FERC audit identifies AI and cybersecurity risks in U.S. electric grid

Federal regulators and state officials are raising concerns regarding the integration of artificial intelligence and cybersecurity vulnerabilities within the United States electric grid. A Federal Energy Regulatory Commission (FERC) audit for fiscal year 2026 identified risks involving inverter-based generation resources operating without proper registration or cybersecurity evaluations. The audit also noted instances where affiliates and third-party providers controlled multiple generation resources through shared centers that met medium-impact criteria, despite being classified as low impact.

Furthermore, the FERC report highlighted risks associated with AI tools, such as chatbots and automated analytical tools, operating within protected environments. Auditors warned that these systems can mimic legitimate user behavior, potentially making malicious activity harder to detect if access controls are not strictly monitored.

In response to these growing technological shifts, Connecticut Attorney General William Tong and Consumer Counsel Claire Coleman have petitioned the Public Utilities Regulatory Authority to investigate how state electric, gas, and water utilities deploy AI. The petition seeks transparency regarding the scope, location, and oversight of AI systems used for grid operations, storm response, and customer services.