Image © Adobe Images
The flood of new AI-related stock and debt issuance is making today’s artificial intelligence boom look increasingly similar to the internet bubble that ended in a brutal market collapse, according to veteran technology analyst Richard Windsor.
Windsor, founder of Radio Free Mobile and a former technology strategist at Nomura, says a series of major transactions suggests the industry is rushing to meet investor demand in much the same way internet companies did during the late 1990s.
"The AI Boom has one saving grace, which is demand, but this has to be profitable to avoid a collapse, and here there are question marks," he says.
The warning comes as investors prepare for a wave of AI-related fundraising by the world's biggest firms.
The issuance comes against a backdrop of elevated sovereign bond issuance by the major developed economies, pointing to increasing competition for global savings.
IPOs to Hoover Up Billions
It's a major risk theme we will be watching in the coming months as it will impact fixed income and foreign exchange, as well as stock markets.
At the centre of the dash-for-cash is the expected IPO of SpaceX, which Windsor says could raise as much as $86BN at a valuation approaching $1.8TR.
Anthropic and OpenAI are also expected to pursue public listings, creating what Windsor describes as three gigantic IPOs arriving in quick succession.
The concern is not simply the size of the deals.
One of the defining features of the internet bubble was a shortage of investable technology stocks, which drove valuations sharply higher until companies began issuing large amounts of new equity.
Google to Hoover Up Billions
"The internet industry started issuing paper in the form of IPOs to meet that demand, and after a while, the demand did not materialise, and the market collapsed," Windsor says.
He argues that AI is now entering a similar phase, with the supply of investment opportunities rapidly increasing.
Alongside the IPOs, Alphabet - Google's owner - has launched a massive equity raise to fund AI spending ($84BN), while Morgan Stanley estimates that AI infrastructure investment will require some $450BN of debt issuance this year alone.
From Break-even to Profit
"This is highly problematic because if the economics of AI compute do not improve, there will be no money with which to pay back the debt," says Windsor, adding:
"The difference with the AI Boom is that demand for AI compute remains very far above the industry's ability to supply it.”
The question now becomes whether companies can make enough money from that demand to justify their surging valuations.
“This time around, it is the profitability of the compute that is being sold that is the problem.”
Windsor argues that much of the industry is currently selling AI compute at or near break-even levels.
That may be enough to justify continued investment today, but it is unlikely to satisfy investors indefinitely.
Severe Punishment for the Unprofitable
“Break-even is not nearly enough to pay back the debt or satisfy investors who at some point will ask where their returns are.”
For that reason, Windsor says the key metric to watch is no longer demand growth but profitability.
“This is why it is the profitability of compute that I am looking at very carefully, as this will determine whether we are at the dawn of a new age or the top of another investment bubble.”
He adds that he has no idea when markets will begin focusing on that issue, but believes the eventual reckoning could be severe.
“At some point it will, and at that time those still losing money will be severely punished by the market.”
