Will the race for artificial general intelligence (AGI) lead us to a land of financial abundance – or will it end in a 2008-style fiasco? Billions of dollars depend on the answer.
The figures are staggering: an estimated $2.9 billion (£2.2 billion) will be spent. spent on data centersthe central nervous system of AI tools; the $4 trillion-plus market capitalization of Nvidia, the company that makes the chips that power cutting-edge AI systems; and the $100 million signing bonuses offered by Mark Zuckerberg’s Meta to top engineers at OpenAI, the company behind ChatGPT.
These dizzying figures are all supported by investors who expect a return on their billions. AGI, a theoretical state of AI in which systems achieve human levels of intelligence across a range of tasks and are able to replace humans in white-collar jobs such as accounting and law, is the keystone of this financial promise.
It offers the prospect of computer systems doing profitable work without the associated cost of human labor – an extremely lucrative scenario for the companies developing the technology and the customers deploying it.
There will be consequences if AI companies fail: US stock markets, strongly driven by the performance of technology stocks, could fall and harm citizens’ personal wealth; debt markets, immersed in the data center boom, could suffer a jolt that ripples elsewhere; GDP growth in the United States, which has benefited from AI infrastructure, could weaken, impacting interdependent economies.
David Cahn, a partner at one of Silicon Valley’s leading investment firms, Sequoia Capital, says tech companies now need to deliver on their AGI promises.
“Nothing less than AGI will be enough to justify the investments currently proposed for the coming decade. » he wrote in a blog published in October.
This means that a lot depends on progress towards advanced AI and the trillions invested in infrastructure and R&D to get there. One of the “godfathers” of modern AI, Yoshua Bengio, says progress in AGI could stall and the results would be bad for investors.
“It is clearly possible that we will hit a wall, that there will be difficulties that we do not foresee at the moment and that we will not find a solution quickly,” he says. “And it could be a real (financial) crash. Many of those currently investing trillions in AI also expect progress to continue fairly steadily at the current pace.”
But Bengio, a striking voice on the security implications of AGIit is clear that continued progress towards a very advanced state of AI is the most likely endgame.
“Blocking progress is a minority scenario, like an unlikely scenario. The most likely scenario is that we continue to move forward,” he says.
The pessimistic view is that investors are supporting an unrealistic outcome: AGI will not happen without further progress.
David Bader, director of the Data Science Institute at the New Jersey Institute of Technology, says trillions of dollars are being spent to expand — tech jargon for making something grow quickly — the technology behind chatbots, known as transformers, in the hope that it will be enough to increase the computing power behind current AI systems, by building more data centers.
“If AGI requires a fundamentally different approach, perhaps something we haven’t designed yet, then we optimize an architecture that can’t get us there, no matter how big it is. It’s like trying to reach the moon by building taller ladders,” he says.
Nevertheless, large American technology companies, such as Google’s parent company, AlphabetAmazon and Microsoft are pursuing their data center projects with the financial cushion to fund their AGI ambitions with cash generated from their highly profitable day-to-day operations. This at least gives them some protection if the wall Bengio and Bader designed appears.
But this boom presents other, more worrying aspects. Analysts at Morgan Stanley, the US investment bank, estimate that $2.9 trillion will be spent on data centers by 2028, half of which will be covered by cash flows from “hyperscalers” like Alphabet and Microsoft.
The rest will have to be covered by alternative sources such as private credit, a corner of the shadow banking sector It is activate the alarm bells at the Bank of England and elsewhere. Meta, owner of Facebook and Instagram, borrowed $29 billion from the private credit market to finance a data center in Louisiana.
AI-related sectors account for about 15% of investment-grade debt in the United States, which is even larger than the banking sector, according to investment bank JP Morgan.
Oracle, which signed a $300 billion data center deal with OpenAIhas seen an increase in credit default swaps, which are a form of insurance in the event a company defaults on its debts. High-yield securities, or “junk debt,” which represent the riskiest segment of the debt market, are also emerging in the AI sector through data center operators CoreWeave and TeraWulf. The growth is also being financed by asset-backed securities – a form of debt backed by assets such as loans or credit card debt, but in this case rents paid by technology companies to data center owners – in a form of financing that has increased sharply in recent years.
No wonder JP Morgan says the AI infrastructure boom will require contribution from all sectors of the credit market.
Bader says: “If the AGI does not materialize on schedule, we could see contagion across multiple debt markets simultaneously – investment grade bonds, high-yield speculative debt, private credit and securitized products – all of which are leveraged to finance this construction. »
AI and technology-related stock prices also play an outsized role in U.S. stock markets. The so-called “magnificent 7” of American technology stocks – Alphabet, AmazonApple, Tesla, Meta, Microsoft and Nvidia – represent more than a third of the value of the S&P 500, the largest stock index in the United States, up from 20% at the start of the decade.
In October the Bank of England warns of the “risk of a strong correction” in the American and British markets due to the dizzying valuations of technology companies linked to AI. Central bankers fear stock markets could collapse if AI fails to reach the transformative heights investors hope for. At the same time, the International Monetary Fund said valuations were heading toward dot-com bubble levels.
Even tech executives whose companies are profiting from the boom recognize the speculative nature of the frenzy. In November, Sundar Pichai, Alphabet’s chief executive, said there were “elements of irrationality” in the boom and that “no company would be safe” if the bubble burst, while Amazon founder Jeff Bezos said the AI industry was in a “kind of industry bubble,” and OpenAI chief executive Sam Altman said “there are many aspects of AI that in my opinion, are rather bubbly at the moment.”
All three, to be clear, are AI optimists and expect the technology to continue to improve and benefit society.
But when the numbers get this high, there are clear risks of a bubble bursting, as Pichai admits. Pension funds and everyone who invests in the stock market will be hit by a collapse in stock prices, while debt markets will also be affected. There is also a web of “circular” deals, such as OpenAI paying Nvidia cash for chips, and Nvidia will invest in OpenAI for non-controlling stock. If these deals fail due to a lack of AI adoption or hitting that wall, it could be complicated.
Some optimists also say that generative AI, the catch-all term for tools like chatbots and video generators, will transform entire industries and justify the spending. Benedict Evans, a technology analyst, says the spending figures are not outrageous in the context of other sectors, such as oil and gas extraction, which is worth $600 billion a year.
“These investments in AI are a lot of money, but it’s not an impossible amount of money,” he says.
Evans adds: “You don’t have to believe in AGI to believe that generative AI is a great thing. And most of what’s happening here isn’t about ‘oh wow, they’re going to create God’. It’s more about ‘this is going to completely change the way advertising, search, software and social media – and everything our business is built on – is going to work’. This will be a huge opportunity.”
Nevertheless, the AGI is expected to reach several billion dollars. For many experts, the consequences of getting there are alarming. The cost of failing to do so could also be significant.