By PYMNTS | October 13, 2026
Goldman Sachs CEO David Solomon’s comments on artificial intelligence flagged a financing risk with implications beyond Wall Street. How will companies fund an infrastructure expansion that could require trillions of dollars, and which projects will justify the expense? The bank’s third-quarter earnings results reflected the strength of its stock market businesses, with equities revenue jumping 72% from a year earlier.
In his opening remarks Tuesday (Oct. 13), Solomon cited studies estimating that data center buildout in the United States could require more than $10 trillion through 2032, averaging roughly 3.6% of annual U.S. economic output. Client conversations indicate that demand for computing power exceeds supply, creating opportunities for Goldman to help arrange financing.
The opportunity comes with limits. Solomon said Goldman is being selective about the financing risks it takes and the capital it commits as projects become larger and more complicated. The eventual demand for computing capacity, and the money needed to supply it, remains uncertain.
During the analyst discussion, Solomon addressed whether financing could become a bottleneck for AI capital spending. Growing borrowing needs could put upward pressure on interest rates, affecting both potential returns and whether individual projects make financial sense, Solomon said.
“There will be a point in time when this recalibrates, resets for sure,” Solomon said. “And you know, when that does happen, of course there’ll be some bumps, but it’s very, very hard to predict that.”
We’d love to be your preferred source for news.
Please add us to your preferred sources list so our news, data and interviews show up in your feed. Thanks!
For now, the shortage of computing capacity points toward more construction and more financing. Solomon said he expects the expansion to draw money from public and private markets through borrowing and investment in company ownership.
Chief Financial Officer Denis Coleman said the biggest companies driving the expansion have substantial cash available from their businesses to help fund it.
Solomon said he is confident in the technology’s expansion when another analyst pressed him about constraints involving power, labor costs and financing. But he acknowledged that higher rates, geopolitical instability and local opposition to data centers could affect how quickly projects advance.
“But this is real, and it’s significant, and it’s not going away, in my opinion,” Solomon said.
Looking toward the fourth quarter, Solomon said technology investment and AI infrastructure should help support fundraising and deal activity, although he warned of potential setbacks. He described the underlying U.S. economy as solid, citing corporate earnings growth, 2.2% GDP growth and a resilient labor market. Globally, renewed Middle East tensions and disrupted energy flows raised oil prices and inflation concerns. Further interest rate volatility could weigh on activity, but Goldman continues to see strong client demand.
Solomon’s outlook suggested companies aren’t simply postponing decisions until conditions settle. Businesses are moving to modernize operations, acquire capabilities and direct money toward their most promising opportunities, supporting demand for advice and financing.
Goldman reported quarterly net revenue of $18.68 billion, up 23% from a year earlier but down 8% from the second quarter. Net earnings totaled $5.29 billion, with earnings per share rising to $17.01 from $12.25 a year earlier. Equities revenue reached $6.44 billion, while investment banking fees increased 21% to $3.22 billion, supported by completed mergers and acquisitions and stock offerings.
For all PYMNTS AI coverage, subscribe to the daily AI newsletter.