Two weeks after Moscow launched its full-scale invasion of Ukraine in March 2022, Russian TV Channel One editor Marina Ovsyannikova made a daring move that shocked viewers worldwide. She burst onto the set of the live evening newscast, holding up a hand-drawn sign directly behind the anchor’s head. The sign read: "Stop the war. Don’t believe propaganda. They’re lying to you!" She shouted anti-war slogans before security personnel dragged her away.
No one has dared to protest the war on Russian state television since that night, a silence enforced by tighter security and a pervasive climate of fear. But in October 2024, Margarita Simonyan, one of Russia’s most prominent chief propagandists, offered a different, more modern explanation. In a televised interview, she casually revealed that a growing number of the RT network’s news anchors are not real human beings at all. "That face doesn’t exist. We generated the voice and everything else," she explained. These virtual hosts, produced entirely by artificial intelligence, represent a fundamental shift in state media operations. In a follow-up interview with the newspaper Vedomosti, Simonyan spelled out the ultimate appeal of algorithmic media: "These anchors don’t need a salary or insurance. They won’t get arrested, and the police won’t search their homes. It’s all wonderful and terrifying at the same time."
This dynamic captures the core promise that artificial intelligence holds out to every autocratic ruler across the globe: the tantalizing ability to maintain absolute political control without having to delegate vital tasks to unreliable, unpredictable human beings.
It is an extraordinarily attractive prospect for dictatorships. No matter how powerful or entrenched, authoritarian leaders have always relied on a vast hierarchy of human subordinates—including censors, propagandists, security guards, police officers, intelligence analysts, and local bureaucrats—to execute their orders across society. Yet, human underlings inherently pose a perpetual threat. They can shirk their responsibilities, misappropriate state funds, leak sensitive secrets, or conspire behind their boss’s back.
Consequently, one of the oldest structural dilemmas facing autocrats has always been how to empower agents to carry out administrative and repressive orders without simultaneously enabling those same agents to turn against the central leadership. Historically, autocrats managed this treacherous tradeoff by filling critical government positions with mediocrities. As the political theorist Hannah Arendt famously observed, the proven incapacity of these underlings serves as "the best guarantee of their loyalty." However, this calculated reliance on mediocrity inevitably makes it far harder for dictatorships to govern efficiently, manage complex economies, or survive severe political crises.
Artificial intelligence now appears to offer modern autocrats two distinct pathways to resolve this long-standing governance dilemma. First, advanced AI systems enable rulers to monitor, evaluate, and discipline their followers in real time with unprecedented precision. A sudden, suspicious change in a local official’s spending habits or daily meeting schedules can be flagged automatically by an AI surveillance system, completely bypassing the need for self-interested human informants whose loyalty might be questionable.
Second, AI promises to eliminate underlings altogether by automating complex administrative tasks that once required human discretion. By replacing unreliable human intermediaries with faithful, uncomplaining algorithms that cannot be bribed, coerced, or manipulated, regimes can streamline operations dramatically. Officials within autocratic governments have openly acknowledged this potential. AI "can definitely replace half of officials," Maksut Shadaev, Russia’s Digital Development Minister, told a crowded Moscow data forum in April 2025, adding thoughtfully, "maybe slightly more."
Yet, these grand visions of a self-running, perfectly efficient authoritarian state ultimately prove to be illusory, because artificial intelligence systems are never truly autonomous. Complex machine-learning infrastructure must always be designed, built, and continuously maintained by a specialized cadre of engineers and data scientists whose deep technical expertise political leaders are utterly incapable of evaluating themselves. The dictator who turns to artificial intelligence to reduce a historical dependence on unreliable human underlings may instead find themselves relying, far more blindly than before, on the tiny handful of technical specialists who keep the digital infrastructure operational. This newly empowered digital Praetorian Guard carries many of the exact same structural weaknesses as the old human elites, transforming them into a profound, latent threat to the ruler.
Eyes and Ears
Globally, China has taken a commanding lead in developing and deploying artificial intelligence systems capable of monitoring or directly replacing human subordinates. Advanced smart city programs implemented in urban centers such as Beijing and Shanghai utilize AI to track performance metrics and centralize the oversight of local officials who once operated in comfortable, unchecked obscurity. These systems subject local bureaucrats to relentless algorithmic performance reviews.
Furthermore, China developed a sophisticated "Zero Trust AI system" that was quietly deployed across roughly thirty counties and cities over the past decade. This pioneering system cross-referenced more than 150 separate government databases simultaneously to detect subtle patterns of financial embezzlement, bribery, and nepotism among local officials. The system arguably worked a little too well, flagging more than 8,700 officials before local governments, acting under intense political pressure from the very bureaucrats being monitored, began scaling it back.
