Jensen Huang Explains Why Nvidia Could Grow 70% as AI Demand Explodes

Nvidia’s Jensen Huang Says AI Demand Could Drive 70% Revenue Growth — Here’s Why He’s So Confident:

Nvidia Could Reach $680 Billion in Revenue as AI Infrastructure Expands.

Nvidia CEO Jensen Huang believes the company is positioned for another extraordinary year of growth, despite intensifying competition in the AI chip market. His confidence comes from something few companies can match: Nvidia has visibility across almost every layer of the global artificial intelligence ecosystem.

At a recent Goldman Sachs Communacopia + Technology conference, Huang offered an unusually detailed explanation of why he believes Nvidia's revenue could grow by approximately 70% year over year next year.

If analysts' current expectations of roughly $400 billion in revenue for Nvidia's current fiscal year are accurate, 70% growth would put next year's revenue at approximately $680 billion.

That is an extraordinary number. But Huang argues that Nvidia isn't simply selling GPUs anymore.

It has become one of the foundational infrastructure platforms powering the global AI revolution.

Nvidia Is No Longer Just a GPU Company:

For years, Nvidia was primarily known for graphics processors used in gaming PCs. That image is now dramatically outdated.

Today's Nvidia is an AI infrastructure company.

Huang emphasized this transformation when explaining the enormous scale of modern Nvidia systems.

A high-end AI computing system is no longer simply a single graphics card that can be installed inside a computer. Today's AI infrastructure combines GPUs, CPUs, high-speed networking, memory, software, cooling, power systems, and thousands of components into massive computing platforms.

Huang pointed out that one of Nvidia's advanced GPU systems can represent millions of dollars in equipment and enormous amounts of power and infrastructure.

That difference illustrates just how dramatically the economics of computing have changed since the early days of the GPU.

The AI boom has transformed the GPU from a component primarily associated with gaming into one of the most strategically important pieces of technology in the global economy.

The AI Infrastructure Opportunity Is Getting Bigger:

One of the strongest arguments behind Huang's optimism is the sheer amount of infrastructure required to operate modern AI.

Generative AI models are becoming larger, more sophisticated, and more computationally demanding.

Large language models, multimodal AI, AI reasoning systems, autonomous AI agents, AI coding tools, and enterprise AI applications all require substantial computing resources. And Nvidia is positioned at the center of this infrastructure expansion.

The company's technology is being used by:

• Major cloud providers

• AI research laboratories

• AI startups

• Hyperscalers

• Neocloud providers

• Enterprise technology companies

• AI-native startups

• Data-center operators

This gives Nvidia an unusually broad view of where AI infrastructure is being built.

Nvidia’s AI Systems Are Becoming Massive:

Huang highlighted Nvidia's increasingly sophisticated computing platforms, including systems built around its Grace CPUs and Blackwell GPUs.

One example is Nvidia's GB200 NVL72 system, which combines 36 Grace CPUs with 72 Blackwell GPUs.

According to Huang, orders for this type of system are experiencing approximately 27% month-over-month sales growth.

That figure illustrates the extraordinary speed at which AI infrastructure demand can expand.

Businesses aren't simply purchasing individual chips.They are building entire AI factories.

These systems require enormous quantities of processors, networking equipment, memory, storage, cooling, electricity, and specialized infrastructure.

As AI adoption increases, the size of these installations could continue to grow.

Why Jensen Huang Believes Nvidia Can “See the Future”:

Perhaps the most interesting part of Huang's argument isn't Nvidia's current sales. It is the company's visibility into future AI demand.

Huang explained that Nvidia works with virtually every major AI ecosystem participant. That includes organizations developing some of the world's most advanced AI models. “Nvidia runs every model,” Huang said, emphasizing the company's role as a foundational platform for AI.

Whether a company is developing a commercial AI model, an open-weight model, an AI coding system, or an AI agent, Nvidia hardware is frequently part of the underlying computing infrastructure.

This gives Nvidia an unusual perspective.

Instead of looking at AI demand from a single company's point of view, Nvidia can observe activity across hundreds of customers, developers, cloud providers, startups, and infrastructure projects.

That could help explain why Huang remains so bullish.

Nvidia Has Its Fingerprints Across the AI Supply Chain:

Nvidia's influence extends beyond selling processors.

