Turning to AI to Safeguard Critical Infrastructure

Tel Aviv University's Harel on Strengthening Anomaly Detection in Energy Systems Michael Novinson (MichaelNovinson) • November 13, 2025

Critical infrastructure faces rising threats that target outdated systems. Yaniv Harel, deputy head of cyber center at Tel Aviv University, said energy systems need stronger defenses that track physical behavior inside production environments and identify deviations before they escalate into cyberattacks.

See Also: Manage End-to-End Risk Including 'No AI Without AI Security'

Systems trained to interpret small signal changes, frequency shifts and other plant-level characteristics give defenders a clearer view of what normal behavior looks like.

But Harel also warns that AI introduces new weaknesses. Organizations often secure production systems but overlook training data stored in labs. Attackers can corrupt that data long before a model reaches deployment. "If I'm an attacker, I would target the training data in your lab, because no one really secured it," he said.

In this ...