026-11-02
- Dev shares prompt to make AI generate texture-free volumetric fire in pure three.js shaders — majidmanzarpour · 2026-11-02
- Students Failed to Spot Deliberate AI Lies in Korean Classroom Experiment (Open Source) — Working_Original9624 · 2026-11-02
- HackerTube's first recommendation nails it: a Lex Fridman UFO deep-dive — daniel_mac8 · 2026-11-02
- Perplexity's Computer builds an unofficial GTA VI map pairing Leonida landmarks with real Florida — AravSrinivas · 2026-11-02
- Non-dev builds Android DJ app with Claude Code: on-device stem separation at 2x realtime — SuryaVardhan1 · 2026-11-02