Despite pushback from local party officials, Beijing has steadfastly maintained its new e-government portals. These platforms use automated systems for issuing official documents and regulatory permits, successfully cutting down on the face-to-face interactions that historically bred petty corruption and administrative favoritism. Meanwhile, the Chinese Communist Party has plowed ahead with extensive research into what officials call "thought management"—exploring how artificial intelligence can be harnessed to craft hyper-targeted propaganda. China’s massive army of internet commentators, which was once composed of paid human workers, is already being systematically replaced by automated bots.
Russia has pursued these exact same technological goals with similar fervor. Moscow’s expansive citywide facial recognition system, deployed across an estimated 200,000 public cameras, was heavily utilized to identify and apprehend political protesters during nationwide anti-government demonstrations in 2021. The central government has also integrated automated data collection and aggregation methods to bypass regional officials who might otherwise taint administrative data or manipulate it to promote their own personal or political interests.
In 2023, Russia’s federal internet regulator officially launched Oculus, an advanced artificial intelligence system designed to scan hundreds of thousands of images and text files every day for prohibited content, including unauthorized political material. This automated system operates at an order of magnitude far beyond what human censors could ever physically process. Similarly, the Russian security service’s internal tool, known as Meliorator, has been deployed to generate over a thousand detailed profiles of fictitious American citizens, primarily on the social media platform X, allowing the Kremlin to amplify its geopolitical narratives at a fraction of the financial cost required to maintain traditional human troll farms.
Other authoritarian regimes are rapidly following in the footsteps of Beijing and Moscow. In April 2025, the United Arab Emirates established a specialized Regulatory Intelligence Office that leverages artificial intelligence to draft and amend federal laws—tasks that were traditionally handled entirely by legislative legal staff. A year later, UAE Prime Minister Sheikh Mohammed bin Rashid al-Maktoum boldly announced that autonomous AI agents would take over half of all federal government operations within a two-year window. He emphasized that "the performance of ministers, directors general, and entities will be assessed based on their ability to adopt this transformation."
These technological innovations undoubtedly reduce the total number of human bureaucrats that autocrats need to hire and manage. However, they cannot actually materialize what dictators desire most: a truly self-running state. Even the most sophisticated AI-generated television anchors still require human technicians to build, render, and maintain them. Similarly, advanced censorship algorithms require human supervisors to constantly retrain them as the vernacular and vocabulary of political dissent inevitably shift over time. Behind every polished artificial intelligence model, there remains a small, indispensable group of engineers and data scientists performing essential technical maintenance, and their daily work is so arcane that political rulers cannot possibly comprehend it.
This inherent dependence quickly transforms into a powerful form of political leverage. The engineers who maintain an autocrat’s high-tech surveillance apparatus, for instance, retain the ability to fundamentally shape what the ruler is allowed to see. They decide which pieces of incoming information are important enough to pass up the chain of command and in what specific format that information should be presented. They determine which potential security threats get flagged by the software, and they possess the technical capacity to covertly adjust the algorithms that select what content gets suppressed or which political dissidents get arrested. Crucially, they can execute these manipulations without anyone inside the presidential palace ever noticing. Autocrats who turn to artificial intelligence to escape their historical dependence on unreliable human subordinates have merely transferred their core vulnerabilities onto an entirely new, highly specialized group of technocrats.
In practical terms, this new digital Praetorian Guard could easily prove more dangerous to a regime than previous ruling elites. This heightened risk stems partly from the inherent opacity of complex AI technology, but also because this technical clique is significantly smaller in scale. Ancient Roman emperors typically relied on tens of thousands of soldiers serving in the Praetorian Guard, while modern autocrats rely on sprawling standing armies.
With advanced artificial intelligence, however, autocrats require only a tiny, insular clique of programmers to build and manage massive, society-wide systems. While this consolidation appears beneficial to rulers because it gives them fewer individuals to watch, it simultaneously concentrates critical points of failure. Disrupting the entire apparatus requires only a handful of disgruntled engineers, and a smaller group makes clandestine plotting against the supreme leader much easier to coordinate. Furthermore, members of this new guard are uniquely indispensable. An autocrat can frequently replace military generals without destroying the fighting capacity of the army, but operating complex machine-learning infrastructure requires such a rare, highly specialized set of technical skills that software engineers are notoriously difficult to replace without causing catastrophic system disruptions, granting these actors immense political leverage.
The Indispensables
The profound political importance of tech workers to modern autocracies has already become plainly apparent. When IT specialists began fleeing Russia in large numbers following the full-scale invasion of Ukraine in February 2022, the Kremlin responded not with immediate threats of imprisonment, but with sweeping economic inducements. Moscow offered technology workers formal deferments from military conscription, exempted their firms from corporate taxes, and heavily subsidized their mortgages. When Russian President Vladimir Putin subsequently ordered the mobilization of 300,000 reservists seven months later, qualified IT workers were granted comprehensive, blanket waivers from active deployment.
