精选理由
有人给Codex一个递归的提示，让它造一个基准的基准，结果它真给你生成了完整的论文PDF，挺有意思的玩法。
用户向Codex模型发出一个递归提示：先构建一个评估AI生成基准能力的基准BenchBench，再推理出BenchBenchBench的含义并运行，最后将结果写成arXiv论文。Codex仅用单次提示就生成了一个完整的PDF文档。这个PDF内容展现了Codex理解复杂嵌套指令并产出的能力，且论文本身具有一定的参考价值。
原文 · Ethan Mollick
As a joke I prompted Codex "Build and run BenchBen…
As a joke I prompted Codex "Build and run BenchBench, a benchmark of now good ai is at creating benchmarks. then figure out what benchbenchbench is and run that. and then write benchbenchbench up as a good arXiv paper." I got a PDF.
But the paper is actually kind of interesting? https://t.co/yIIEvBxGw7