A coalition of major Wall Street firms is pushing for a $7.5 trillion investment in artificial intelligence infrastructure over the next five years, according to a proposal circulating among financial institutions. The plan would fund a sweeping buildout of data centers, semiconductor fabrication plants, and energy systems needed to support the next generation of AI models.
The scale of the ask
The $7.5 trillion figure dwarfs most previous infrastructure campaigns. To put it in context, that's roughly the combined annual GDP of Japan and Germany. The money would be raised through a mix of private capital, debt issuance, and public-private partnerships, the document shows. The firms argue that without this level of spending, the U.S. risks falling behind in the global AI race.
What the money would buy
The bulk of the spending would go toward building hyperscale data centers — facilities that can house hundreds of thousands of specialized AI chips. Those chips, mostly graphics processing units from companies like Nvidia and AMD, are in short supply and consume enormous amounts of electricity. The proposal also allocates funds for upgrading the power grid and building new natural gas or nuclear plants to keep the lights on. Another chunk would go to advanced manufacturing for the chips themselves.
Who's behind the push
The proposal was drafted by a group of investment banks, asset managers, and private equity firms that collectively manage trillions of dollars. They've been in talks with technology companies and energy providers to gauge interest. The document stresses that the buildout must be coordinated across sectors to avoid bottlenecks. It also warns that delays in permitting and construction could derail the timeline.
Why now
Demand for AI computing power has exploded since the release of ChatGPT in late 2022. Companies like Microsoft, Google, and Amazon are already spending billions on data centers, but the Wall Street group argues that the current pace is insufficient. They point to projections that AI workloads will grow 10x or more in the next few years, requiring a massive upfront investment in physical infrastructure. The proposal also cites national security concerns, noting that China is investing heavily in its own AI supply chain.
The proposal is still in its early stages. The firms plan to present it to regulators and lawmakers in Washington over the coming months, but they haven't set a formal deadline for securing commitments. One unresolved question is how much of the $7.5 trillion will come from public sources versus private markets. Another is whether the energy grid can handle the load. For now, the Wall Street group is focused on building consensus among its members and potential partners.
