NVIDIA’s rise in high-performance computing is a near-perfect embodiment of the American Dream. Three friends, including Taiwanese immigrant Jensen Huang, tired of life at a major computer company, decide to risk it all to compete in the highly competitive computing hardware field. They hatch a plan while eating at a Denny’s in 1993, as Huang had worked as a waiter at Denny’s diners in his youth. Huang himself considers their proposal an “unfundable idea” with “zero percent chance of success.” After securing funding anyway, the new company faces many setbacks and barely survives. The company then grows rapidly, eventually becoming a $3 trillion company that is now driving the artificial intelligence (AI) revolution and the record rise of the U.S. stock market.
But the Biden administration is now planning to punish the company for this surprising success. On September 3, the Department of Justice sent a subpoena to Nvidia, deepening its antitrust investigation into the company’s business practices and the AI market more generally. This came shortly after European authorities raided Nvidia’s overseas headquarters in a separate antitrust investigation.
What’s surprising about this response is that these regulators likely didn’t even know the company existed until recently. Until a few years ago, Nvidia was just a maker of advanced 3D graphics processing units (GPUs) for gamers. Nvidia’s stock price is soaring now, but it has been trading below $1 a share for years.
But as the cryptocurrency and AI revolution took hold, the company pivoted. It realized it could provide a new type of high-performance GPU badly needed to meet the world’s ever-growing demand for data processing. Over the past decade, Nvidia’s transformation has helped fuel an explosion of private-sector-led AI investment, reaching $335 billion, more than triple China’s financial stake in the industry.
At a time when many are concerned about the growth of Chinese algorithms and computing power, Nvidia’s role in expanding America’s AI dominance should earn it praise and admiration from political leaders. But the Biden administration appears intent on punishing the company for growing too big, too quickly.
Biden’s antitrust policy team doesn’t seem to understand the wisdom of Joseph Schumpeter, who taught us 80 years ago to reject short-term market snapshots and understand that the most important competition comes from the unexpected: What really matters, Schumpeter argued, is “competition from new goods, new technologies, new sources of supply, and new types of organizations” because such innovations “attack not the limits of incumbent firms’ profits or their output but their very foundations and their very life.”
Schumpeterian competition is a race for the prize of profit, which drives entrepreneurs like the founders of NVIDIA to take bold risks to unseat market leaders with superior innovation. Depriving successful companies of that prize, which Schumpeter called “the most powerful engine of progress,” would be a huge disincentive for current or future companies to take such risks at all. The result would be technological stagnation and reduced economic productivity and growth.
Nvidia is a perfect example of how Schumpeter’s “perpetual storm of creative destruction” still shakes the tech economy in relentless and unpredictable ways. So it’s no wonder that Huang frequently expresses his “worry and concern” about so many other competitors trying to take market share from his company. Nvidia’s unofficial motto is “We’re 30 days away from bankruptcy.”
Huang’s anxiety is understandable. The race for high-performance computing is fierce. “The key is that you have a lot of options,” said Lisa Su, CEO of AMD, one of Nvidia’s main rivals. “I think we’re going to get to a point where there’s not just one solution, but multiple solutions,” she told reporters in December. She noted that chip designs are becoming more sophisticated every day. Amazon, Apple, Google, Meta, Microsoft and Tesla are also developing their own AI hardware solutions that will power the next generation of computing and robotics.
While antitrust regulators seem concerned about Nvidia’s “full-stack” computing dominance in AI hardware and software, the reality is that every segment of the AI ecosystem is booming. Forbes’ recent survey of the world’s 50 most promising AI innovators received more than 1,900 applications, more than double last year’s, leading the magazine to conclude that “the artificial intelligence sector is more competitive than ever.” FirstMark’s 10th annual map of the AI and machine learning business landscape features a record 2,011 company logos for 2024, including 578 new entrants and many new business segments. The first version in 2012 had just 139 company logos and a handful of market segments.
Compare the rise of NVIDIA to Intel’s recent failure. Intel once assumed the role of disruptive innovator, with former CEO Andy Grove’s famous belief that in a competitive world, “only the paranoid survive.” In contrast, Intel’s motto today is more like “only the politically connected thrive.” The company is all but entwined with the Biden Administration’s Commerce Department and its industrial policy spending rush, following the massive CHIPS and Science bill passed in 2022.
The bloated law allocated $39 billion in direct subsidies for semiconductor investment, plus another $75 billion in loans and loan guarantees. Intel received much of this generous aid, positioning itself as the nation’s semiconductor backbone of last resort. But Intel now finds itself “experiencing weaker-than-expected sales and bleeding cash,” and is considering more drastic measures, such as scaling back factory expansion plans that were supposed to be facilitated by government funds.
Some have suggested a “CHIPS 2” bill might be necessary, but that would only further politicize the AI and computing markets, especially as Washington seeks to attach a variety of DEI conditions to grant funding, including labor and childcare mandates and the burden of environmental permitting that has stymied many other infrastructure projects. Even advocates of tech industry policy acknowledge that adding “woke” mandates to the mix would make it even more costly and inefficient, even as it is inefficient enough as it is.
Intel and the tech industry should have listened to former Cypress Semiconductor CEO TJ Rogers. In 2000, Rogers implored fellow Silicon Valley innovators to avoid “normalizing relations” with big government. “The political climate in Washington runs counter to the core values that drive our success in the international marketplace and risks turning entrepreneurs into nationalistic businessmen,” he argued. Other tech leaders of the time took the same approach. Steve Jobs, for example, actively avoided Washington, and it was not until years after his death that Apple opened a full-fledged policy office in Washington, DC. Today, every major digital technology leader (except Elon Musk) has an army of lobbyists in Washington.
Did NVIDIA invite its own downfall by avoiding political maneuvering? The Chinese Communist Party certainly hopes so, and has invested heavily in expanding its own AI capabilities, especially in military applications. China is working to develop its own “national champion” to rival NVIDIA in achieving the CCP’s avowed goal of overtaking the United States to become the world’s AI leader by 2030. Beijing may already be ahead of us in that race, but the CCP is tying the hands of its domestic AI innovators with another constraint: freedom of expression. Like the internet, AI systems require a free flow of information to fully flourish. Beijing cannot tolerate such freedom and has moved to impose a massive censorship regime, stifling its own AI capabilities in the process.
This gives the United States a significant strategic advantage in both the AI industry and geopolitics. But to leverage that advantage, the United States must avoid heavy-handed AI regulation and antitrust threats. As the Lexington Institute noted in a recent report, U.S. national security requires a strong, innovative technology sector as part of “the new arsenal of democracy.” Thriving U.S. companies not only strengthens America’s international competitiveness, but also helps more pluralistic values, rather than those of the Chinese Communist Party, shape the world’s information platforms and markets.
Unfortunately, the Biden administration is undermining these goals by over-regulating AI in general through executive orders while trying to punish AI market leaders like NVIDIA. In his massive AI executive order in October 2023, more than 100 pages long, Biden sought to turn the 74-year-old Defense Production Act into a regulatory order that limits algorithmic innovation, even though it is a law intended to expand industrial production. This follows the administration’s 2022 AI Bill of Rights, a tough document that argues that algorithmic systems are “highly harmful” and will “exacerbate discrimination” without extensive preventative bureaucratic oversight. Overseeing this entire effort was “AI czar” Kamala Harris, and if she wins the presidential election in November, she may pursue this “algorithmic fairness” agenda even more aggressively. If so, it would be the biggest gift China could hope for as it tries to catch up in the global AI race.
Photo: Costfoto/Future Publishing via Getty Images
