The Golden Age of AI Is Experiencing Its First Significant Slowdown: A Market Correction or a Bursting Bubble?

Generative AI has become a major topic in the tech world since ChatGPT debuted in late 2022. Big tech companies and startups have invested billions of dollars in its development, but recently, doubts have been growing about whether we’re in a speculative bubble that’s about to burst, similar to the dot-com bubble 25 years ago.

Is it a real revolution or an overvalued fad?

The Recent Context

AI has captured the attention of users, companies, and investors. Companies like Nvidia and many others have seen huge increases in their stock market value in recent months. For instance, Nvidia hit a peak market value of $3.3 billion in June and is now one of the Big Tech players.

Even Alphabet CEO Sundar Pichai stated in 2018, long before the current stage, that AI would be as important to humanity as electricity or fire. However, in recent weeks, we’ve seen several worrying signs.

• On Monday, August 5, the shares of several tech companies fell sharply. The “Magnificent Seven” collectively lost almost $1 trillion in value in a single day.
• Nvidia has delayed its B200 chips, dampening confidence in the industry’s pace of innovation.
• Warren Buffett, arguably the world’s greatest investor, has halved his Apple stake.
• The Goldman Sachs report “Gen AI: too much spend, too little benefit?” published on June 27, has brought the current lack of return on AI investments to the forefront.

Several recent events have sparked debate on whether AI is a speculative bubble about to burst or not.

The Arguments of the Skeptics

Gary Marcus is a renowned AI expert who has become a regular critic of the current AI hype. For some time, he’s been using his X account to point out the word “bubble” in reference to today’s generative AI situation. In a recent article in his newsletter, he delved into it further:

“Generative AI itself won’t disappear. But investors may well stop forking out money at the rates they have, enthusiasm may diminish, and a lot of people may lose their shirts. Companies that are currently valued at billions of dollars may fold, or stripped for parts.”

Marcus’ skepticism is based on several factors:

1. Generative AI doesn’t work as well as it’s touted to... and perhaps never will.
2. Language models such as GPT-4o still have problems, such as hallucinations or committing very basic errors.
3. It’s not entirely clear yet how to effectively monetize these technologies.
4. Training larger and larger models is unsustainable in terms of infrastructure and resources.

Chrys Taylor, a reporter at Mashable, has noted patterns resembling previous tech bubbles:

“Drive to Silicon Valley on any of the Bay Area’s main arteries right now, and you’ll notice nearly every billboard pumping a product ‘driven by AI.’

On the same drive five years ago, you’d see the same scene with the word ‘blockchain.’ Ten years ago: ‘big data.’ Twenty-five years ago: literally any word followed by ‘.com.’”

Additionally, Elliott Management, a hedge fund managing $70 billion in assets, claimed in a letter to its clients published by the Financial Times that Nvidia is in a “bubble” and the AI driving its share price is “overhyped.”

Elliott expressed skepticism about the Big Tech (Nvidia’s main clients) continuing to purchase Nvidia GPUs in high volumes, and argued that many AI applications wouldn’t be profitable, work effectively, or be reliable. They would also consume too much power.

The firm also mentioned that AI hasn’t yet delivered the promised significant increase in productivity, with few practical uses, limited to “summarizing notes of meetings, generating reports and helping with computer coding.”

The Optimistic Voices

At the other end of the debate, AI advocates argue that the technological revolution we’re currently experiencing is real and can be compared to the emergence of the Internet. Pichai emphasized this when he likened its potential impact to that of electricity or fire.

On one hand, they point out that the recent drops in the stock market are natural corrections following the significant increases earlier this month. The crucial difference lies in determining whether this is a simple correction or the start of the burst of the bubble.

On the other hand, technology companies are demonstrating concrete results with AI. Nvidia has broken its revenue and profit records, while Alphabet reported nearly 30% growth in Google Cloud revenue due to increased demand for AI services. There are many examples.

AI advocates also argue that current investments are laying the foundation for future advancements. According to Amy Hood, Microsoft’s chief financial officer, the company’s investments in data centers “will be monetized over 15 years and beyond,” Reuters reported.

And of course, companies like OpenAI and Google DeepMind are consistently making breakthroughs in the capabilities of their AI models.

A Question of Balance

The reality may lie somewhere in between extreme hype and outright skepticism, although the exact position is unknown. Axios reporter Felix Salmon provides a more nuanced view:

“Whether you think the stock market is telling a tale of boom or gloom depends mostly on whether you're looking at its level or at its direction.”

In other words, while the current valuations of AI companies may seem exaggerated, it doesn’t mean that the technology itself is a fraud. This situation is similar to what happened with the Internet after the dot-com bust: Expectations were overblown, but the Web continued to grow and eventually took over. First, there was a correction, and then more sustainable long-term growth.

In a few years’ time, it’s quite possible that the current situation will follow the same path.

Jeremy Bowman at The Motley Fool agrees that what we’re experiencing isn’t so much the burst of a bubble but rather a market correction.

“Rather than the AI bubble bursting, the recent decline in AI stocks seems to be best explained by two factors. First, there’s a market correction in AI stock prices. After this year’s rally, investors believe the sector may be getting overheated and are selling these stocks and booking profits...

Secondly, there’s a market rotation happening. Investors are moving money out of large-cap tech stocks like Alphabet, Microsoft, and Nvidia and, anticipating rate cuts from the Federal Reserve, putting it into small-cap stocks, which have badly underperformed in the AI era.”

It’s becoming increasingly clear that there’s some level of hype surrounding AI. Warnings from major players like Elliott Management and recent stock market declines indicate that investors are starting to question the rapid growth of companies like Nvidia.

Nevertheless, it’s hard to deny the transformative potential of AI in various industries. The challenge ahead for investors, companies, and the media (ahem…) will be to differentiate between the practical and profitable applications of AI and the empty promises and unwarranted hype.

In the years to come, there will be companies that fail to deliver on their promises, but there’ll also be dominant players that will shape the future of AI. Just like Gmail emerged in 2004 when mail giants were already established, we might see similar trajectories in the AI space. As such, I=it’s uncertain whether companies like OpenAI or Nvidia will maintain their leadership in the long run.

This article was written by Javier Lacort and originally published in Spanish on Xataka.

Image | Kid Circus

Related | The AI Bubble Is Starting to Deflate: Big Tech Had Its Worst Day on the Stock Market Since the Rise of ChatGPT