AI Attacks on OT and Critical Infrastructure

Artificial intelligence is changing how industrial control systems, SCADA and critical infrastructure are attacked, and how they must be defended. AI Attacks on OT & Critical Infrastructure, a Premium Published Edition in the OT Rapid Response Series, is the practitioner reference for that shift.Part I sets out the AI-enabled adversary: the new OT threat model, how AI changes adversary economics, reconnaissance of industrial environments, and AI-assisted credential and identity attacks. Part II works through the attack surfaces and techniques one by one: engineering workstation compromise, PLC logic manipulation and safety impact, AI-generated industrial malware, autonomous agents in offensive operations, deepfakes and operator social engineering, digital twin abuse and simulation poisoning, AI against industrial protocols, ransomware with AI-assisted targeting, cloud-to-OT attack paths, and supply-chain and model dependency risk.Part III turns to defence, detection and recovery: industrial AI data poisoning, prompt injection in operational AI, securing AI copilots for engineers, human-in-the-loop safety controls, detection of AI-accelerated intrusions, OT SOC analytics for AI threats, incident response for AI-enabled attacks, and recovery and trusted-state reconstruction. Part IV covers governance for AI in critical infrastructure, board-level AI/OT risk decisions and future scenarios of autonomous industrial conflict.Three master extensions close the volume: quantitative risk and consequence engineering, purple-team control assurance, and evidence, metrics and data engineering. The book references the IEC 62443 family, NIST SP 800-82, the NIST Cybersecurity Framework, MITRE ATT&CK for ICS and CISA guidance throughout, and is written for defensive cybersecurity, engineering assurance, resilience, audit and governance use across power, water and manufacturing.Key topics include: AI-enabled OT threat modelling, industrial reconnaissance, PLC and safety system attacks, AI-generated malware, autonomous offensive agents, deepfake operator manipulation, digital twin security, industrial protocol attacks, OT ransomware, cloud-to-OT paths, AI supply-chain risk, OT SOC detection, incident response and recovery, and board-level AI/OT governance.