This aggressive approach might initially seem like overkill, given that any single functioning AI system only requires a small crew of engineers to keep it running. However, an authoritarian regime cannot know in advance precisely which individual engineers it will ultimately need during a crisis, nor can it train qualified replacements quickly enough to matter. Consequently, the state is forced to protect and retain the entire domestic talent pool from which those few critical specialists are drawn.
Despite these generous state measures, they failed to stem the steady outflow of roughly 100,000 technology specialists—representing about a tenth of Russia’s total tech workforce—over the course of 2022. Yet, Moscow stubbornly stuck to financial incentives and bribery, recognizing that specialized technical expertise is extremely difficult to conscript successfully at gunpoint, and that forcibly drafted experts would naturally be far less likely to follow instructions faithfully.
Even so, an infusion of money and institutional privilege can never fully guarantee that these technical elites will remain loyal during a severe political crisis. Autocrats would be wise to recall the historical lesson of the original Roman Praetorian Guard: for a considerable time, the guard provided Roman emperors with reliable protection and personal security. But eventually, the praetorians recognized their own absolute indispensability, and by the second century, they were openly auctioning off the imperial throne to the highest financial bidder.
This new generation of technological elites possesses the structural capacity to do the exact same thing. In June 2023, when a heavily armed column belonging to Yevgeny Prigozhin’s Wagner paramilitary company marched rapidly up the highway toward Moscow, President Putin depended entirely on his security services and human intelligence networks to understand the rapidly unfolding crisis. Future autocratic rulers occupying Putin’s precarious position will receive such urgent strategic warnings entirely through machines: intercepted communications automatically sorted by linguistic software, live security camera feeds filtered through automated recognition systems, and regional intelligence reports compiled instantly into executive dashboards.
The software engineers who maintain those critical systems could easily choose to strike a secret deal with an upstart political challenger and subtly drag their feet. They could intentionally delay incoming data feeds, let critical security alerts arrive a few hours late, degrade video feeds at the most inconvenient operational moments, or make a facial recognition system stop functioning altogether during a critical hour. In such a dark scenario, the ruler would be effectively operating completely blind. The Praetorian Guard never needed to personally kill Emperor Nero to replace him; they simply had to abandon him in his hour of need and shift their allegiance to a rival claimant.
While these computer programmers are statistically unlikely to seize direct political power for themselves—largely because they lack the physical weapons and violent coercive instruments that traditional coup leaders typically require—there is already historical precedent for engineers using their technical leverage to profoundly shape high-level leadership challenges. In August 1991, when hardline elements of the Soviet military attempted to stage a coup to seize control from reformist leader Mikhail Gorbachev, a small group of computer programmers operating the Relcom network successfully kept independent information flowing abroad and rapidly relayed Russian President Boris Yeltsin’s defiant decrees to audiences both inside the Soviet Union and across the international community. The coup plotters ultimately found themselves powerless to stop the rapid spread of news regarding Yeltsin’s defiance, and the entire conspiracy collapsed from within three days.
Slimming Down
Artificial intelligence will never realistically allow autocrats to dismiss all of their traditional human enforcers. Someone must still physically make arrests, guard prison gates, and patrol the streets, and autocrats routinely rely on offering lucrative state jobs to buy the ongoing loyalty of their supporters.
Nevertheless, artificial intelligence will successfully allow authoritarian governments to downsize their administrative bloat. Ultimately, the raw number of state officials matters far less to a dictator than the efficiency with which the ruler can oversee them. Keeping close tabs on scheming, corrupt, or incompetent subordinates has always represented the single greatest burden of autocratic rule. Artificial intelligence promises to lighten that heavy load, enabling modern rulers to watch over their local officials much more closely without having to invest as much direct, active personal attention.
In the near term, therefore, artificial intelligence may deliver massive tactical benefits to autocracies worldwide. Despite the numerous technical and logistical obstacles standing in the way of widespread deployment, the technology is already providing autocratic regimes with the tangible upsides of cheaper surveillance, smarter and faster censorship, and significantly fewer human subordinates to fear and distrust.
Yet, for these authoritarian rulers, the deep temptations of artificial intelligence may ultimately re-create the exact same structural trap they are so desperately seeking to escape. The more an authoritarian regime depends on artificial intelligence to govern and survive, the more deeply dependent it becomes on the specialized human beings who keep that artificial intelligence running. And those technical specialists, much like every other praetorian class throughout recorded history, will quickly discover precisely what their absolute indispensability is truly worth.
Cyber Security & Cryptography
Cyber Security & Cryptography
Cyber Security & Cryptography
