“AI Data Centers” or Bitcoin Mines? Critics Say Utilities Are Rebranding to Tap Taxpayer Cash
By [James P. Hard] — October 17, 2025
Across the United States, a growing number of “AI data center” projects are drawing scrutiny from analysts and local watchdogs who claim the label is being used as a cover for large-scale Bitcoin mining operations — with electric utilities and developers seeking to justify billions in taxpayer-backed infrastructure.
Over the past year, energy-hungry data centers have been pitched as the backbone of America’s artificial intelligence boom. Companies and utilities are promising job creation, technological advancement, and regional competitiveness. But behind the AI rhetoric, critics say, many of these projects look strikingly similar to crypto-mining facilities: massive arrays of high-density servers running nonstop, drawing enormous amounts of power while creating few long-term jobs.
“It’s a rebranding exercise,” said one former utility executive who requested anonymity. “Bitcoin mining has a bad reputation — politically, environmentally, and financially. But call it an AI data center, and suddenly it’s an ‘innovation hub’ that deserves subsidies.”
Power-Hungry Promises
Electric companies in several states have lobbied regulators to expand generation capacity and build new substations to serve what they describe as “AI-driven computing clusters.” These requests often come with proposals for taxpayer incentives, discounted energy rates, or special financing.
But independent energy economists note that many of these proposed facilities are being built by or leased to companies with long histories in cryptocurrency mining. In some cases, public filings show that the firms’ existing operations consist almost entirely of Bitcoin-mining rigs, not AI or cloud-computing hardware.
A Convenient Rebrand
For utilities, the shift in terminology has strategic value. Expanding grid capacity for crypto mines is politically risky — but promoting “AI growth infrastructure” plays well with both local officials and federal policymakers seeking to boost U.S. competitiveness against China’s AI push.
“Everyone wants to be seen as supporting the next big thing,” said energy policy analyst Marla Chen. “Calling it AI makes it sound futuristic and socially useful. The reality is, most of these facilities are just power-intensive server barns doing computational hashing.”
Taxpayer Exposure
Critics warn that taxpayers could end up subsidizing the costs of private digital-asset speculation. In several states, utilities have asked ratepayers to fund transmission upgrades or offer “economic development” incentives on the assumption that AI data centers will create thousands of jobs — numbers that rarely materialize in practice.
Instead, most sites employ only a handful of technicians once operational, while consuming electricity equivalent to small cities.
The Blurred Line Between AI and Mining
Industry insiders admit that the line between AI computing and crypto mining can be blurry. Both depend on dense arrays of chips, high cooling demands, and stable power. Some mining firms are now experimenting with mixed-use models — operating Bitcoin rigs during off-peak hours and leasing GPU clusters for AI workloads when demand arises.
But experts caution that these hybrid claims are often overstated. “The hardware and networking required for genuine AI inference or training is completely different from ASIC-based mining systems,” said David Konrad, a data-infrastructure consultant. “You can’t just flip a switch.”
A Call for Transparency
As the line between digital-asset mining and AI compute infrastructure continues to blur, regulators and taxpayers are being urged to demand clearer disclosures — including hardware specifications, workload breakdowns, and power-usage audits.
“AI has become the new buzzword that unlocks government money,” said Chen. “If we’re not careful, we’ll end up underwriting speculative crypto operations that do nothing for the real economy.”