Will Nvidia’s AI Bet Continue to Pay Off in 2025? #NewsUnitedStates
CashNews.co
Nvidia (NASDAQ:NVDA) bears have been proven wrong continuously in the past two years, and the stock is up 923% since. This company went on to become the largest in the world before declining, but it is on another rally that could make it the biggest company yet again. Very few people would’ve thought that such a thing would be possible even just a year ago.
That said, the stock market is–and always will be–a weighing machine in the long run. The bulls may have been right so far, but that does not mean that Nvidia will keep its shine forever. No stock in history has defied gravity for years and years without a major correction of some sort. Conversely, the bearish argument here is taking a very long time to play out. There’s competition on the horizon, but the numbers coming out of Nvidia’s quarterly reports are simply too high for the market to not reward.
Anyone telling you with absolute certainty about where this train is heading is lying. However, what we can do is dig deeper and see what can go right and what can go wrong. You can then form your own opinion–with a lot more confidence. Let’s start!
We’re going to assume that you already have a decent idea of what Nvidia does. On that note, we’re going to jump into the most important part of Nvidia’s business: Compute & Networking.
Compute & Networking is Nvidia’s largest and fastest-growing segment. It generated $22.68 billion in revenues in Q1 fiscal 2025, up by an extraordinary 5.1X year-on-year. The segment further accelerated in Q2 fiscal 2025 and reached $26.3 billion in revenue. This marked a 16% sequential increase and a staggering 154% year-over-year growth.
The success here stems from Nvidia’s flagship H100 data center GPUs. These GPUs are poised to get even better with the upcoming H200 and Blackwell B200 architectures. Speaking of Blackwell, it claims to deliver up to 30 times greater inference performance and consume 25 times less energy for massive AI models compared to its predecessors.
Now let’s see what the competition has to offer.
The underdogs AMD and Intel are making significant strides against Nvidia. However, they’re more focused on trying to undercut Nvidia’s chips instead of trying to outperform them. AMD’s Instinct MI300X chip focuses on inference workloads and Intel’s Gaudi 3 accelerator is positioning itself as a more cost-effective alternative.
AI hardware isn’t the only thing Nvidia is working on, though. Nvidia’s competitive moat extends beyond hardware. Its CUDA software platform has become the industry standard, creating what CEO Jensen Huang calls a “virtuous circle.” The company is expanding its ecosystem with new products like Spectrum-X Ethernet for AI and NVIDIA AI Enterprise software.
Moreover, Nvidia’s networking arm–which reported revenue of $3.2 billion in Q1 fiscal 2025 (a 242% year-over-year increase)–is becoming increasingly important. Nvidia’s Spectrum-X product line is expected to become a multibillion-dollar business within a year. In Q2, its Networking revenue hit $3.7 billion, up 114% year-over-year.
Nvidia is clearly miles ahead of the competition so far. There’s no denying it. That said, being overkill does have its own drawbacks. There’s no guarantee that the broader AI industry will grow in a way that would let Nvidia continue to post such insane growth figures. It’s a good idea that we look into the growth potential here more closely.
According to Fortune Business Insights, the global artificial intelligence market is projected to grow from $621.19 billion in 2024 to $2,740.46 billion by 2032. This is at a compound annual growth rate (CAGR) of 20.4%.
Looking specifically at Nvidia, Bank of America analyst Vivek Arya believes that Nvidia can capture a significant portion of the AI market. He estimates that with Nvidia maintaining its market share, this translates to $272 billion in AI computing revenues for Nvidia by 2030. For reference, Nvidia’s total revenue for fiscal year 2024 was $61 billion.
However, not all analysts are quite as bullish. The consensus among analysts suggests that Nvidia will generate roughly $1.29 trillion in cumulative revenue between fiscal years 2025 and 2030. While still an impressive figure, it falls short of the most optimistic projections.
How could this play out? Well, let’s take a look.
All of these are based on what analysts estimate right now. We could dig into each of these estimates separately, but that would take us outside the scope of this article. Here are what the estimates for its Data Center/AI segment look like for the next two years.
Nvidia’s growth going forward clearly has a lot of variability but should remain fast-moving in the near term unless competition really starts to ramp up.
Let’s dig deeper into the catalysts that could drag growth into the lower end of the estimates we’ve presented.
Macroeconomic factors can throw a wrench into those estimates. Plus, there’s no knowing if AI hype will continue to hold steady. AI startups are flush with cash due to enthusiastic funding rounds from venture capitalists. However, they can get impatient fast if profitability does not improve in the coming years.
Another catalyst that is quite under-mentioned is the reliance of Nvidia on a few key customers. This level of customer concentration is considered “highly unusual” for a company of Nvidia’s size. Most other mega-cap companies don’t have even a single customer accounting for 10% of revenue. Let’s look at what each customer is trying to do:
Nvidia’s revenue from these customers is likely to peak in 2024-2025 and then gradually decline. New customers and markets are likely to offset some losses from major tech companies, but the customer makeup here makes Nvidia prone to any cooldown in how these big software companies view AI.
This is probably the most difficult part of our analysis, as we have to rely on “guesstimates.” There are many variables at play with Nvidia, but the current valuation looks fair, or even slightly overvalued. We do not see any signs of Nvidia being a deeply undervalued company. The market has been more than generous. (Read our article about what NVDA’s current valuation priced in)
However, this also does not mean that Nvidia is out of steam. Momentum has proven analysts wrong–again and again–and it can continue to do just that if the broader tech market keeps rallying.
If you are a momentum investor, Nvidia is a good play. Even if you are a value investor and you believe in the prospects of AI, going long on NVDA stock is not a bad idea. You could also start taking some profits if you’re sitting on big gains.
What we wouldn’t suggest doing is shorting NVDA. The momentum here can get you burned pretty quickly.
Regardless, all of this is short-term. Our long-term analysis is that NVDA stock is heavily dependent on the broader tech industry and macro factors. This is a cyclical stock that has a lot of room for correction if things do go wrong.
You’re paying 36 times forward earnings for this business. The growth so far certainly warrants that premium. The problem arises when we consider whether or not that growth will stick around in the coming years.
If the AI segment’s growth starts to falter, the market will have a very hard time justifying this premium. Moreover, any slowdown in the development of AI models or a better GPU chip by a competitor could erode Nvidia’s margins quickly. Wall Street is only paying 36 times trailing sales due to its sky-high margins. However, this will not be the case if margins take a dive. This has happened before during previous downturns.
We have good reason to believe that any future recession would erode margins significantly. This is because AI startups and software companies will likely cut corners when it comes to GPUs if VC/public funding starts to take a hit. But again, that’s if a recession happens.
Whether or not you should buy Nvidia depends on how much risk you’re comfortable taking. The upside here has been more and more momentum-driven, so value-focused investors should scrutinize NVDA stock more before going long.
If you’re an aggressive investor, going long on NVDA is not a bad idea, but you should keep in mind that any pullbacks in the broader market could hurt you a lot more.
According to my base case, I see Nvidia $140-220 a year from now. That said, a recession could easily drag it below $100.
While we acknowledge the potential of NVDA as an AI play, our conviction lies in the belief that some AI stocks hold greater promise for delivering higher returns, and doing so within a shorter time frame. If you are looking for an AI stock that is more promising than NVDA but that trades at less than 5 times its earnings, check out our report about the cheapest AI stock.
READ NEXT:8 Best Wide Moat Stocks to Buy Now and30 Most Important AI Stocks According to BlackRock.