The company sits within a much larger AI ecosystem that includes:

Semiconductor manufacturers → memory suppliers → Nvidia → cloud providers → data centers → AI developers → enterprise customers

Nvidia also works closely with original equipment manufacturers, neocloud providers, hyperscalers, and AI startups.

Huang said Nvidia tracks projects involving land, electricity, data-center buildings, and other infrastructure around the world.

This gives the company insight into how much AI computing capacity is being planned before that capacity necessarily appears in traditional financial reports.

In other words, Nvidia may be able to see the next wave of AI infrastructure spending while it is still being developed.

The AI Chip Competition Is Getting Tougher:

Nvidia's confidence comes at a time when competition in the AI semiconductor market is becoming more intense.

Major technology companies are developing their own AI chips.

Amazon, Google, and Microsoft are investing heavily in custom silicon designed to reduce their dependence on third-party processors.

AI companies are also becoming more interested in specialized computing infrastructure. Meanwhile, companies such as Cerebras and Etched are pursuing alternative approaches to AI acceleration.

This creates an obvious question:

Can Nvidia maintain its dominance as AI customers gain more alternatives?

Huang's answer appears to be yes.

Nvidia's advantage isn't just its GPU hardware.

It has built a massive ecosystem around its chips, including networking technology, software, development tools, AI libraries, and platforms.

That ecosystem makes replacing Nvidia hardware more complicated than simply purchasing another processor.

The Secret Weapon May Be Nvidia’s Software Ecosystem:

The AI chip market is often described as a hardware competition.

But Nvidia's real competitive advantage may be much broader.

Its software ecosystem has become deeply embedded in AI development.

Developers have spent years building applications and machine-learning systems around Nvidia's technology stack.

This creates an ecosystem effect.

The more developers use Nvidia technology, the more software becomes optimized for it. The more software is optimized for Nvidia, the more attractive Nvidia hardware becomes.

That creates a powerful feedback loop.

For competitors, challenging Nvidia therefore requires more than producing a faster AI chip.

They also need to build an ecosystem that developers actually want to use.

What About Nvidia’s “Circular Deals”?

Another interesting part of Huang's comments involved Nvidia's investments in AI companies. Nvidia has invested in various companies across the AI ecosystem, including companies that may subsequently purchase Nvidia hardware.

This has led to criticism and comparisons with so-called circular financing arrangements from earlier technology cycles.

Huang dismissed the criticism with humor.

His argument was essentially that Nvidia isn't investing simply to create artificial demand. According to Huang, Nvidia looks for companies with genuine customers and contracts before making investments.

He said Nvidia has seen approximately** $100 billion** worth of contracts associated with these companies.

The distinction is important.

If an AI startup receives investment, then uses the money to purchase AI infrastructure, that can create a feedback loop between investment and hardware demand.

Whether this model proves sustainable over the long term remains an important question for investors and the broader AI industry.

The AI Boom May Eventually Become More Efficient:

Despite his optimism, there is an important reason to remain cautious.AI infrastructure demand cannot grow exponentially forever.

As AI models mature, companies will likely become more efficient.

AI developers are already working on:

• Smaller AI models
• More efficient inference
• Model quantization
• Specialized AI chips
• Better data-center utilization
• Lower token consumption
• AI model routing
• Distillation techniques
• More efficient AI agents

Today, companies may spend enormous amounts of money experimenting with AI. Tomorrow, they may learn how to accomplish the same tasks using significantly less computing power.

That could eventually slow infrastructure growth.This is one of the biggest long-term questions facing Nvidia.

AI Agents Could Create the Next Wave of Computing Demand:

There is another development that could work in Nvidia's favor: AI agents.

Generative AI initially became popular through conversational interfaces. Users asked questions.

AI responded.

The next phase involves AI systems that can actually perform tasks.

AI agents can potentially research information, write software, interact with applications, manage workflows, communicate with customers, and perform complex multi-step operations. That could dramatically increase the amount of AI inference required by businesses.

Instead of one AI query producing one answer, an autonomous AI agent could make dozens or hundreds of model calls while completing a single business task.

This creates a potentially enormous new source of computing demand.

For companies building enterprise AI, private AI agents, AI automation, and AI-powered applications, infrastructure efficiency will become increasingly important.

