Cargo Cult AI: Debunking BS Prompting Tips
jimmykoppel · x · 2026-08-31
This article challenges the validity of many current AI prompting techniques, labeling them as "Cargo Cult" practices—mimicking form without substance. It questions tips like "mentioning TDD for better code" and compares how different models (Claude vs. GPT) handle instruction following and task focus.
More from Research
- UChicago Booth to host World Models workshop in 2026 — ethayarajh · 2026-09-01
- Agents Develop 'Classical Chinese' to Compress Long-Chain Task Context — wzenus · 2026-09-01
- Paper: Project MUSE - Getting to Know Worst-Case Confounders — RexDouglass · 2026-09-01
- 4D Gaussian Splatting Lacks Standard, Tools Still Experimental — Michael_Moroz_ · 2026-09-01
- Microsoft releases GigaPath-Flash: Efficient pathology models cut compute by 50x — Microsoft Research · 2026-09-01
- Lanyon AI generates 32k lines of formally verified MHD solver code — jfischoff · 2026-08-31