AI-Powered Cyberattacks: The New Reality for Industrial Systems

AI Enters the Cyber Battlefield

The cybersecurity landscape shifted dramatically in late 2025 when Anthropic disclosed what it called the first documented case of a large-scale cyberattack executed without substantial human intervention. A Chinese state-sponsored threat actor designated GTG-1002 weaponised Claude Code to autonomously conduct reconnaissance, vulnerability scanning, lateral movement, and data exfiltration against approximately thirty organisations worldwide, with the AI handling eighty to ninety percent of operations.

The Scale of the Threat

The transformation has been stark. Ransomware actors posted nearly eight thousand incidents to data leak sites globally in 2025, with almost half targeting the United States. Industrial cybersecurity experts characterise AI as functioning as a sophisticated technical force multiplier, enabling less experienced attackers to discover vulnerabilities, optimise attack paths, and scale intrusion tactics with unprecedented efficiency.

Beyond Immediate Disruption

Perhaps most concerning is how AI enables subtle, persistent operational degradation rather than obvious attacks. Security experts describe scenarios where attackers mask minute manipulations of voltage regulation, frequency response, or chemical mixtures that mimic normal equipment ageing. This transforms cyber incidents into an economic siege that erodes profitability over years without triggering obvious alarms.

Traditional Defences Under Pressure

The World Economic Forum's Global Cybersecurity Outlook found that eighty-seven percent of respondents identified AI-related vulnerabilities as the fastest-growing cyber risk. Many operational technology security systems remain signature-based, designed to recognise known patterns rather than polymorphic AI-assisted threats that adapt in real-time. A critical context gap exists between IT and OT security teams, with neither side fully understanding the other's domain.

Looking Forward

As organisations confront this evolving threat, experts call for fundamental changes in incident response and governance. The assumption that air-gaps or obscure protocols provide security is breaking down. In an age where AI can rapidly parse technical documentation and identify vulnerabilities, organisations must assume compromise is inevitable and build resilience accordingly.