USA: The US Department of Energy has awarded $750,000 to University of Arkansas researcher Qinghua Li to develop tools designed to protect AI systems used in power grid operations from cyberattacks.
AI is increasingly used to detect faults, classify grid events and support rapid operational decisions. However, adversarial attacks can manipulate input data and cause AI models to misidentify problems or trigger incorrect responses, potentially increasing the risk of wider grid disruption.
Li’s project, “An Autonomous Framework for Adversarial Robustness and Resilience of AI Models in Power Grids,” will develop an autonomous toolkit capable of identifying and defending against hidden vulnerabilities in AI systems supporting grid operations.
Unlike research focused on individual AI models, the project will examine complete workflows made up of multiple connected AI models, aiming to identify weaknesses across the wider decision-making process.
The research forms part of the Department of Energy’s Genesis Mission, a national initiative launched in 2025 to use AI and supercomputing for scientific research and national security.
Xintao Wu, professor of electrical engineering and computer science at the University of Arkansas, is a co-investigator. Brookhaven National Laboratory and Southern Methodist University are also participating in the project.
The work is expected to support faster vulnerability detection and improve the security of AI-enabled power grid infrastructure, while also providing training opportunities for engineering students.
Source: University of Arkansas
