Big Changes in the World: Davos 2026

by Sandy Pentland

Alex “Sandy” Pentland is a HAI Center Fellow and faculty lead for digital society at Stanford HAI and the Stanford Digital Economy Lab. Reid Hoffman, co-founder of LinkedIn, described Sandy’s new book Shared Wisdom: Cultural Evolution in the Age of AI as a “must-read for anyone who wants AI to strengthen human insight, not override it.”

Here, Sandy shares his insights and opinions from the 56th World Economic Forum held in Davos, Switzerland from January 19-23. 2026.

At the end of World War II America was the clear economic and military superpower and it imposed the idea of a rule-based world order implemented by international organizations like the UN. But, as Canadian Prime Minister Carney said, we all knew it was at least partially a lie, and that the powerful could always claim exemption when it suited them.

And then came Donald Trump, who declared most of this world order to be a fraud. Privately, and with surprising uniformity, senior political leadership (Ministers of State, etc.) that I have talked with say that while Trump is unpredictable (and more), his actions have made them realize that they have lost their national sovereignty. They are no longer in charge of their digital infrastructure, their financial assets, or their military defense. Consequently, Carney’s call for a new world alliance of mid-sized nations that can compete head-to-head with America and China is appealing to many leaders.

A prime example of this newly emerging world order is the free trade agreement between the EU and India. After languishing for 17 years, it has suddenly been revived, creating an open market of almost 2 billion people and nearly one-quarter of world GDP. Incidentally, it is rumored that India spent $100 million for this year’s Davos meeting, in part to promote their open-source Citizens Stack public digital infrastructure, modeled after their very successful India Stack. Citizen Stack is claimed to have already signed up to a dozen other counties, providing a standardized protocol for digital identity, payments, medical services, and contracts. AI using this data will be formidable, and will act as a general productivity accelerant.

How are nations responding to AI and political decentralization?

Most smaller countries hope that AI can help them deal with all the political change and allow them to build thriving economies. However, they are not looking to the big frontier models of American hyperscalers, now that it is dramatically cheaper and easier to create specialized AI agents they are looking to create more focused “sovereign” AI agents for their nations’ trade, banking, government processes, and citizen services.  They are looking at data as a productivity factor, and are mindful that to limit AI power, you have to limit its access to proprietary data.

To address these concerns, India’s Citizen Stack is intended to be customized by each participating country to suit their particular needs, forming an interoperable digital trade area for AI agents. Similarly, China is dramatically increasing its investment into the Belt and Road trade alliance, specializing their AI tools for each application, thus creating another interoperable trade area for AI agents.

Many companies, including those in the finance, health, and trade industries, are also following this path. Rather than uniform services across the entire world, they are deploying systems customized for local conditions and allow users to leverage proprietary data, but are also interoperable to support international trade.

Economic structure in the age of AI: Nation-scale businesses

Many nations are looking for a practical vision of how AI might help deal with all this change and lead to a thriving economy. They recognize that the core competence of AI technology is to allow people to coordinate large amounts of information and enable large number of complex transactions, and they hope that AI’s capabilities may help their nations’ systems become far more sophisticated and efficient.

As described in the recent Stanford HAI book Flash Teams or my recent book Shared Wisdom, AI tools already allow small teams of people can set up and run large, complex organizations. Consequently, CEOs and C-level managers that I talk to expect to have smaller core, in-house employee teams, but (perhaps surprisingly) they also expect to have many more personnel conducting localized project work across the globe, much like many of the big Indian software companies do today.

This suggests that we might be headed toward a world of closely held regional data and distributed, less centralized companies and startups building and selling specialized and localized services. What might this world look like? Perhaps like the churning, entrepreneurial environment of San Francisco today: government statistics say that SF has a working-age population of 600k people, 15k venture funded startups, and 90k small and medium size businesses (SMEs) including startups but also businesses like restaurants, stores, services, etc. Most citizens are part of a small specialized and localized business.

Income equality in the age of AI: AI sovereign wealth funds

A world of small, distributed, and changeable organizations may imperil social safety nets, which today are mostly funded through employment relationships. People suggest that we might need universal basic income (UBI) to help move to the continually churning economy that AI seems to portend. But raising taxes to pay for such universal income seems unlikely.

An interesting alternative is Universal Investment, where every company that is created deposits 10% of its founding shares in a Sovereign Wealth fund like the one that Alaska uses to spread funds from its oil wealth to all citizens. When companies are created, they are worth nothing, so giving up 10% of the valueless shares is not worth paying lawyers and lobbyists to avoid — especially if sharing the founding shares provides future tax certainty. If this scheme had been in force over the last 30 years, this 10% share of all new US companies would be worth more than $9 Trillion dollars and generate enough income to pay each citizen $3000 each year — similar to the Alaskan fund in its best year. If AI improves productivity as expected, the yearly payment could quickly rise to a very decent living wage for every citizen.