It’s no exaggeration to say Nvidia has become an engine of global growth. The company now accounts for 3.6% of global GDP growth, a larger footprint than the entire stock markets of the UK, France, or Germany. That kind of dominance means every earnings report isn’t just about one company – it’s a pulse check on the entire AI industry, and by extension, the broader market. The latest results once again showed breathtaking growth, but also a few warning signs.
For Q2 FY2026, revenue surged 56% year-over-year to $46.7 billion, with data centre revenue jumping 56% to $41.1 billion. Earnings climbed 54% to $1.05 per share, beating estimates by 4%.
Still, the stock slipped – why?
For one, expectations were sky-high.
Data centre sales came in just shy of consensus, and the absence of China’s H20 shipments (delayed for now) left some analysts disappointed.
Secondly, past dominance doesn’t guarantee future success – especially as competition heats up. Profit margins are already showing cracks: not long ago, Nvidia kept $0.78 of profit on every $1 of sales. Today, that’s slipped to $0.71 and trending lower.
The reason? Big Tech is cutting its dependence. Amazon, Alphabet, Microsoft, and Meta are all designing their own chips.
• Amazon Web Services launched its in-house Graviton4 chip in July, part of a broader push from its Annapurna Labs.
• Google has been rolling out Tensor Processing Units (TPUs) for data centres, AI, and even smartphones.
• Apple builds its own Apple Silicon processors, while still tapping into Google’s TPU ecosystem.
• And earlier this year, Meta began testing its custom Meta Training and Inference Accelerators (MTIA).
Each of these chips is designed with one clear purpose: reduce reliance on Nvidia.
Nevertheless, management guided Q3 revenue to about $54 billion – right in line with Wall Street’s $53.4 billion estimate, but below the most bullish calls above $60 billion.
Importantly, this guidance does not include H20 sales to China, which could add $3–5 billion per quarter once shipments begin. CEO
Jensen Huang also emphasised that licenses are secured to resume business there.
Despite the “slower” growth, CEO Huang struck a confident tone, reiterating that AI spending is only accelerating. Nvidia estimates $3–4 trillion in AI infrastructure investment by the end of the decade . Even if Nvidia captures only a slice of that pie, it remains one of the most important companies in the world. With a 32x forward PE and 50% projected earnings growth, the stock hardly looks “expensive” compared to many tech names.
So, is the AI boom over?
Not quite. If anything, they show that the boom is maturing. Early hypergrowth is giving way to a new phase defined by competition, regulation, and more realistic expectations.
For investors, that means the easy money may have been made – but the opportunity is far from over. Nvidia is still a key player in AI’s future, just no longer the only one.
The next wave of winners may be found in the broader AI ecosystem – not just the chipmakers. That includes cloud providers like Amazon Web Services, Microsoft Azure, and Google Cloud, which power the massive computing needs behind AI models. It also includes software developers building AI tools, platforms, and applications that make these models usable in the real world.
Then there are the companies supplying the infrastructure that makes AI possible: data centre operators, networking equipment makers, and storage providers. Even firms that produce specialised cooling systems, high-speed memory, or advanced servers are set to benefit as AI workloads grow.
In other words, while Nvidia is still at the heart of AI today, many other players are quietly carving out profitable niches. Savvy investors who understand the full ecosystem may find opportunities that are less risky but still tied to the explosive growth of AI. Make sure you’re following South African Investor.
Not a subscriber to Money Morning?
You can get free daily recommendations like these with Money Morning eletter. Just sign up here.