“Despite the high price tag, the technology is still a long way from being useful,” Jim Covello, a senior equity analyst at Goldman Sachs and a 30-year veteran who has covered technology companies, wrote in a recent report on AI. “When you build too much of something that the world isn't useful or ready for, it usually ends badly.”
Covello's comments contrast with another report published by Goldman Sachs a little over a year ago, when some of the bank's economists said AI could automate 300 million jobs around the world and increase global economic output by 7% over the next decade, sparking a flurry of reports about AI's disruptive potential.
According to Barclays, Wall Street analysts expect big tech companies to spend about $60 billion a year on developing AI models by 2026, but will only generate about $20 billion in annual revenue from AI at that point. That level of investment would be enough to power 12,000 products the size of OpenAI's ChatGPT, Barclays analysts wrote in a recent report.
Get caught up in
Stories to keep you up to date
OpenAI released ChatGPT in November 2022, kicking off a race in Silicon Valley to develop new AI products and get people using them. Big technology companies are spending tens of billions of dollars on the technology. Retail investors are soaring the shares of these companies and their suppliers, especially Nvidia, which makes computer chips used to train AI models. Since the beginning of the year, Google's shares have risen 32%, Microsoft's 20%, and Nvidia's shares have risen more than 150%.
Venture capitalists are pumping billions more into thousands of AI startups. The AI boom is helping to contribute $55.6 billion that venture investors pumped into U.S. startups in the second quarter of 2024, the highest single quarterly amount in the past two years, according to venture capital data firm PitchBook.
Tech executives say AI will change every aspect of modern life, just as the internet and mobile phones did. AI technology has advanced dramatically and is already being used to translate documents, write emails and help programmers write code. But some companies that predicted an AI boom last year are now worried about whether the tech industry will be able to recoup the billions of dollars it's investing in AI anytime soon, or ever.
“We expect to see a lot of new services, but probably not 12,000,” Barclays analysts wrote. “We sense Wall Street is becoming increasingly skeptical.”
April: Meta, Google, Nvidia Both companies signaled their commitment to AI, telling investors on their quarterly earnings calls that they will spend more on building out data centers to train and run AI algorithms. Google reiterated Tuesday that it will spend more than $12 billion per quarter on building out AI. Microsoft and Meta are due to report their own earnings next week, which could give further hints at their AI roadmaps.
Pichai said Tuesday that it will take time for AI products to mature and become more useful. He acknowledged that AI is expensive, but said the data centers and computer chips the company is buying could be repurposed for other uses even if the AI boom slows.
“For us, the risks of underinvesting are significantly greater than the risks of overinvesting,” Pichai said. “There are much more significant drawbacks to not investing to lead here.”
A Microsoft spokesman declined to comment. A Meta spokesman did not respond to a request for comment.
Unrealistic expectations
Vinod Khosla, co-founder of computer network systems company Sun Microsystems and one of Silicon Valley's most influential venture capital investors, compared AI to personal computers, the Internet and mobile phones when asked about the magnitude of its impact on society.
“These are all fundamentally new platforms. In each of these, the new platform will create an explosion of applications,” Khosla said. The rapid entry into AI could create a financial bubble in which investors lose money, but that doesn't mean the underlying technology won't continue to grow and become more important, he said.
“The dot-com bubble happened because prices went up and down, according to Goldman Sachs. I don't think internet traffic has decreased at all.”
He says many startups will fail as AI transforms the way people work, do business, and interact with each other. But the industry as a whole will benefit from AI. He predicts that AI will eventually create trillion-dollar businesses, including humanoid robots, AI assistants, and programs that can perfectly replicate the jobs of highly paid software engineers.
But so far, AI hasn't helped VCs make more money on their investments: VC exits (initial public offerings or acquisitions of tech startups) totaled $23.6 billion in the second quarter, down slightly from $25.4 billion the previous quarter, according to PitchBook.
David Kahn, a partner at venture firm Sequoia Capital, wrote in a blog post last month that to replenish all the money currently being invested in AI, the tech industry would need to generate roughly $600 billion in annual revenue, but it's nowhere near that figure.
“Speculative fever is part of technology and there's nothing to be afraid of,” Khan said, “but we need to stop buying into the delusion that's spread from Silicon Valley to the rest of the country and the world, which says we can all get rich quick.”
Microsoft and Google are seeing growing revenues, especially from their cloud businesses that sell access to AI algorithms and storage space to use them. Executives at both companies say AI is driving new interest in their products and will be a big revenue driver in the future. But some analysts say that beyond OpenAI's ChatGPT and Microsoft's coding assistant GitHub Copilot, few standalone products have been huge successes.
“Twenty months later, Wall Street is growing increasingly skeptical, given the two big consumer and enterprise successes of ChatGPT and GitHub Copilot,” Barclays analysts wrote in the report.
Vineet Jain, CEO of AI and data management company Egnyte, said the cost of developing and running AI programs will come down as other companies compete with Nvidia and the technology becomes more efficient. Right now, he said, he doesn't expect to see any AI-related revenue this year because the cost of delivering AI products is too high. But that will change as costs come down and demand continues to rise, Jain said.
“The value proposition is certainly there, but expectations are still unrealistic at this point,” he said, referring to the frenzy to sell AI products to consumers and businesses.
Some startups are already fading from the heyday of the early AI boom. Inflection AI, a startup founded by veterans of Google's famed AI lab DeepMind, raised $1.3 billion last year to expand its chatbot business. But in March, the company's founder left for Microsoft, poaching some of his top employees to the tech giant. Other AI companies, including Stability AI, one of the first companies to develop a widely used AI image-generating tool, have been forced to lay off employees. The industry is also facing litigation and regulatory challenges.
Large companies like Google and Microsoft will be able to keep pouring money into AI products until demand grows, but smaller startups that have raised large amounts of venture capital may not be able to survive the transition, Jain said.
“It's like a soufflé that keeps rising and rising and then it has to come down a little bit.”
