Sovereign AI And The Geopolitics Of Compute

Punnam Raju Manthena, Cofounder & CEO at Tekskills Inc. Partnering with clients across the globe in their digital transformation journeys.

​If you’re a business leader, your AI sovereignty is your capability to independently build, deploy and regulate your AI using your own infrastructure, data, models and workforce. A nation’s job is to strengthen its digital infrastructure, such as national payment systems, to avoid depending on foreign financial engines. Considering the geopolitics of compute, this may be akin to oil reserves. Soon, chips, GPUs and cloud infrastructure could be the powerful geopolitical assets that decide whether a country is economically dominant and secure, rather than its oil or gold reserves.​

In either case, businesses and countries get to work with chosen vendors, based on cost and capability considerations. This approach may be incomplete and “burn a hole in your pocket” in the long run. It may be better that you look at the technical fit, data governance and total spend as well.​

​What is AI sovereignty?

Generally, you might think that AI sovereignty is all about frontier AI models: the models at the forefront of your capability. That’s a big no. There are aspects, including jurisdictional access, data residency and vendor availability, that can hit deployments.

First, jurisdictional access means you juggle your deployments with legal compliance, complicating them. When you have your software and data across multiple countries, you need to comply with diverse and sometimes conflicting laws about who can access, audit or seize your data.

Data residency tells you where your drives and servers are located, while sovereignty tells you the laws that govern that data and whether there is any foreign government that may be able to legally access it.​

AI sovereignty can be an uncomfortable trade-off.

Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?