Live

Albania’s AI “Minister” and the Temptation of Algocracy

In a world-first move, Albania has appointed an artificial intelligence system as a cabinet-level “minister” to oversee all government procurement. This virtual official, named Diella, is tasked with deciding which private suppliers win public contracts a responsibility involving over $1 billion worth of goods and services each year.

Prime Minister Edi Rama introduced Diella in September 2025, touting her as “the first cabinet member who isn’t physically present, but is virtually created by AI,” and promised she will help make Albania “a country where public tenders are 100% free of corruption” . The hope is that an incorruptible algorithm will succeed where human officials often failed, eliminating the bribery and favoritism that have long plagued Albanian public tenders .

This bold step towards what some call “algocracy” – government by algorithm  has captured global attention. If democratic institutions can’t deliver honest and efficient governance, might an AI system do better? It’s a compelling proposition for a country beleaguered by corruption. Albanian media lauded Diella’s appointment as “a major transformation… introducing technology not only as a tool, but also as an active participant in governance” .

This is huge and it could actually set a precedent, Bojana Zorić, a policy analyst at the European Union Institute for Security Studies, said. However, not everyone is convinced by this techno-utopian fix.

Skeptics abound on social media; one Albanian citizen commenter quipped, “In Albania, even Diella will be corrupted,” expressing doubt that an algorithm can stay immune to the country’s entrenched graft culture . The world is now watching closely to see whether Diella will truly root out corruption or simply add a new layer of opacity to government decisions.

A Crisis of Democracy and the Lure of AI

Albania’s experiment comes at a time of widespread disillusionment with democratic governance in many countries. Across 12 high-income democracies, a majority of citizens (median of 64%) say they are dissatisfied with the way democracy is working according to a study conducted by Pew Research Organization in June 2025.

From Europe to Asia and the Americas, protests regularly erupt over governments seen as distant, ineffective, or corrupt. Broken trust in institutions is visible in “turnstiles ablaze, smashed storefronts and streets choked with tear gas,” as people vent their frustration with leaders who seem out of touch . In this climate of democratic malaise, the idea of handing certain decisions to unbiased algorithms starts to sound appealing.

At the same time, AI technology is rapidly becoming more powerful and prevalent. Advanced models can now outperform humans in specialized tasks like medical image analysis or complex problem-solving . The public is growing familiar with AI through tools like chatbots, and interestingly, many are open to trusting AI over politicians.

In global surveys conducted in 2025 by Global Collective Intelligence Project (CIP), people consistently said an AI chatbot could make better decisions on their behalf than their elected representatives . In other words, as democratic institutions struggle, confidence is shifting toward artificial intelligence.

This pattern is not new  historically, when democracy fails to deliver, populations often turn to strongmen or authoritarian alternatives. Now that age-old reflex is pointing to algorithms: a hope that competent, data-driven AIs might succeed where messy human politics has not .

It’s easy to see the temptation. Algorithms appear neutral and efficient. They won’t take bribes, trade favors, or get bogged down in partisan bickering. Faced with a government that can’t fix chronic problems, many citizens find the promise of an “AI fix” seductive.

Why not let a hyper-rational machine make decisions instead of squabbling politicians? Albania’s Diella is a dramatic test of that proposition, and other countries with weak institutions are likely watching, perhaps ready to follow suit if it shows success.

Why Algorithms Can’t Replace Democratic Values

Despite the allure, simply replacing democratic deliberation with algorithmic decision-making is a dangerous trade-off. Democratic governance isn’t just about efficiency; it’s about reconciling competing values and interests in society. Algorithms are excellent at optimisation, but they cannot decide fundamental moral and political questions.

For instance, allocating a national budget involves value judgments,  how to weigh education vs. healthcare vs. defense. An algorithm can crunch numbers to maximize output, but it cannot tell us whose priorities should come first or what fairness means in budgetary terms.

“Algorithms can optimise efficiency, but they can’t decide between competing values the very choices at the heart of democratic politics,” tech leader Eric Schmidt recently warned .

Another problem is the lack of transparency and accountability. If an AI like Diella decides all procurement, on what basis is it choosing winners? Without clear public rules and the ability to audit or challenge AI decisions, citizens are left in the dark.

