TheChek News Flow

The Takeaway

With AI and big tech firms seeking to raise hundreds of billions in equity and debt this year to build data centers, any high-profile misstep could throw the broader AI boom off track. That includes the potential public debuts of SpaceX and possibly AI firms Anthropic and OpenAI. Poor performance from those initial public offerings could dampen broader enthusiasm for AI investment, according to a speaker at The Information’s AI Financing event in New York on Monday.

Ashley MacNeill, co-head of capital markets at Vista Equity Partners, warned that “there’s a ton of ways” the SpaceX IPO “can go awry” and “only a handful of ways this goes right.” She said that the unprecedented nature of the IPO—it is aiming to raise $75 billion, far more than any other IPO—meant it would confront issues previous public offerings had never encountered.

MacNeill pointed out for example, that the SpaceX offering will equate to almost 10% of daily equity trading volume, and it’s not clear the market can handle all that capital in “one shot,” let alone as huge gobs of capital start trading when lockup periods end.

“Three of the biggest IPOs…potentially in history [are] happening all within potentially the same calendar year,” said Alexa von Tobel, founder and managing partner of Inspired Capital. “The performance of some of those IPOs may actually be like real cold water on reality.…There’s a state of the world where it actually makes the capital around it more cautious.”

Monday’s event looked at the fast-growing need for capital in the AI market, including from credit markets and equity markets, and bankers and investors have had to come up with a variety of creative structures to raise the money.

Morgan Stanley’s global head of debt capital markets, Anish Shah, noted that the markets would need to finance between $300 billion and $400 billion in annual spending for AI. That will make up about 10% of the bond and equity capital markets, he said, marking explosive growth for a type of financing that was basically nonexistent a year ago.

Plenty of speakers were bullish about the potential of AI. Billionaire investor Glenn Hutchins, who is on the board of AI data center firm CoreWeave, said the AI boom is not a bubble but “one of the biggest changes in economic and human organization in history.”

“If you’re not an investor in that, if you’re not on the right side of that, both with your capital and with your time…you risk real obsolescence,” Hutchins added.

Another investor, Jon Redmond, a portfolio manager at Discovery Capital, also took an upbeat view, noting he was “excited” about the SpaceX IPO. He believes that once SpaceX’s IPO prospectus becomes public, which is expected in a few weeks’ time, “people are gonna really start to realize how big the opportunity is” for the company.

Still, Redmond acknowledged that the giant IPOs could affect other stocks as investors sell some investments to raise money to buy the newly public companies. He said historically large IPOs were followed by a market sell-off: “I think you’re gonna have something very similar. It will probably be tech led, is my guess—Mag seven led, if I had to take a wild guess.”

Speakers were divided about the future for enterprise software. Uncertainties about how AI will disrupt the industry have pummeled public stocks.

Hutchins predicted the software sell-off would continue as part of a broader reckoning with AI disruption that could play out across the economy: “A massive tsunami is about to hit the global economy and software firms are just the sunbathers on the beach that were first to be hit.”

Still, some businesses could actually see their valuations surge as more companies shift to usage-based pricing, according to PwC partner Alex Baker, who leads the firm’s US technology, media and telecommunications deals.

“Where agents become the new user, potentially operating 24/7, a hundred times more actions per day—if you have a true moat and a valuable platform, you’re worth more than you were in the previous environment, right?” he said.

Inspired Capital’s von Tobel said she was spending more time looking for companies that are growing more slowly but have an edge that makes them stickier with customers and indicates their tech can’t be replaced as easily.

“If a company shows up and says, ‘I’ve existed for one month and I went from zero to $100 million of ARR,’ that does not excite us,” von Tobel said. “We look for businesses where actually, the early days are hard. You’re finding hard insights or there’s significant IP—we’ve backed a lot in the infrastructure space, quantum computing, photonics, et cetera, where the ideas are 42-patents-pending-type ideas, where it’s not so easy for a competitor to walk in the door and replicate it.”

The AI build-out isn’t a great fit for traditional construction or project financing lending, said Morgan Stanley’s Shah, because the capital needs are simply too big and the pace of development is too fast for those kinds of markets to keep up with. Instead, bankers and investors have had to get creative with new types of lending deals.

One example is Meta Platforms’ joint venture with Blue Owl to build the Hyperion data center campus in Louisiana, which Shah worked on and called a first-of-its-kind transaction. That deal was funded with $27 billion in investment-grade debt but structured so that Meta could keep the debt off its own balance sheet.

Brookfield Asset Management has focused on going deep into data center customer contracts to understand who’s actually on the hook for data canter payments, who the various counterparties are and whether the project can be covered out of the borrower’s own cash flow. Delays in construction—and how borrowers’ own customers might penalizethem—are another risk lenders have to keep in mind.

“There’s generally a point where it has to be operational for that contract to kick in. And if you’re not operational, you may actually have to pay penalties attached to that, and then the debt does not wanna be exposed to that,” said Hadley Peer Marshall, chief financial officer and co-head of infrastructure credit at Brookfield.