A group of investors is gathered outside a convention center in San Jose late in the afternoon following an AI conference session. The smell of roasted coffee wafts from a nearby café, and the phrase “the next Nvidia” keeps coming up in conversations.
In the world of tech investing, it’s practically a catchphrase.
Nvidia has been doing something uncommon in Silicon Valley history for the past few years. The business not only profited from a technological revolution, but it also served as its catalyst. Originally created for video games, its graphics processing units are now used to run massive artificial intelligence models in data centers all over the world. Investors were left staring at charts that resembled mountain cliffs rather than stock performance as Nvidia’s valuation skyrocketed.
The hunt has started, of course.
The same question is being asked by retail traders, venture capitalists, and Wall Street analysts. Who will follow Nvidia if it emerges as the key player in the AI era? It’s an alluring concept. According to history, a new giant is always emerging somewhere, just out of sight.
These days, a peculiar blend of impatience and excitement can be seen when strolling through venture offices in Palo Alto or Midtown Manhattan. Investors discuss chip startups in the same way that early internet investors discussed browser companies. It’s a familiar atmosphere. It resembles the late 1990s, when everyone was speculating about which company would control the internet economy. Many of the guesses made back then were incorrect.
| Category | Details |
|---|---|
| Company | Nvidia Corporation |
| Founded | 1993 |
| Headquarters | Santa Clara, California, United States |
| CEO | Jensen Huang |
| Core Business | Graphics Processing Units (GPUs) and AI computing infrastructure |
| Market Position | Dominant supplier of AI training chips for data centers |
| Estimated Share of AI Data Center Chips | Around 80%+ in recent estimates |
| Key Competitors & Challengers | Broadcom, AMD, custom chip startups, hyperscaler AI chips |
| Major Industry Trend | Shift toward custom AI processors and AI infrastructure expansion |
| Reference | Nvidia official website |
| Reference | CNBC coverage of Nvidia AI chip investments |
The unexpectedness of Nvidia’s dominance contributes to its allure. For many years, the company was primarily recognized by computer graphics engineers and gamers. Few people, even in Silicon Valley, could have predicted that GPUs would become the central component of artificial intelligence.
However, Nvidia appeared less like a graphics company and more like the infrastructure of the digital age once big AI models started training on enormous clusters of chips.
Everything was altered by that change.
Investors now appear to be certain that the next trillion-dollar opportunity will appear somewhere along the same supply chain. However, it’s still unclear exactly where that opportunity is. Custom AI processors, or chips made for particular models rather than general computing, are thought to be the cause. Others contend that the true winners might be found further down the infrastructure, providing massive AI data centers with memory, networking, or even power.
The next Nvidia might not even look like Nvidia.
Think about Broadcom, a business that subtly controls the market for custom ASIC chips. Its processors are made especially for hyperscale businesses developing massive AI systems. Artificial intelligence-related revenue has been increasing quickly; some executives estimate that sales of AI chips will reach tens of billions of dollars annually in a few years.
The story is still far from resolved, though.
Another possibility is beginning to emerge inside data center construction sites throughout the American Southwest. Massive amounts of electricity are used in AI computing. It becomes clear that the true limitation might not be chips at all when you stand outside one of these facilities at dusk, with rows of gray cooling towers humming softly. Perhaps it’s power.
Investors are starting to take notice.
Some hedge fund managers now discuss energy infrastructure firms with the same zeal that they used to reserve for software startups. It’s not glitzy. However, the economics start to make sense when a single AI facility can consume as much electricity as a small city.
The market might still be underestimating how tangible this AI revolution is.
As the frenzy develops, it’s difficult to ignore how quickly tech investing narratives take shape. The natural response was to imitate Nvidia’s success once it became evident. Look for a different chip manufacturer. Support the upcoming GPU champion. Hold off until the stock chart repeats itself. However, technological advancements are rarely linear.
While the largest profits surfaced in the fiber-optic networks, semiconductor suppliers, and hardware companies that enabled the web, many investors pursued internet companies during the dot-com boom. It’s possible that something similar is happening once more.
Indeed, chips are necessary for artificial intelligence. However, it also requires a lot of electricity, data centers, networking hardware, memory, cooling systems, and rare metals.
It’s possible that the next giant is already developing somewhere within that network of infrastructure.
It appears that investors are aware of it. For this reason, the same agitated question can be heard in conference rooms, venture capital meetings, and trading desks.
Who will take over as Nvidia’s successor?
To be honest, no one truly knows. And maybe the search is so appealing because of that uncertainty. There’s a sense that the next chapter of this tale hasn’t been written yet as the AI economy grows, with factories constructing servers, engineers testing new chips, and investors poring over balance sheets late at night.
