Tallinn Digital Summit 2026 live: Estonia debates AI agents and the future of government

Estonia is using this week’s Tallinn Digital Summit, with selected sessions streamed live, to make the case for a new kind of government: one where AI agents do more than answer questions – they carry out public services under human oversight.

The idea is central to the two-day meeting, which brings political leaders, technologists, businesses and security experts to Estonia’s National Library on 5 and 6 November. Its question is how democracies can thrive in what organisers call the “agentic era”, as AI systems begin to act on people’s behalf.

The promise of an agentic state

An AI agent might file a form, arrange a service or complete a routine administrative task. In theory, that could mean less paperwork and fewer unnecessary interactions with government. But as public services become more autonomous, they could also become harder to scrutinise. Who is accountable when an AI agent makes a consequential decision? What happens when it misfires, or when citizens’ own agents overwhelm systems built for people?

Those questions have become more urgent as governments try to turn AI investment into practical results. The summit’s background materials note that, despite the technology’s much-discussed economic potential, fewer than five per cent of organisations – and fewer governments still – have seen substantial financial gains from AI.

Estonia is presenting itself as a place to test what comes next. Its Eesti.ai programme aims to use AI to double the value of Estonian labour by 2035, while growing the economy by 25 per cent within five years and 50 per cent within a decade. The AI Leap initiative, a public-private partnership, is bringing AI tools and technology education into schools.

The challenge is to make those ambitions work in real services while keeping people in control. Estonia’s foreign minister Margus Tsahkna argues that digital policy is now inseparable from national security. Russia’s war against Ukraine has shown how digital infrastructure can help a country maintain public services under attack, while cyber threats and AI-enabled disinformation expose new vulnerabilities.

Ukraine’s experience features prominently in the summit programme. A session on the “Agentic State” will examine the country’s Diia.AI service, designed to answer citizens’ questions and complete public services on their behalf. Another will explore projects developed at an Estonia–Ukraine hackathon, including tools for government software security and public-sector work.

The conference will also examine how to test AI before connecting it to the systems people depend on. A planned “Tallinn Stress Test” will use a cyber-range demonstration to assess how AI behaves when linked to data, identities and other tools. The point is to test the system in context, rather than assume that a model’s performance in isolation tells the whole story.

The case for global rules

Estonia’s justice and digital affairs minister Liisa Pakosta is making the case for international rules to match the technology’s global reach. In a contribution prepared for the summit, she argues that AI needs an international body with a role comparable to that of the International Atomic Energy Agency. Her proposals include shared standards for AI-generated content and a network for reporting emerging risks.

The agenda puts that proposal alongside a broader debate over whether regulation can keep pace with competition. Researchers and policymakers, including computer scientist Stuart Russell, are due to discuss safeguards, human oversight and the possibility of international standards. Other sessions will focus on Europe’s ability to retain technology companies, attract capital and build the infrastructure – from chips to computing power – needed to compete.

The race for digital sovereignty

That competition has a geopolitical edge. AI may give smaller countries new ways to build influence, but it also deepens reliance on global technology providers and concentrated supply chains. The summit’s debate over digital sovereignty asks whether governments can secure control of essential systems through national capacity, regional cooperation or trusted commercial partnerships.

Estonia’s digital state has long made it a case study for governments seeking to move services online. The test now is whether that model can extend to systems that make decisions and take action – without making public power less visible or less accountable.

The summit’s defining question is not simply how quickly governments can adopt AI. It is whether they can make the technology useful, secure and answerable to the people it serves.