While AI bubble talk fills the air today, with fears of overinvestment that might pop at any time, something of a contradiction is brewing on the bottom: Corporations like Google and OpenAI can barely construct infrastructure fast enough to fill their AI needs.
During an all-hands meeting earlier this month, Google’s AI infrastructure head Amin Vahdat told employees that the corporate must double its serving capability every six months to fulfill demand for artificial intelligence services, reports CNBC. Vahdat, a vp at Google Cloud, presented slides showing the corporate must scale “the following 1000x in 4-5 years.”
While a thousandfold increase in compute capability sounds ambitious by itself, Vahdat noted some key constraints: Google must give you the option to deliver this increase in capability, compute, and storage networking “for essentially the identical cost and increasingly, the identical power, the identical energy level,” he told employees through the meeting. “It won’t be easy but through collaboration and co-design, we’re going to get there.”
It’s unclear how much of this “demand” Google mentioned represents organic user interest in AI capabilities versus the corporate integrating AI features into existing services like Search, Gmail, and Workspace. But whether users are using the features voluntarily or not, Google isn’t the one tech company struggling to maintain up with a growing user base of shoppers using AI services.
Major tech firms are in a race to construct out data centers. Google competitor OpenAI is planning to construct six massive data centers across the US through its Stargate partnership project with SoftBank and Oracle, committing over $400 billion in the following three years to achieve nearly 7 gigawatts of capability. The corporate faces similar constraints serving its 800 million weekly ChatGPT users, with even paid subscribers usually hitting usage limits for features like video synthesis and simulated reasoning models.
“The competition in AI infrastructure is probably the most critical and in addition the costliest a part of the AI race,” Vahdat said on the meeting, in response to CNBC’s viewing of the presentation. The infrastructure executive explained that Google’s challenge goes beyond simply outspending competitors. “We’re going to spend rather a lot,” he said, but noted the true objective is constructing infrastructure that’s “more reliable, more performant and more scalable than what’s available anywhere else.”
