“A new landscape is emerging where smaller countries are gaining influence through targeted investment, strong partnerships, and focused policy,” says Vaishali Rastogi, global leader of BCG’s Technology, Media and Telecommunications practice.

New Regulatory Realities

Since 2023, legislative mentions of AI rose by 21.3 percent across 75 countries. Adoption largely depends on regulatory environments, as policy can bolster or dampen enterprise use of new tech. Policy also dictates the barriers between operations, production, and workflows across borders.

​​“There’s a risk when competition becomes siloed,” says Rastogi. “Without coordination, we may see a fractured AI landscape with incompatible standards, duplicated efforts, and uneven access.”

According to reports from the BCG Henderson Institute, around 40 percent of large companies have at least some of their tech teams in more than 10 countries. As regulation occurs in silos around the world, business decision-makers are suddenly faced with those challenges, including the unavailability (or banning) of certain technologies produced in specific countries. For example, Italy has banned all Chinese AI platform products. Russia, China, and Iran are restricted from receiving advanced chips from the US. These kinds of regulatory and practical subtleties between countries mean that a company could reasonably be faced with building different, isolated tech stacks in each country where they have a presence.

“Corporate leaders are in a very challenging situation,” Lang says.

“Today, when you do a vehicle navigation system, you have to be sure that if you use it in China, it uses digital maps and the data is being saved on Chinese cloud computers. If you use that navigation system in Europe, you need to make sure that the data is compliant with EU regulatory law. And if you want to use that navigation system in the US, you have to comply with the new ICTS regulation, starting on January 1, 2027—meaning that if you happen to have Chinese code in your navigation or your driver systems, after 2027, you can’t sell the car. So the reality is that if I’m a car manufacturer, I already have three different regions where I need to have three different tech stacks to satisfy regulations,” Lang continues. “And that’s just the beginning. You need to understand the realities. You need to be aware that the one-size-fits-all approach and the single global product will not work, and that you will need to adapt.”

Because of the new geopolitical realities of technology and AI, business is moving from an old paradigm that allowed companies to concentrate resources, production, teams, and supply chains where headquarters are managed to a new normal—where three or four of those global locations exist within different regulatory bodies.

Building Something That Lasts

Lang remembers a time not too long ago—only about a year and a half ago—when the question about AI was: Will it last? “Now I think everyone understands something really is coming,” he says.

The reaction within the business world has been a flurry of interest and toe-dipping into AI strategy, but also a lot of trepidation. The sentiment of “we want to do something, we just don’t know what,” reverberates through many a conference room.