The public sours on AI and data centers as Anthropic, OpenAI look to IPO and tech keeps spending

Artificial intelligence is losing the popularity contest in the United States, and that's bad news for some of Big Tech's power players.

The trend has implications for richly valued companies like OpenAI and Anthropic, which are both on the path to go public. It also threatens Amazon, Google, Microsoft and Meta, the so-called hyperscalers planning to spend hundreds of billions of dollars to compete for data centers to power AI.

The latest signal of negative sentiment was extreme.

OpenAI CEO Sam Altman was targeted at his San Francisco home last week, where prosecutors say a man threw a lit Molotov cocktail at his driveway gate. The San Francisco District Attorney said the crime was motivated by hate of AI technology.

The suspect, 20-year-old Daniel Moreno-Gama of Texas, faces charges including attempted murder. He is also accused of threatening to burn down OpenAI's headquarters.

Altman responded to the attack over the weekend, acknowledging a time of "great anxiety about AI" and called for a de-escalation of rhetoric and tactics.

"I empathize with anti-technology sentiments and clearly technology isn't always good for everyone," Altman wrote. "But overall, I believe technological progress can make the future unbelievably good, for your family and mine."

Altman recently floated several policy remedies to try and get ahead of any broader economic anxiety, ranging from a public wealth fund to a four-day workweek and changes to payroll taxes in favor of taxing automation.

Anthropic CEO Dario Amodei, however, is among those who have called out the risks of large-scale disruption from AI.

The conversation is now moving from niche tech circles to the public.

AI is already expected to become a central campaign issue during midterm elections, and polling data now suggests Americans are feeling more pessimistic about the technology.

A March NBC News survey found 57% of registered voters believe the risks of AI outweigh the benefits, and a Quinnipiac University poll reported that 55% expect AI would do more harm than good in their daily lives.

Those results echoed in a Pew poll that found a majority of Americans are more concerned than excited about increased use of AI.

Data centers and servers are the real brain power behind AI, and the tech megacaps have committed roughly $700 billion this year to build them out across the U.S. The more data centers and "compute" available, the smarter these AI models can become.

But the energy demand of data centers has already become a focus point of state and local campaigns, and has resulted in widespread pushback.

A report from Data Center Watch found that in 2025, at least $156 billion in data center projects were blocked or delayed amid local opposition and litigation. 

On Wednesday, Maine passed a bill to create the first state-wide data center ban, which is now heading to the governor's desk for signature. Voters in the St. Louis suburb of Lester, Missouri, voted out several city council members over support for a proposed data center.

Finally, the risk could spill over to impending public offerings.

OpenAI's valuation may be especially dependent on a data center buildout, which it has described as a strategic advantage. Political sentiment could also weigh on potential shareholders.

The startup is among the companies planning to reserve a portion of its IPO for individual investors, according to CFO Sarah Friar.

"Everybody wants to own part of a rocket company," Friar said, referring to SpaceX and its plans to hold almost 30% of its allocation for retail. "I hope everyone wants to own part of ChatGPT. It helps when you're a consumer brand."