OpenAI Report: AI Coding Agents Can Accelerate Scientific Research Software But Cannot Determine Scientific Correctness

TechFlow reports, August 1: Field reports from OpenAI and academic partners show that programming agents such as Codex and Claude Code can accelerate legacy research software by more than 60 times. RustQC reduced runtime from 15 hours 34 minutes to 14 minutes 54 seconds, and HelixForge is 59.6 times faster than BamSurgeon. However, agents cannot reliably determine whether results are scientifically correct, often confidently outputting incorrect code, requiring humans to define testing and validation standards.