On Friday, US Representatives Greg Casar (Texas) and Maxine Waters (California) introduced a House resolution urging Congress to deny further funding to AI companies and financial institutions facing insolvency or financial collapse due to heavy AI investments.

In the resolution, Rep. Casar who serves as Congressional Progressive Caucus (CPC) Chair and Waters from House Financial Services Committee, have invoked the 2008 Global financial crisis and said that the “AI bubble” is set for a similar course as “artificial intelligence-related firms hold trillions of dollars in debt, much of which is raised in opaque and unregulated private credit markets.”

Senator Bernie Sanders also criticized the mounting AI debt by major U.S. Tech companies and referred to it as “high-stakes gamble.”

Companies like Microsoft, Meta, Google, Amazon, SpaceX, Oracle & Nvidia have racked up $3 trillion in debt on a high-stakes gamble on AI.

Big Tech Oligarchs: If your AI bet fails, don’t come begging to taxpayers. You won’t be bailed out — period. End of Discussion.

— Sen. Bernie Sanders (@SenSanders) October 30, 2026

“Companies like Microsoft, Meta, Google, Amazon, SpaceX, Oracle & Nvidia have racked up $3 trillion in debt on a high-stakes gamble on AI.  Big Tech Oligarchs: If your AI bet fails, don’t come begging to taxpayers. You won’t be bailed out — period. End of Discussion, ” posted Sanders on X.

According to a latest Bloomberg report, five U.S. companies Alphabet Inc (Google), Meta, Microsoft, Amazon and Oracle have committed more than $3 trillion in total, a mix of short-term spending and future pledges, in their race to build “fleets of data centres.”

So, who pays this money? In this article, we will examine the rising debt of AI investments, why the “gold rush risk” should concern us and whether we are heading towards an AI bubble burst.

The $3 Trillion AI Debt: What it means?

Multiple market reports have warned us that the AI boom in the U.S. is being financed by borrowed money even as tech giants commit hundreds of billions of dollars to build data centres, buying advanced chips and expanding computing infrastructure.

The money is coming from institutional bond investors, banks, private credit funds, insurance companies and other financial institutions. These lenders expect repayment through future revenues from cloud computing, AI services and long-term infrastructure contracts.

The scale of commitments is already substantial. Microsoft, Meta, Oracle, Amazon and Alphabet have collectively committed approximately $1.09 trillion in future data-centre lease payments, according to a Reuters analysis published in August 2026.

Separately, AI cloud provider CoreWeave has accumulated around $50 billion in debt. Google alone has made commitments upto $1 trillion in AI investments followed by Meta at $800 billion in investments, in the form of technical equipment, leases and energy.

It’s crucial to understand here that the $1.09 trillion figure represents future lease commitments and not conventional outstanding debt.

Why are House Democrats Introducing Resolution Against Mounting AI Debt?

In their resolution introduced in the house of representatives, Casar and Waters invoked the 2008 global recession crisis stating that back then Congress had appropriated hundreds of billions of dollars to financial institutions responsible for propelling the housing bubble.

“ financing terms of many artificial intelligence-related data centers as well as the costs and revenues of major artificial intelligence firms are not publicly available and artificial intelligence-related corporate borrowing to fund data center investments are multiples higher than projected revenues,” read an excerpt from the resolution.

The representatives said that Congress should oppose “any measure to rescue insolvent artificial intelligence-related firms, or to shield the senior executives of such firms from accountability for reckless conduct that contributed to the failure of such firms.”

The representatives warned that an AI bubble burst can have severe repercussions like the 2008 global financial crisis as the “AI bubble pulls an increasing volume of resources from other sectors of the economy, diverting investment from other productive uses.”

The representatives also criticized the Trump administration’s decision to appoint David Sacks to the Chair of the President’s Council of Advisors on Science and Technology, and accused him of conflict of interest.

“White House’s artificial intelligence and cryptocurrency czar was personally invested in at least 449 companies with ties to artificial intelligence while pushing to eliminate government obstacles facing artificial intelligence companies, reflecting the administration’s broad deregulatory focus and conflicts of interest related to artificial intelligence.”

So, Who Clears the $3 Trillion AI Debt?

In a latest report by Boston Consultancy Group (BCG), it is argued that $2.6 trillion annual bill for AI infrastructure by 2030 can be paid for, but the money will ultimately have to come from businesses, consumers and technology companies.

In its October 29 report, “Who Will Pay the $2.6 Trillion AI Infrastructure Bill?” BCG estimates that the global AI ecosystem, excluding China, will need approximately $2.6 trillion in annual spending by 2030 to make the infrastructure expansion economically sustainable. This includes the costs of data centres, advanced chips, electricity, model development and returns for investors.

According to BCG report, enterprise customers are expected to shoulder the largest share, contributing $1.8 trillion, or nearly 69% of the total. BCG estimates that companies will fund this through a combination of reallocating existing IT budgets towards AI and additional spending.

Similarly, consumers are expected to contribute another $366 billion, primarily through advertising, e-commerce and direct subscriptions to AI products. Direct consumer subscriptions are expected to account for a relatively small portion, with the report estimating a potential $37 billion market for paid AI applications. Meanwhile, hyperscalers are expected to contribute $204 billion through internal AI usage, while emerging technologies, including robotics and physical AI, could generate another $200 billion.

As US lawmakers begin questioning the possibility of another corporate bailout, the AI industry faces a difficult test: it must demonstrate that the economic value generated by artificial intelligence can eventually justify the enormous financial commitments being made today.

Also Read: AI Spending Faces a Reality Check as Companies Tighten Budgets and Ration Tokens