The courtship of AI companies has ended: initiatives to impose a direct tax on data centers have begun
As AI company valuations soar, concerns over environmental impact and lack of job creation have prompted US states to impose new taxes and construction bans on data centers.

The value of companies in the AI field continues to break records, and in the United States, there are concerns that enormous profits will remain in the hands of a few, while the public is left with environmental and economic damages: high energy consumption and unemployment. These concerns have turned AI policy into a top priority ahead of the November midterm elections.
Virginia has become the most prominent example of this trend. In late June, the state made a precedent-setting decision to become the first to impose a direct tax on data centers based on their electricity consumption. For tech giants like Amazon and Microsoft, this is a relatively negligible amount compared to their future profits, but for the state budget, it represents a significant addition of up to $600 million in the coming year.
"How do we let the richest companies in human history come here and not play by the same rules as everyone else?" asked Democratic Senator Louise Lucas, who initiated the legislation. She argues that the rapid boom of AI entrepreneurs places a heavy burden on local infrastructure, and the industry must share the fruits of its growth with the public.
Virginia is not alone. According to a report by Aterio, the number of data centers under construction in the US has jumped from about 100 in early 2021 to nearly 1,000 this year. Nebraska and Ohio have stopped granting generous tax incentives, and New York Governor Kathy Hochul was the first to impose a general ban on data center construction for the coming year. Arizona followed suit to allow authorities time to examine the environmental and budgetary implications.
States and cities previously courted tech giants like Amazon, Meta, Alphabet, and Microsoft with tax breaks, expecting economic prosperity. However, these expectations have not always materialized. Data centers serving the AI industry require less manpower but consume vast amounts of water and electricity. "What is the return on investment if no jobs are added?" Hochul wondered last month.
At the national level, Senator Bernie Sanders has proposed that the government hold a 50% stake in large AI companies via a one-time tax paid in shares to a new sovereign wealth fund. While unlikely to materialize soon, the Financial Times reported that the Trump administration has explored options to share AI revolution benefits with the public, including talks with OpenAI about transferring a 5% stake to the government. Experts like Ben Harris of the Brookings Institution have criticized this, warning of potential exploitation and conflicts of interest.
Meanwhile, Malaysia has successfully leveraged AI demand for economic growth, reporting a 6% annual growth in the second quarter, driven by a 7.5% expansion in the manufacturing sector—largely dedicated to chip production—and a 6.6% expansion in construction supported by data center development.





