Jensen Huang Says Hyperscalers Get This ‘Almost Always Wrong’ — Nvidia Has a Workaround

The biggest AI infrastructure players may have a planning problem. NVIDIA Corp‘s (NASDAQ:NVDA) CEO Jensen Huang says hyperscalers are "always almost wrong" because they plan capacity once a year while AI demand is changing too quickly to fit an annual calendar.

His workaround? Build a much larger network of nimble regional clouds.

• NVIDIA shares are consolidating. What should traders watch with NVDA?

Hyperscalers Can’t Move Fast Enough

The unusual admission came during the All-In Summit, when the hosts pressed Huang on Nvidia’s support for neocloud companies and the need for more competition at the hyperscale layer.

Huang said Nvidia would be happy with five hyperscalers, but argued the industry ultimately needs 50, 100 or even 1,000 neoclouds. The reason is speed: hyperscalers typically plan their infrastructure once a year, while AI market dynamics are moving so rapidly that those forecasts can quickly become obsolete.

That creates an opening for regional cloud providers that can react faster to local demand and infrastructure constraints.

Nvidia’s Distributed AI Bet

Huang said regional clouds know their markets and can move quickly to secure land, power and existing facilities, something that is harder for companies operating from Silicon Valley or Seattle to coordinate globally. Nvidia is consequently building what Huang described as a large-scale distributed network of companies securing those critical resources.

The strategy is already moving beyond the U.S. Huang pointed to activity in Australia and Southeast Asia, saying Nvidia is helping bring additional gigawatts of capacity online through regional cloud partners.

That could matter enormously for Nvidia. The company does not necessarily need every AI workload to run through a handful of hyperscalers if it can help create hundreds of smaller customers that collectively consume massive amounts of Nvidia hardware.

It also reinforces a broader shift in the AI infrastructure trade. The bottleneck is increasingly not simply who has the best chips, but who can secure the electricity, land and data center capacity needed to deploy them.

For investors, Huang’s neocloud strategy offers a different way to think about Nvidia’s next growth phase: instead of betting on fewer giant AI customers, Nvidia is helping build an army of smaller ones.

And in an industry moving faster than an annual planning cycle, that flexibility could become a competitive advantage.

Image via Shutterstock