Details
- AWS unveiled Agentic Grid Planning on AWS in September 2026, offering utilities access to AWS-managed AI agents to accelerate transmission interconnection studies and grid planning workflows.
- The program is being developed in close collaboration with Duke Energy, whose engineers use the agents with existing physics-based simulation software, grid models, scripts, and engineering standards.
- AI agents automate data preparation, study execution, and workflow coordination while engineers retain control over study design, deterministic modeling, and all final engineering decisions.
- The initiative targets growing interconnection backlogs, with more than 2,000 GW of generation and storage seeking U.S. transmission interconnection at end-2025 and ERCOT large-load requests reaching roughly 410 GW in March 2026.
- AWS positions the program as a tool for utilities and grid operators to meet new regulatory requirements such as FERC Order No. 2023’s first-ready, first-served cluster studies and study deadlines, supporting faster, auditable, AI-assisted planning.
Impact
By embedding AI agents directly into established grid planning toolchains, AWS is pushing AI deeper into critical energy infrastructure workflows, potentially easing engineering bottlenecks created by the interconnection backlog. Over the next 12–24 months, this approach could influence how utilities staff planning teams, prioritize digital twins and workflow automation, and allocate capex amid tightening FERC timelines and rising load growth.
