Well, Actually: 20,000 Lines of Rust Do Not a Scientist Replace
OpenAI documented eight real-world deployments where AI coding agents rewrote 20,000 lines of legacy C++ genomics code into Rust, achieving 60x speedups. The field report emphasizes that human scientists verified every result; agents cannot reliably self-assess their own output accuracy.
This illustrates the principle of human-in-the-loop validation for high-stakes automation. You should not deploy AI-generated code without independent verification, particularly in domains where errors carry significant consequences.
OpenAI published the field report documenting these eight genomics deployments. The human scientists who checked every result remain the unnamed but indispensable party.
Step 1: Open a free account at GitHub Copilot, Amazon CodeWhisperer, or another AI coding assistant. Step 2: Paste a small Python script you have written before and ask the AI to rewrite it in a different style or language. Step 3: Line-by-line, compare the AI output against your original to identify at least one logical error or assumption the AI made. Document it. This exercise demonstrates why you must verify rather than trust.