[ad_1]
Years of unbridled AI optimism have given way to strains of skepticism, even within the business and investment communities, as calls of an AI bubble have grown as of late, drawing comparisons to the dot-com boom and bust at the turn of this century.
“The concept of an AI bubble is not entirely new,” Ram Bala, associate professor of AI & Analytics at Santa Clara University, told Newsweek. “For more than a year, there has been this discussion [as] the investment numbers almost began to look a little unreal…from billions to trillions.”
In the last few months, chip companies saw slowed sales and stock growth, though Nvidia’s recent earnings announcement has assuaged some concerns. Also, the efficacy of AI in the workplace is not as great as most people thought at this point, and the vast environmental costs of this technology are becoming increasingly apparent.
A Bank of America survey found that 45 percent of global fund managers said there was an “AI bubble” that could negatively impact the economy. An MIT study made waves with the finding that 95 percent of enterprise generative AI deployments do not achieve financial returns. The International Energy Agency reports that one ChatGPT request uses 10 times more energy than a Google search, and the rise in demand for data centers is a potential strain on the world’s water supply.
Those heavily invested in the future of automation and generative technology may have hoped to see greater adoption at this point. The lack of workplace adoption, identified by MIT, Gartner and banking analysts, is driving some of the bubble talk. In many industries, business leaders seem to struggle with the change management focus needed to empower employees to adopt new tech-enabled workflows.
“It will take longer than I think currently predicted to see the gains,” Hatim Rahman, associate professor of management and organizations and sociology at Northwestern University, told Newsweek. “Because this is not a plug and play technology. This is a technology that requires fundamentally rethinking change management, adoption of culture, people processes, which, research for decades has shown, takes time.”
The proliferation of AI also stokes fears of job loss at a scale that would be ruinous to the economy. While the labor market is certainly unstable and layoffs are occurring at a variety of different companies, attributing that instability to AI at this point would be premature, and inaccurate.
“In the last few years, so many people have talked about [jobs] going away, almost every one of those predictions was wrong,” Kian Katanforoosh, CEO of AI startup Workera and a lecturer on machine learning at Stanford University, told Newsweek. “People overestimate the technology and underestimate the human capacity that is needed to integrate that technology. I see that every single day.”
Katanforoosh acknowledged that AI has a lot of hype right now, and some people have been benefiting in the investment market. Most of the beneficiaries, however, may be at large chip-making and technology giants, rather than AI-powered startups and their early investors.
“Companies that get a massive valuation just for putting AI in their mission statement but fail to deliver could still go to zero,” Samuel Hammond, chief economist at the Foundation for American Innovation, told the Los Angeles Times. “But most of the stock market’s growth is being driven by the large-cap tech stocks like Nvidia and Google.”
Today, the internet is a pretty crucial aspect of our personal and business lives, but the pile of investment behind its future was at times misguided. Like the internet, or generative AI, it is a common notion to perceive an emerging technology as capable of changing the world, but following through on that nugget with a successful investment strategy is a different animal.
Observers note that government investment, across the world, in data centers, serves to mitigate the financial risks of the infrastructure investments occurring to advance AI.
“The question is more about specific numbers, did we go a little bit too high? Now, there’s a correction. In my view, that’s what’s happening,” Bala said. “A short term correction.”
The nature of a bubble, whether it is around tulips, businesses with prominent web domains or AI tech company stocks, is that people buy into their financial future, literally, and get burned when the bubble bursts.
“Jumping on a bandwagon is predicated on this idea that there is going to be some returns,” Bala continued. “If those returns don’t pan out, that’s when there is a collapse,” like in a housing bubble, “when prices are going up, people keep investing more and more in housing, and the only way that is sustainable is if the house prices keep going up.”
If consumer and enterprise demand for emerging AI technology does not rise, a lot of people are going to lose a lot of money. But we’re “still in the very early innings,” Bala cautions. Like with the internet, investments in infrastructure may go unused, but eventually they are filled.
Right now, adoption into workflows and wide-scale reshaping of work or consumer processes has yet to occur. But perhaps it is on the horizon, just in a timeline longer than expected.
“People are very slow to change,” Katanforoosh said. “We’ve seen that in prior cycles of technology.”
[ad_2]