Who’s Handing Power Over to AI

An image illustrates a superintelligent AI cyborg making decisions over humans. Albania may be the first country to seat an algorithm as a c...

An image illustrates a superintelligent AI cyborg making decisions over humans.

Albania may be the first country to seat an algorithm as a cabinet minister, but it is far from alone in wiring artificial intelligence into the machinery of the state. The difference is that most nations are doing it quietly, in fragments, and behind thicker curtains.

AI Ministers 

In the United Arab Emirates, Omar Sultan Al Olama serves as Minister of State for Artificial Intelligence, tasked with redesigning the nation’s digital bureaucracy around machine learning. While he hasn’t handed decision-making to AI yet, he is building the infrastructure that could carry algorithmic governance in the future.

Spain has gone regulatory, creating AESIA, one of Europe’s first dedicated AI oversight agencies, to audit and license government algorithms. It is a legal skeleton meant to ensure accountability before machines move closer to decision-making authority.

AI Enforcement

Tax agencies are some of the earliest adopters of AI in governance. In the United States, the IRS uses AI to sift through hedge fund filings and detect hidden evasion schemes. Canada goes further, requiring agencies to file algorithmic impact reports before deploying new models. Spain is rolling out AI tools to catch fraud patterns in real time. Italy is testing machine learning to flag fake VAT claims and has built a chatbot to assist auditors. India is scaling up AI-driven crackdowns on phantom deductions, while Armenia is piloting systems to scan invoices for suspicious activity before human review.

France has pointed algorithms at aerial imagery to detect undeclared swimming pools, instantly boosting tax revenue. Latvia runs a tax chatbot named Toms, answering citizen queries since 2020 and scaling the reach of the bureaucracy.

Automation in Public Services

Romania has equipped its rural investment agency with robotic process automation and AI to pull documents from state databases and accelerate EU funding to farmers — not glamorous, but highly effective. Estonia is pushing forward with KrattAI, a vision where every citizen can interact with the state through a single voice interface. Denmark is preparing to roll AI tools across all public services. Singapore’s GovTech unit is building AI-powered applications for ministries, while South Korea and Japan are embedding AI into welfare, healthcare, and administrative services.

Nepal has adopted a National AI Policy, outlining a blueprint to modernize its bureaucracy with AI, laying down legal guardrails before deploying systems at scale. No algorithm holds decision-making authority yet, but the groundwork is ready.

From Tools to Authority

So far, these systems are assistants rather than rulers. They pre-fill forms, score risks, triage cases, and flag anomalies. They shape outcomes invisibly and constantly, but they stop short of writing laws or signing contracts. Albania’s Diella is different. She is tasked not only with advising but also with awarding — deciding who receives public funding. This crosses the line from AI as a tool to AI as authority.

Right now, no country has fully handed political power to algorithms. What we see is a two-track model: administrative AI running in the background, and experiments like Albania where AI steps into decision-making. For this to expand globally, laws would need to define accountability, regulators would need teeth like Spain’s AESIA, and models would have to prove resilient against manipulation, bias, or sabotage.

The image from popular culture about apocalyptic fear of AI, often referred to as the "AI doomsday".

Governments are also testing AI in defense and security. The Saudi Ministry of Defense has piloted AI for logistics and predictive maintenance of military equipment. Japan’s Ministry of Defense is trialing AI for cyber defense and satellite surveillance. Even the United States Department of Defense is exploring AI-driven decision support for battlefield simulations, though under strict human oversight.

In the field of healthcare governance, Britain’s NHS uses AI to triage patient calls through its NHS 111 service, while Singapore’s HealthHub integrates AI to predict patient needs and optimize hospital resources. These systems are not just administrative helpers — they are gradually shaping policy decisions by influencing how resources are allocated.

While Albania is leading with symbolic boldness, other nations are building the pipelines, safeguards, and cultural shifts that could allow AI to sit at the decision-making table. The quiet revolution is already happening: governments are turning into hybrid human-machine systems. What is missing is a unified framework for rights, accountability, and transparency that can keep pace with the speed of deployment.

One thing is certain: AI is no longer an experiment for government — it is the new operating system of the state. Whether it whispers from the background or takes a ministerial seat, AI governance is here, and growing stronger every day.