For years, investors have worried about one thing when it comes to Nvidia: competition.
The concern seems logical. When a company becomes this dominant, rivals eventually catch up, prices come down, and profit margins shrink. It is a pattern that has played out repeatedly across the technology industry.
Yet according to Gil Luria, Head of Technology Research at DA Davidson, Nvidia may be an exception, at least for the rest of this decade.
His argument is simple: the world’s largest AI customers still do not have many alternatives.
The AI Gold Rush Still Runs Through Nvidia
The biggest buyers of AI chips today are the hyperscalers.
These include companies like Microsoft, Amazon, Google, and Meta, which are spending hundreds of billions of dollars building AI infrastructure.
While these companies are trying to diversify their suppliers, the reality is that most of their AI workloads still run on Nvidia hardware.
That dependence matters.
When customers have limited alternatives, suppliers gain pricing power. And pricing power is what allows Nvidia to maintain some of the highest profit margins in the entire semiconductor industry.
Nvidia recently reported:
- Revenue growth of 85% year-over-year
- Quarterly revenue of $81.6 billion
- Adjusted gross margin of roughly 75%
To put that into perspective, a 75% gross margin means Nvidia keeps approximately 75 cents of every dollar of revenue after covering production costs.
Very few hardware companies operate at that level.
Why Competition Hasn’t Hurt Nvidia Yet
Investors often point to AMD and Broadcom as potential challengers.
Both companies are making significant investments in AI chips.
Both have relationships with major cloud providers.
And both are becoming increasingly important players in the AI ecosystem.
But according to Luria, they remain in the early stages of competing with Nvidia’s complete platform.
The key word here is platform.
Many investors focus only on the chip itself. Nvidia’s advantage extends far beyond silicon.
Its strength comes from a combination of:
- Hardware
- Software
- Developer tools
- AI frameworks
- Networking infrastructure
- Years of ecosystem development
The result is that switching away from Nvidia is not as simple as replacing one chip with another.
For hyperscalers, moving workloads to alternative solutions can involve significant engineering effort, software adjustments, and operational risk.
That makes negotiations much harder.
Even the largest technology companies in the world are not necessarily in a strong position to demand lower prices when they remain heavily dependent on Nvidia’s ecosystem.
The Market Wants More Than Good Results
Interestingly, while Nvidia’s business remains strong, investor expectations have become incredibly demanding.
The AI boom has trained markets to expect exceptional numbers every quarter.
A good example came this week with Broadcom.
Broadcom delivered strong revenue growth and continued AI momentum. Under normal circumstances, the results would have been viewed positively.
Instead, the stock suffered a sharp decline after investors felt its AI chip outlook was not strong enough.
This highlights a broader trend.
The market is no longer rewarding companies simply for participating in AI.
Investors now want:
- Faster growth
- Bigger forecasts
- Larger market opportunities
- Clear evidence of AI monetization
Anything less can trigger significant volatility.
What Could Threaten Nvidia’s Margins?
While Luria believes Nvidia’s margins are relatively safe through 2030, investors should remember that no competitive advantage lasts forever.
Several developments could eventually pressure margins:
Custom AI Chips
Major cloud providers are increasingly developing their own chips.
Google has TPUs.
Amazon has Trainium.
Microsoft and Meta are investing heavily in custom silicon initiatives.
If these solutions become powerful enough for broader workloads, Nvidia’s pricing power could weaken.
AMD Gains Share
AMD continues improving its AI offerings and has already secured important customer wins.
Even modest market share gains could create more pricing competition over time.
Software Becomes Less Dependent on CUDA
A significant part of Nvidia’s moat comes from CUDA, its software ecosystem.
If developers become more platform-agnostic and alternative software stacks mature, switching costs could decline.
AI Spending Slows
Nvidia’s margins also benefit from extraordinary demand.
If AI infrastructure spending eventually normalizes, customers may become more price-sensitive.
The Bigger Picture
The debate around Nvidia is no longer whether AI is real.
That question has largely been answered.
The real debate is how much of the AI value chain Nvidia can continue to capture.
Today, Nvidia remains the central supplier powering the world’s largest AI infrastructure projects.
The company is not just selling chips. It is selling an ecosystem that many customers still find difficult to replace.
As long as hyperscalers remain dependent on that ecosystem, Nvidia’s pricing power remains intact.
That is why some analysts believe gross margins above 70% are not a temporary phenomenon, but a feature of Nvidia’s business model for the remainder of the decade.
Whether that proves correct will depend on one thing: how quickly competitors can give customers a genuine alternative.
For now, the alternatives exist. But according to Nvidia’s financial results and the continued buying behavior of hyperscalers, they are not yet strong enough to challenge the market leader.