What Nvidia’s Growth Means for Enterprise AI:

The Nvidia story isn't only about semiconductor companies and Wall Street.It also has major implications for businesses adopting artificial intelligence.

Companies increasingly want AI systems that can:

• Answer customer calls
• Handle website conversations
• Qualify leads
• Schedule appointments
• Process documents
• Automate repetitive tasks
• Assist employees
• Analyze business data
• Write and maintain software
• Connect different business systems
• Operate as intelligent AI agents

As these systems become more sophisticated, the underlying AI infrastructure becomes increasingly important.

Businesses need to consider performance, scalability, privacy, reliability, inference costs, security, and integration when deploying AI.

This is where customized and private AI infrastructure can become especially valuable.

Why Private AI Infrastructure Is Becoming Important:

The growth of AI infrastructure also raises an important issue for enterprises: data control.

Organizations handling sensitive customer information, proprietary software, financial information, legal documents, intellectual property, and internal business data may not want every AI workload processed through a public AI environment.

Private AI infrastructure can give businesses greater control over:

Data privacy, security, model deployment, customization, infrastructure, and intellectual property.

This is an increasingly important part of the enterprise AI conversation.

The future may not be about one AI model winning everything.

Instead, businesses could use a combination of public models, private models, specialized AI systems, and autonomous AI agents depending on the task.

Nvidia’s Biggest Advantage: Being Everywhere:

Perhaps the most important takeaway from Huang's comments is that Nvidia doesn't need to predict exactly which AI company will win.

It can potentially benefit from the growth of the entire ecosystem.

• If OpenAI succeeds, Nvidia can sell infrastructure.

• If Anthropic succeeds, Nvidia can sell infrastructure.

• If Google expands its AI operations, Nvidia can benefit in areas where its technology is used.

• If thousands of AI startups grow, Nvidia can potentially benefit from their infrastructure requirements.

That creates a powerful position.

Nvidia isn't necessarily betting on one AI winner. It is betting on the continued expansion of AI itself.

Can Nvidia Really Grow 70%?

That remains the billion-dollar—or potentially hundreds-of-billions-of-dollars—question. Huang says he is confident.

The company has enormous exposure to AI infrastructure, and demand for advanced computing remains strong.

But technology markets have a history of changing rapidly.

Today's dominant platform can become tomorrow's legacy technology.

Competition from custom AI chips, cloud providers, semiconductor startups, and increasingly efficient AI models could eventually put pressure on Nvidia's growth.

There is also the possibility that AI companies will discover ways to deliver significantly more intelligence with fewer computing resources.

For now, however, the demand story remains extremely strong.

The Bigger AI Picture:

Nvidia's rise tells us something much bigger about the artificial intelligence industry. The AI revolution isn't being driven by software alone.

It requires an enormous physical infrastructure consisting of:

GPUs + CPUs + networking + memory + data centers + electricity + cooling + software + AI models + AI applications.

Nvidia has positioned itself at the center of this infrastructure stack.

That is why Jensen Huang believes he has unusual visibility into the future of AI. The company isn't simply watching the AI revolution from the outside.

It is supplying the infrastructure on which much of that revolution is being built.

And if AI adoption continues moving from experimentation to large-scale enterprise deployment, the demand for computing could remain enormous for years to come.

Final Thoughts: Nvidia Is Betting on the AI Economy:

Jensen Huang's bullish outlook isn't based simply on Nvidia selling more GPUs.

His argument is that Nvidia has become a foundational platform for artificial intelligence.

The company works with AI laboratories, hyperscalers, cloud providers, neoclouds, startups, manufacturers, and data-center operators. That network provides Nvidia with an unusually broad view of where AI infrastructure demand is heading.

The biggest question is whether that advantage will survive the next phase of the AI industry.

Competition is coming. Custom silicon is improving. AI models are becoming more efficient. And businesses will eventually demand greater computing efficiency.

But right now, the AI infrastructure buildout remains one of the largest technology expansions in history.

And Nvidia remains directly in the middle of it.

For Otherworlds AI, the lesson is clear: the next phase of artificial intelligence will not simply be about bigger models. It will be about building scalable AI infrastructure, intelligent AI agents, private AI systems, and practical automation that can deliver measurable business value.

The AI race is becoming an infrastructure race—and Nvidia intends to stay at the center of it.