The artificial intelligence boom has been in full swing for the last three years, and in that time it has attracted unprecedented levels of investment. With little sign of a slowdown anytime soon, it’s no wonder that warnings of the industry entering a speculative bubble have reached a fever pitch, sparking fears of an imminent economic meltdown.
Technology giants are spending billions of dollars, not only to keep pace with the use of AI chatbots like ChatGPT and Google Gemini, but also to prepare themselves for the emergence of enterprise automation, where machines are expected to take over much work that’s now done by humans. Ultimately, the total cost of the AI buildout could run into trillions, with a significant amount of it coming from venture capitalists, debt and some increasingly inventive financing arrangements.
By now, conventional wisdom has it that the AI market is heading into uncharted territory, and even those at the forefront of the boom acknowledge that they’re taking a big risk. Yet, they remain steadfast in their belief in AI’s ultimate destiny to reshape the world, transform the way we work and accelerate human progress.
That may be so, but never before has so much cash been thrown at such an unproven technology. The reality is that the leading players in AI are still searching for a profit-making business model, perhaps because they feel as if they have little choice but to delay monetization in order to keep up with their rivals in terms of user adoption. The belief in AI is so great that no one dares risk falling behind and being sidelined, even if it means taking on an enormous financial risk.
Throughout November, the unease over the future of AI finally came to the fore, as seen by the sudden volatility in the world’s most valuable tech stocks.
The fears are certainly not unfounded. Zeev Farbman, CEO of the AI video software company Lightricks, said in an op-ed recently that there’s no doubt in his mind that a correction is coming. “Ongoing dramatic market swings among AI-driven tech giants, with public companies losing more than $1 trillion in valuation in less than a month, clearly illustrate that valuations are disconnected from fundamentals,” he said.
Farbman is far from alone in sharing his fears that the AI hype could soon turn into an economic train wreck. The amount of spending on building out data centers and renting powerful GPUs and other accelerators is truly off the scale, said Crazy Stupid Tech’s Fred Vogelstein. He believes that it’s not merely a bubble but the biggest technology mania in history – and one that will inevitably end badly.
Even more worrying, Vogelstein believes we could be just months away from experiencing a dramatic correction. “We’re months, not years, from the end of the party,” he wrote. “We may be even closer than that. Nvidia posted better than expected earnings on Wednesday. And it briefly looked like that would buoy all AI stocks. It didn’t.”
Can investors weather the storm?
Pretty much everyone who knows anything about the financial markets has been issuing warnings of an AI bubble burst, but at the same time, almost everybody agrees that the technology is about to transform the very fabric of our society. There’s an overwhelming consensus that AI will spark hundreds of innovations over the next few years, including many that are yet to be imagined. So investors don’t want to miss out.
Meta Platforms’ founder and CEO Mark Zuckerberg has no qualms about the risks he’s taking. Speaking on the Access podcast in October, he said that the risk of being too slow to build out AI infrastructure was bigger than the risk of “misspending” a few hundred billion dollars.
That might be fine for Zuckerberg, but what about the rest of us? Most investors don’t have a majority share in a $1.65 trillion company as a safety net in case things go wrong. So how can they shield themselves from the prospect of an AI market that suddenly goes pop?
Investors might do well to remember what happened during the dot-com era, which was the last major bubble that emerged in the tech industry at around the time the internet began to go mainstream. The best known survivor is Amazon.com, which not only weathered the storm but went on to become one of the richest and most famous companies in the world, creating fabulous wealth for its founder and early backers.
Amazon survived because it hit upon a unique business model that provided genuine value to the masses – cheaper, better quality products delivered directly to their door. That meant Amazon drummed up enough business to maintain cash flow and continue growing while many of its peers crashed and burned.
Lessons can also be learned from some of the dot-com bust’s most spectacular failures. Pets.com famously raised millions of dollars in venture capital, but it was never even remotely profitable. It was based on a flawed business model that involved excessive spending on marketing and selling products at a loss to try and outcompete its rivals. It also struggled with the high costs of shipping bulky items to customers, and eventually ran out of cash when venture capital dried up.
Tiz Gambacorta, a former institutional fund manager and now co-founder of the financial media firm Eunice.io, wrote in Forbes about the need to identify companies with some kind of competitive moat. “Those with the most value – and resilience – will boast high-quality datasets, high-profile partnerships and high-visibility integrations with existing software already used by millions of people,” he said.
But it’s still possible to make riskier bets, Gambacorta said, so long as investors do their homework. For those who really want to cash in on the AI boom, the trick is to try and find companies that are deploying AI to solve real-world problems that nobody else is. He points out that AI is a technology that is going to impact potentially hundreds of industries, including cybersecurity and the energy sector.
“Compelling opportunities can be found in smaller rocks working behind the scenes to make this new age a reality,” he insisted.
Farbman stressed caution and the importance of carefully evaluating how AI firms build, optimize and integrate their models into real workflows to create actual products and services. “Investors should look for metrics such as economics by workload when looking at AI applications,” he said. “These, not simply the talents or the proprietary nature of the model itself, are the key elements of value.”
Proprietary models are probably a much riskier proposition in general, Farbman added. He said investors should be able to find greater value in open-source models, which can be just as powerful as those hidden in black boxes. Open-source models “often out-iterate closed APIs because researchers and developers can adapt them locally,” he explained. “That adoption compounds into a moat around a company or a product, helping to guarantee profits and success.”
AI isn’t going away
No one knows for sure what will happen next in the AI market, or if it’s really going to crash. But when so many economists unite to warn of the risks, smart investors are wise to pay attention.
Few AI-only companies are making any real revenue, mostly because they haven’t even hit upon a killer application yet. The AI industry remains very much in its infancy, and though the potential is there, many of the startups raising millions of dollars in VC money are ultimately destined to fail.
But not all of them will, and when the AI bubble bursts, the hype will go away, leaving the underlying technology here to stay. Those who keep their heads and put their money behind the applications offering genuine value will be the ones left holding sway.
