OpenAI and the U.S. Department of Energy’s Pacific Northwest National Laboratory (PNNL) have partnered to deploy AI in accelerating federal permitting processes, aiming to streamline critical infrastructure development. This collaboration, detailed in a recent OpenAI News announcement, focuses on using coding agents to tackle the extensive documentation and review cycles that often delay projects like energy facilities and transportation systems.
The initiative targets the National Environmental Policy Act (NEPA) review process, notorious for its multi-year timelines. Through its PermitAI™ team, PNNL, alongside OpenAI, developed DraftNEPABench. This benchmark assesses how well AI models perform on NEPA workflow tasks, such as drafting environmental impact statements.
Nineteen subject matter experts evaluated generalized coding agents and found they could reduce drafting time by 1 to 5 hours per subsection, equating to a potential 15% reduction in overall drafting time. These agents, utilizing command-line interfaces, synthesize hundreds of pages of technical and regulatory content, verify facts across multiple sources, and draft structured reports meeting specific legal and technical criteria.
Modernizing federal permitting is crucial for U.S. economic competitiveness, particularly in sectors like energy, advanced manufacturing, and transportation. By enhancing AI government efficiency, these tools could free human experts to focus on complex judgment, oversight, and decision-making, moving beyond static PDFs to interactive, AI-generated reports and visualizations.
While the benchmark focuses on well-specified drafting tasks and available context, real-world deployments will involve more ambiguity and require expert feedback. However, the ambitious goal remains to reduce federal infrastructure project approval times from months to weeks, fostering faster development and strengthening U.S. competitiveness.