When people don’t understand how decisions that affect them are made, they will feel disenfranchised and powerless. The result can be alienation and anger  the same problems that corrupt or unresponsive democratic institutions create, but with even less accountability.

In a democracy, a bad decision can be appealed or a leader voted out. But who do you blame when an algorithm makes a bad call? An opaque AI governance system might actually deepen public distrust, not alleviate it.

There is also the broader context of how algorithms operate in society today. Far from being neutral, many of the AI-driven systems we interact with (like social media algorithms) are designed with profit motives that exploit human weaknesses. Outrage and division generate engagement, so automated content feeds often amplify the most extreme and polarising material to keep us clicking.

Facebook’s own internal research showed that its news feed algorithm “promotes divisive content because outrage drives clicks”. In general, anger is the emotion that social media algorithms reward most , leading to echo chambers and heightened polarisation. If we hand more governmental functions to algorithms without safeguards, we risk importing these conflict-amplifying dynamics into public life.

An AI tasked with, say, allocating social benefits might make choices that optimise some metric but inadvertently reinforce biases or fuel social divides  and do so in a black-box manner. In short, algorithmic governance could substitute one kind of democratic deficit (unresponsive leaders) with another (unaccountable machines). With human dignity and agency treated as afterthoughts, polarization would only deepen and trust in the system erode further .

A Closing Window – and a Choice to Make

The window of opportunity to set this positive course is finite. With each passing month that traditional democracies continue to flounder mired in gridlock, corruption, or inefficiency the appeal of quick algorithmic solutions will only grow.

If people see no improvements in governance but do see AI systems getting smarter, there is a real risk that “algocracy” gains traction by default. Authoritarian-leaning leaders or struggling governments might seize on AI as a way to centralize power and claim technocratic legitimacy (“the computer decides, not me”).

Indeed, countries where institutions are weak or corruption is rampant are most likely to attempt an Albania-style AI power grab, hoping to bypass messy politics. Whether Diella ultimately succeeds in scrubbing corruption from Albania’s procurement or ends up adding another layer of opacity remains to be seen . But either way, the experiment has fired the starting gun  and imitation is likely in places desperate for a governance miracle cure.

This puts a spotlight on democratic nations like the United States and those in Europe: Will they demonstrate how to integrate AI into accountable, human-centric governance, or will they let the algocratic narrative run unchecked?

In truth, the future of democracy may depend on proving that we can have the best of both worlds,  a government that is both of the people and as competent and data-informed as the smartest algorithm. The choice we face is not AI or democracy. It is how we marry AI with democracy.

As one Albanian critic of Diella implied, simply handing power to an algorithm without transparency is a false solution. But as Taiwan’s example shows, using algorithms to empower more citizens in decision-making can renew democracy’s promise.

Conclusion: AI as a Democratic Ally, Not a Replacement

The coming years will test whether free societies can adapt and improve using AI, or whether they will be tempted to outsource governance to machines.

The lesson from Albania’s bold gamble and Taiwan’s digital innovations is that we shouldn’t abandon democratic principles for efficiency’s sake. Democracy’s problems from corruption to polarisation won’t be solved by ceding authority to code. Instead, we should be asking how AI can help us reimagine civic participation, increase transparency, and hold leaders (and algorithms) accountable.

Done right, AI could revolutionize 21st-century democracy  making it more inclusive, agile, and informed. Imagine town hall meetings supplemented by real-time AI analysis of every attendee’s feedback, or legislatures where AI briefings ensure representatives truly grasp public sentiment and policy impacts before they vote. These advances won’t come automatically; they require deliberate design and oversight to keep AI as the servant, not the master.

As a recent New York Times opinion put it, “The future of democracy doesn’t require us to reject AI. Quite the opposite. We need AI to make democracy work for the 21st century but we must be careful not to ask AI to decide for us, rather to help us govern ourselves better.”

In short, the goal is to reinvigorate democracy with AI, not replace it. The world’s first AI minister has taken office  now it’s up to the rest of us to ensure that “We the People” remain in charge, with AI as a powerful tool of our collective will, rather than a substitute for it.

 

Written by

Shape the conversation

Do you have anything to add to this story? Any ideas for interviews or angles we should explore? Let us know if you’d like to write a follow-up, a counterpoint, or share a similar story.

Write a counterargument article Write a related article