Cerebras CEO Andrew Feldman says the AI industry has done a poor job of selling data centers to the public. He has a different message in mind, modeled after Microsoft President Brad Smith.
“These can be clean, they can make jobs, they can be good for communities,” Feldman said during a recent episode of Harry Stebbing’s “20VC” podcast. “We can do this thoughtfully.”
Feldman, fresh off the chipmaker’s blockbuster IPO, said AI companies need to be better neighbors when approaching communities where they plan to build massive facilities to house the thousands of advanced chips needed to power AI models.
“There’s no reason why we can’t add these to communities and have the community benefit from it,” he said. “And we have to do some thinking, we have all the heavy equipment out there —build a football field for the local school, build a school, add a church or a synagogue to the community. We can be good neighbors at very, very low cost.”
Data centers also need to be better stewards of local resources, Feldman said, and AI and tech companies should pick up the tab, not taxpayers.
“In some cases, they tried to pawn off costs on the local community or use outdated financial arrangements that left the community holding the bag,” Feldman wrote in an email to Business Insider. “And in others they were wasteful of resources. This is not cool. And none of this needs to be the case.”
Feldman said one way companies can better serve their communities is by building closed-loop facilities that reduce water usage. Last June, Business Insider reported that 40% of the nation’s planned and existing data centers are in some of the US’s most water-stressed areas.
The Cerebras CEO cited Smith’s plan, “Building Community-First AI Infrastructure,” as a model for how the industry should be talking about its plan. Smith outlined five steps Microsoft would follow in developing its AI infrastructure, including paying its own way to ensure residents don’t face higher electricity prices, reducing its water consumption, creating jobs, and partnering with local nonprofits and universities on job training.
“Whether it was canals, railroads, the electrical grid, or the interstate highway system, each era produced its own conflicts over who bore the burdens of progress,” Smith wrote. “One enduring lesson is that successful infrastructure buildouts will only progress when communities feel that the gains outweigh the costs.”
Data centers aren’t the only thing hurting AI’s popularity.
Polling shows Americans are concerned about AI-related job displacement. A March Quinnipiac University poll found that 7 out of 10 Americans believe that advancements in AI will lead to a decrease in the number of job opportunities.
It’s why OpenAI CEO Sam Altman and others in the industry have chafed at the growing number of companies whose executives have cited AI as their motivation for recent layoffs. Like Altman, Feldman also sees “AI washing,” or blaming the unpopular technology for job cuts, as afoot.
“I think to date, most of the layoffs were ‘AI-washed,'” Feldman said. “They were because we did boneheaded hiring during COVID. It is actually because a great deal of productivity gains have occurred over the years that we’re just now harvesting.”
Feldman said that companies that don’t figure out how to harness AI-driven productivity gains will be outpaced. He said Cerebras wants to hire more engineers, not fewer.
“If you are an engineering organization that can’t see how to take advantage of vastly more productive engineers, I don’t think you’re long for this world,” he said. “I mean, the list of things I want our engineers to do is 50 times as much as we have engineers.”
