DraftNEPABench
HoldTools
A benchmark for measuring how AI coding agents perform on federal permitting tasks.
Why it's here
Placed in Hold: 1 article(s) of evidence from 1 source(s), led by research-stage coverage, with 0 in the last 30 days. Confidence 24%. Low accumulated evidence, so it defaults conservatively pending more signal.
Evidence (1)
- 6OpenAI Blog·2/26/2026researchOpenAI and PNNL launch DraftNEPABench for federal permitting
OpenAI and Pacific Northwest National Laboratory introduced DraftNEPABench, a benchmark for evaluating how AI coding agents can speed up federal permitting work. The effort targets NEPA drafting and infrastructure review workflows, with reported potential to cut drafting time by up to 15%.