The AI boom is consuming energy at a staggering rate, forcing a political and economic reckoning. Tucker Carlson notes a proposed data center in rural Utah would draw nine gigawatts - more than twice Utah's total current consumption. He argues this demand is driving a reversal of climate-first energy policy, with elected officials and financiers now pushing for massive fossil fuel expansion.
Kevin O'Leary frames this buildout as a national security imperative, comparing it to a nuclear arms race against China, which is constructing 400 gigawatts of coal power specifically for AI. Carlson counters that beating China by adopting its surveillance-state methods is a hollow victory, risking the erosion of democratic consent as projects are greenlit by obscure military boards without public benefit.
The energy crisis is reshaping the Bitcoin mining industry. Harry Sudock explains that AI and Bitcoin have different energy profiles: AI needs stable, high-quality power near cities with fiber, while Bitcoin mining can thrive on stranded, intermittent power at the grid's edge. CleanSpark is using mining as a bridge strategy, securing land and power contracts to monetize electricity immediately while waiting the years required to build out fiber and attract AI data center tenants.
This positions miners as infrastructure developers. Rory Murray details CleanSpark's financial engine: selling short-dated covered calls against daily Bitcoin production to generate operational cash, creating a recursive loop where an appreciating asset funds expansion without equity dilution. Institutional lenders are recognizing Bitcoin's safety as collateral, with loan rates compressing from 10% to roughly 6% as they realize its 24/7 liquidity allows for near-automatic liquidation.
“AI needs mega-campuses near cities with high-speed fiber. Bitcoin is the 'cockroach' of compute, thriving at the edges of the grid using Starlink.”
- Harry Sudock, What Bitcoin Did
Behind the AI spending frenzy, the bond market is screaming. Simon Dixon points to the 30-year Treasury yield breaching 5%, a stress level not seen since 2007, driving mortgage rates to 7%. He argues the AI trade is the only thing propping up the US economy, with massive data center projects consuming liquidity and electricity while the underlying financial structure deteriorates.
Neil Dutta from Forward Guidance confirms the precarious dependence. He says non-residential construction, including data centers, is now a major driver of employment growth. This capital expenditure fuels a financial accelerator loop: AI spending boosts corporate earnings, which lifts stock prices, which juices consumer spending through a wealth effect. However, real disposable income growth is anemic, meaning the entire structure relies on continued hyperscaler investment.
The United States is attempting to build this new infrastructure with a severe handicap. Turner Caldwell of Mariana Minerals states the US is 50 years behind China in critical mineral supply chains. Drew Baglino of Heron Power adds that the grid itself relies on pre-World War II mechanical systems, creating a fragile, overbuilt network. Their solution is vertical integration and extreme automation - treating mining and grid modernization as software problems to bypass labor shortages and sluggish permitting.
“The U.S. is 50 years behind China on critical minerals supply, a gap that persists even if permitting and finance accelerate.”
- Turner Caldwell, The a16z Show
The societal implications are profound. Carlson warns that replacing intellectual labor with AI threatens the human drive to create, potentially leading to a crisis of purpose beyond financial hardship. This Week in AI host notes a cynical training-layoff loop emerging, where employees at firms like Cloudflare train AI models that eventually automate their roles, creating a 'purgatory' for mid-managers and developers.
Derek Thompson sees a consolidation of power among a new billionaire class. He warns that the revenue growth at companies like Anthropic - which reached a $450 billion annualized run rate shortly after a $30 billion valuation - is not a bubble. The resulting concentration of wealth could allow billionaires to account for half of all political spending, demanding a new minimum basic tax to prevent democratic capture.
The scramble for energy to power AI is dismantling old political orthodoxy, creating new financial engines, and exposing structural weaknesses in the American economy and grid. It is a race not just for compute supremacy, but for the foundational resources - power and minerals - that will determine who controls the future.
