Artificial intelligence is giving autocratic rulers a powerful new way to automate surveillance, eliminate unreliable human subordinates, and monitor citizens in real time. However, this reliance on complex machine-learning infrastructure creates a dangerous dependency on elite engineers, shifting vulnerabilities onto a new digital Praetorian Guard.
The Algorithmic Bureaucracy of Beijing and Moscow
Autocrats have long faced a structural dilemma: they must empower subordinates to execute policy, but those same agents pose a constant threat of subversion, corruption, or disloyalty. Historically, regimes solved this tradeoff by appointing mediocrities whose incompetence guaranteed obedience. Today, authoritarian states are turning to machine learning to bypass human intermediaries entirely.
China pioneered this approach through smart city monitoring programs and its “Zero Trust AI system,” which cross-referenced over 150 government databases to flag embezzlement and nepotism. China also developed a “Zero Trust AI system,” which it deployed across roughly 30 counties and cities over the past decade. This system cross-referenced more than 150 government databases to catch embezzlement and nepotism by local officials.
Russia has pursued parallel automated controls. Moscow integrated a citywide facial recognition infrastructure spanning 200,000 cameras to identify and arrest dissidents during the 2021 protests. In 2023, the federal internet regulator deployed Oculus, an AI system engineered to scan hundreds of thousands of images daily for prohibited political content. Meanwhile, the Russian security service utilized a profiling tool called Meliorator to generate over a thousand fictitious profiles on X, formerly Twitter, to distribute Kremlin narratives at a fraction of the cost required by traditional human troll farms.
Other authoritarian states are accelerating similar transformations. In April 2025, the United Arab Emirates established a Regulatory Intelligence Office using AI to draft and amend federal legislation. Sheikh Mohammed bin Rashid al-Maktoum, the UAE Prime Minister, announced that autonomous AI agents would assume control of half of all federal government operations within two years, tying ministerial evaluations directly to their adoption rates.
How Artificial Intelligence Replaces Human Intermediaries
Yet these systems remain fundamentally dependent on human labor. Complex models require continuous maintenance, retraining, and architectural oversight from specialized engineers and data scientists. Because political leaders lack the technical literacy to evaluate this specialized work, the autocrat’s reliance on unreliable human underlings is merely replaced by a blind dependency on a small, indispensable cadre of programmers.
The Vulnerability of the Digital Praetorian Guard
This dynamic mirrors the historical vulnerability of Roman emperors relying on the Praetorian Guard. While modern autocrats require fewer enforcers to maintain algorithmic infrastructure, this extreme concentration of technical talent creates a single point of catastrophic failure.
Machine-learning systems demand rare expertise that cannot be easily or rapidly replaced. When thousands of IT specialists fled Russia following the full-scale invasion of Ukraine in February 2022, the Kremlin responded not with coercion, but with economic inducements, exempting tech firms from taxes, subsidizing mortgages, and granting full conscription waivers to IT workers—even as Vladimir Putin ordered the mobilization of 300,000 men later that year.
Despite these concessions, roughly 100,000 specialists departed over the course of 2022. Money and privilege cannot guarantee long-term loyalty when technical elites recognize their own indispensability. In a crisis scenario, such as the June 2023 march toward Moscow led by Yevgeny Prigozhin’s Wagner paramilitary company, future rulers will rely entirely on automated data feeds, facial recognition outputs, and software-sorted communications. Engineers who control those pipelines could manipulate data feeds, delay security alerts, or degrade recognition systems, effectively blinding the regime without ever needing to pick up a weapon.
Related reading
