What It Takes to Put a New Institution on Davos Map

What Davos taught me about legitimacy

Legitimacy, in high-stakes institutional spaces, is not something you arrive with. It is something you construct — moment by moment, room by room — through decisions that are mostly invisible to anyone watching the final result.

I was reminded of this in January, when I helped launch the first GovTech Day at the World Economic Forum Annual Meeting. On paper, it looked like an event. In practice, it was something harder: an attempt to bring a relatively young policy agenda into one of the most competitive attention economies in global governance. And what the experience made clear is that the conventional playbook — show up with a strong message, good speakers, and a polished agenda — is necessary but nowhere near sufficient.

Here is what actually matters.

The shift underneath the event

The easiest mistake in environments like Davos is to start with format. A summit. A day. A platform. But formats don't travel far there. What travels is a shift — a reframing that makes people feel like they are suddenly seeing something they had been looking at the wrong way.

For GovTech Day, that shift was this: governments are not simply digitising services anymore. They are beginning — often anxiously, often without quite admitting it — to rethink what the state does in an age of AI. The tension crystallised around what some are starting to call the "agentic state": systems where government doesn't just support decision-making with technology, but executes routine administrative processes through AI-driven workflows. That is not a technology question. It is a question about accountability, about where authority sits when there is no human in the loop.

Once you name it that way, the conversation changes register entirely. You stop talking to technologists. You start talking to people who think about power.

Credibility is assembled, not carried

No one walks into Davos with credibility intact. You borrow it before you earn it — and the borrowing is deliberate.

That meant being forensic about who was in the room and, just as importantly, who was seen to be there together. Ministers and senior officials from Singapore, Estonia, and the UAE. Global technology players alongside multilateral voices. Individually, none of this is unusual. But together, it creates something more important than a speaker list: a signal that this is already a shared agenda among serious actors, not an emerging conversation looking for its first believers.

There is something almost paradoxical about this. You construct the appearance of an established consensus in order to create that consensus. But it is not dishonest — it is how institutional change actually moves. The co-presence is itself evidence. When a Singaporean and an Estonian minister are willing to share the same framing on the same stage, that tells you something real about where the field is heading.

The most underrated asset is proof

There is a point in every policy conversation at this level where theory stops working. It comes faster than you expect.

This is where Ukraine's Diia ecosystem became central — not as a case study, but as a live system. Tens of millions of citizens, real services, functioning under real pressure. In a room full of people debating what AI-enabled government could look like, that kind of proof does something no framework can: it closes off the "has anyone actually done this?" escape hatch. It moves the conversation from conditional to indicative. From could to has.

The lesson extends beyond GovTech. In high-level institutional settings, the most underrated strategic asset is a concrete example that makes abstraction uncomfortable. Not a pilot. Not a roadmap. A thing that exists and works.

Communications is not the last step — it is the connective tissue

Most organisations treat communications as a downstream function: strategy first, then messaging. That model falls apart in multi-stakeholder environments moving at speed.

What held GovTech Day together was not the public programme. It was the alignment work behind it — between speakers, between institutions, between national contexts that had subtly different framings of the same agenda. In environments like Davos, inconsistency is structurally expensive. A fractured narrative signals that the coalition behind an idea is thinner than it appears. And that signal, once it gets out, is almost impossible to walk back.

Strategic communications, at its most demanding, is not about messaging. It is about maintaining a shared interpretive frame across actors with different interests, different vocabularies, and different audiences at home. It is infrastructure, not decoration.

The shift that is still looking for its language

Running underneath all of this was something broader that nobody quite named directly.

We are moving out of the era of government procurement and into something closer to long-term co-creation — governments and companies building, iterating, and maintaining systems together over time. You could hear it in how organisations like Capgemini, IBM, and Microsoft described their roles: not suppliers, but partners embedded in delivery ecosystems.

But this shift carries real tensions. As collaboration deepens, so do questions about accountability, sovereignty, and control. Who is responsible when a jointly-built system fails? Who owns the infrastructure the state comes to depend on? These are not abstract questions. They are the questions that will define the next generation of public-private relationships — and they do not have clean answers yet.

What I took from Davos is that the most important work happening in global governance right now is not in the formal sessions. It is in the effort to build the shared language, the shared trust, and the shared proof of concept that makes hard institutional change feel — to the people who have to authorise it — like a reasonable next step rather than a leap into the unknown.

That is slow, unglamorous work. It is also, I think, the only kind that lasts.