It sounds like something out of a Dystopian Philip K. Dick short story: A national government run on artificial intelligence (AI) and algorithms.
But it is happening to large degree in Albania, with decidedly Big Brother results.
This sci-fi yarn started in September, when the European country appointed an AI system to a cabinet-level positon, igniting fresh debate over the role AI should play in government decision-making.
The system, named Diella (Albanian for “sun”) was designated “Minister of State for Artificial Intelligence” by Prime Minister Edi Rama and tasked with combating corruption in government contracting. The appointment immediately sparked controversy, with Albania’s opposition party dismissing the AI as a “propaganda fantasy” and mounting a legal challenge against its appointment.
Yet Albania’s bold experiment reflects a broader global trend of governments integrating AI into their operations to streamline bureaucracy, boost efficiency, and support decision-making processes.
“Governments need a minister for tech, and that includes AI,” says Jamie Dobson, founder of Container Solutions.
“IBM famously said that a computer cannot make a decision because it cannot be held to account. Diella undoubtably has a human operator behind it and in any case sounds nothing more like an automated system to help with procurement,” Dobson said in an email. “That human, when things go wrong will be held to account. With that been said, some humans are so devoid of courage that an ‘AI’ – or chicken – would make better decisions. So maybe the Albanian government are onto something.”
“By creating the world’s first AI minister, Albania is not merely embracing the future, but trying to do its part in designing it,” Rama said in a recent speech. “Diella is far from being a gimmick.”
The numbers support his view. By the end of 2024, U.S. federal agencies reported more than 1,700 AI use cases — including note-taking, document summarization, and regulatory comment review — more than double the number from the previous year. While many applications rely on traditional machine-learning techniques for basic tasks, generative AI systems that produce novel content are rapidly gaining ground. A July report from the Government Accountability Office (GAO) found genAI use cases increased ninefold across 11 federal agencies between 2023 and 2024.
President Trump’s AI Action Plan, released in July, identifies AI adoption as a key priority, stating that “transformative use of AI can help deliver the highly responsive government the American people expect and deserve.” The U.K. estimates that AI assistants could save taxpayers up to £45 billion by making the public sector more efficient.
Diella: Symbol or Substance?
Despite its title, Diella’s ministerial role is far more limited than the designation suggests. Prime Minister Rama elevated the system to signal his administration’s commitment to fighting corruption and embracing advanced technology as Albania seeks European Union membership.
The country has spent the past decade implementing aggressive anti-corruption measures, including vetting every judge and establishing specialized prosecution bodies. Public procurement, which accounts for roughly one-third of government spending globally, remains a prime target for reform.
“Diella never sleeps, she doesn’t need to be paid, she has no personal interests, she has no cousins—because cousins are a big issue in Albania,” Rama said, describing the AI system as an incorruptible watchdog.
But Diella has not yet been deployed in its ministerial capacity. When operational, it will assist at four stages of the contracting process: drafting terms of reference, specifying eligibility criteria, setting price limits, and verifying submitted documents. Crucially, human procurement experts will approve all AI recommendations, with all actions logged and monitored.
But critical questions remain about how the system will function in practice. When unsuccessful bidders contest awards, which is a common occurrence in government contracting, the chain of accountability becomes murky. Is the prime minister responsible? A human supervisor? The AI vendor? The government has yet to clarify who answers for the algorithm’s decisions.
Transparency advocates caution against viewing Diella, as the system is known, as a silver bullet for corruption. While officials tout the AI as impartial and efficient, humans still control what data feeds the system and the parameters that govern its decision-making. “An AI system can’t take a bribe,” one observer noted, “but people can still manipulate its outcomes.”
The system’s reliability, experts say, hinges not on the technology itself but on the governance framework surrounding it, and Albania has yet to disclose its planned framework.
For now, Diella remains more symbol than substance, and power stays firmly in human hands.
With AI capabilities advancing every few months and adoption accelerating among both citizens and officials worldwide, the question is no longer whether AI will play a larger role in government, but how quickly — and how responsibly — that transformation will unfold.
