Nvidia — the most valuable company in the world thanks to its place atop the AI food chain — generated a staggering $120 billion in profits last year.

This includes an eye-popping $43 billion during the three-month period ending in January, one of the strongest quarters of any business ever recorded.

Investors didn’t seem to care.

Shares of Nvidia fell more than 5% Thursday, following the results. With a total market value of roughly $4.5 trillion, the company’s one-day loss amounted to roughly $256 billion worth of market capital.

Nvidia’s stock price decline is part of a broader phenomenon dubbed the “AI scare trade” that is percolating in certain corners of the stock market.

This bearish play threatens the very driver that has powered broader, double-digit gains across the benchmark S&P 500 for the past two years.

And while the stock market might look broad, its gains are increasingly concentrated in just a handful of mega-cap names, including Nvidia. In other words, the entire market’s performance is heavily tied to the performance of these select companies.

Nvidia says its growth story is very much still intact.

“We have now seen the inflection of agentic AI and the usefulness of agents across the world and enterprises everywhere,” Nvidia CEO Jensen Huang said during the company’s quarterly earnings call Wednesday, referring to AI chatbots like OpenAI’s ChatGPT and Anthropic’s Claude.

“You’re seeing incredible compute demand because of it,” he said. “In this new world of AI, compute is revenues.”

Compute refers to the processing power that is needed to train and operate AI models. Nvidia’s chips, each of which is around 30 inches square, underpin the massive data centers needed to run AI chatbots and agents.

Nvidia’s dominance in the AI chip race also means that more companies than ever are dependent upon its products, at a time when AI is evolving faster than even its early adopters say they could have imagined.

Over the past five years, this has sparked a massive investor rush to buy a piece of Nvidia, whose shares have surged nearly 1,300% since the start of 2021.

Driven by a mixture of FOMO and faith in AI’s growth-at-any-cost business model, these investors and others like them have piled into virtually any company with even a tangential relationship to the AI industry.

All this time, Nvidia has led the charge. But so far this year, Nvidia’s share price is barely positive. Some firms, including HSBC, have argued that in order to justify another leg higher in the company’s stock price, Nvidia needs a “new narrative,” such as a meaningful expansion in AI demand or pricing power.

But more broadly, the AI scare trade visited upon Nvidia on Thursday underscores a growing unease around the future of AI.

After a multiyear boom for public and private companies alike, AI is now facing tougher scrutiny. Questions persist around whether or not the AI boom is starting to look more like a bubble.

Likewise, investors are uncertain whether AI can generate the kind of near-term returns necessary to justify the massive investments — and soaring share prices — coursing through the tech world.

“Artificial intelligence stands to become one of the most consequential technologies in generations, if not in the history of humankind, with enormous implications for the economy,” Moody’s economist Mark Zandi wrote in a new report Wednesday. “However, the specifics of how it will shape the future remain highly uncertain and are the subject of immense debate.”

That debate includes growing concerns over how AI agents will affect vulnerable industries like cybersecurity and software — and potentially upend traditional business models that have worked for decades.

Shares of software companies like ServiceNow and Synopsys have fallen sharply amid those fears, declining roughly 20% and 15% over the past month, respectively. Salesforce is down nearly 25% this year.

So far this year, companies in the software industry have been the largest drag on the S&P 500.

AI has “started to call into question how exactly software companies are really going to compete and provide something superior in this environment,” Melissa Otto, head of Visible Alpha research at S&P Global, told NBC News.

Nvidia’s Huang attempted to push back on this narrative in an interview with CNBC on Wednesday.

The “markets got it wrong” when it comes to the AI-driven panic around software, he said. Huang argued that AI will enhance productivity and expand what software can do, rather than kill the whole industry.

Huang’s attempt to assuage investor fears didn’t move the needle much, though. On Thursday, the tech-heavy Nasdaq fell nearly 1.5% on the back of Nvidia’s slide. Software giants like Synopsys dropped 5% while shares of Microsoft and Alphabet also traded lower.

Beyond software, investors are contending with other existential anxieties. Many of them were captured this week in an essay posted on Substack by a small research firm called Citrini Research. The post warned that AI adoption would lead to a stock market crash, a sharp pullback in consumer spending and widespread white-collar layoffs by 2028.

The report painted a vivid picture of an economic doomsday scenario caused by AI, effectively animating investors’ vague, simmering fears. Payment giants like Mastercard and American Express were hit particularly hard after the post named them as potential casualties in a lower-spending, AI-disrupted economy. Shares of the two payment giants rebounded slightly Thursday.

Many Wall Street analysts say it’s too soon to panic.

“While we take concerns about the AI trade and private markets and other matters seriously, we think it’s premature to assume that’s the kind of risk we face today,” Lori Calvasina, head of U.S. equity strategy research at RBC Capital Markets, wrote in a client note earlier this month.

Kristy Akullian, BlackRock’s head of iShares investment strategy for the Americas, added in a separate note Thursday that the recent sell-off “is predicated on still uncertain existential risk,” rather than any immediate changes to company earnings or business fundamentals.

Nonetheless, this existential risk is one that investors are taking more seriously now than they did six months ago.