Anubhav Sachar
Skip to content
Navigate
HomeAboutAll InsightsContact
Insights — eight domains
Artificial Intelligence12 topicsGeopolitics5 topicsDigital Economies5 topicsInnovation & Technology5 topicsLeadership & Strategy5 topicsMy Journey3 topicsFuture Horizons3 topicsGlobal Business Strategies6 topics
Ventures
World AcceleratorVisit site Global Development 50Visit site
Connect for partnerships
Digital Soft Power

Digital Soft Power Is Won in Standards Committees

The unglamorous institutional work of setting technical standards determines more geopolitical outcomes than most summits. A case for treating standards bodies as strategic terrain.

Where the decisions are actually made

A technical standard is a decision about how a technology will work everywhere it is adopted. It fixes interfaces, defines what counts as compliant, and determines whose existing products need modification and whose do not.

These decisions are made in working groups by engineers, over years, in meetings that receive no coverage. The states that take this seriously send consistent, well-briefed delegations for a decade. The states that do not send whoever is available, and then discover that the rules of a market they intended to enter were written without them.

Why it compounds

Standards influence is unusually cumulative. Chairing a working group provides agenda control. Agenda control shapes which proposals receive serious consideration. Serious consideration produces adopted text. Adopted text becomes the baseline that the next revision starts from.

A state that invests consistently accrues a position that cannot be bought quickly. A state that arrives late finds that the relevant questions were settled two revisions ago and that reopening them requires a coalition it has not built.

The AI dimension

This is now playing out in AI governance. Evaluation methodologies, model documentation requirements, safety-testing protocols and interoperability specifications are all being defined at present, largely in technical fora.

Whoever defines what an acceptable evaluation looks like has substantial influence over which systems can be deployed in regulated sectors — which is a more consequential form of influence than most declaratory AI policy achieves. It is also considerably cheaper to contest, which makes it an unusually good investment for middle powers.

What effective participation costs

Less than most states assume. A credible standards presence requires a small number of technically strong people, funded continuously, with the mandate to build relationships over years and the institutional memory to maintain positions across changes of government.

The binding constraint is rarely money. It is continuity — the willingness to fund something whose returns arrive after the ministers who authorised it have moved on. That is a governance problem before it is a budget problem, and it is solvable by any state that decides to solve it.