Nvidia Shrinks OpenAI Bet as Investor Pressure Grows While Anthropic Revenue Surges
Nvidia is scaling back the financial commitment it had planned for OpenAI’s massive Ohio data-center project, highlighting growing investor scrutiny over the chipmaker’s increasingly close financial ties with the AI companies that buy its hardware. The change comes as Nvidia faces questions about whether it should continue taking large financial positions in customers while simultaneously supplying them with the chips and infrastructure needed to expand.
At the same time, Anthropic’s latest revenue numbers are providing fresh evidence that at least some of the enormous spending driving the AI boom is being matched by rapidly growing demand. Anthropic’s revenue reportedly jumped from $4.7 billion to more than $11.5 billion in a single quarter, complicating arguments that the AI industry is being driven mainly by speculative spending. The contrasting developments highlight the central question facing the AI market: whether today’s extraordinary infrastructure investment will eventually translate into sustainable revenues and profits.
Nvidia Cuts Its Planned OpenAI Exposure
Nvidia has reportedly reduced the size of its financial guarantee for OpenAI’s planned Ohio data center.
The commitment was previously expected to reach around $250 billion but has now been reduced to just under $120 billion, according to reporting cited by The Decoder.
The change represents a major reduction in Nvidia’s potential financial exposure.
| Key Detail | Reported Figure |
|---|---|
| Original Nvidia guarantee | About $250 billion |
| Revised guarantee | Just under $120 billion |
| Reduction | More than 50% |
| Beneficiary | OpenAI |
| Project | Ohio AI data center |
| Main concern | Nvidia’s financial exposure to customers |
| Broader issue | AI infrastructure investment and bubble concerns |
The move comes as investors increasingly question whether Nvidia should simultaneously act as a semiconductor supplier, investor and financial backer of the AI ecosystem.
Why Investors Are Pressuring Nvidia
Nvidia has become one of the biggest beneficiaries of the AI infrastructure boom.
Its GPUs power many of the world’s leading AI models and data centers. But the company has also become increasingly involved in financing the ecosystem that creates demand for those GPUs.
That creates a potentially circular relationship.
The AI Investment Cycle
Nvidia
↓
Invests in AI companies
↓
AI companies buy Nvidia GPUs
↓
Data centers expand
↓
Nvidia generates chip revenue
↓
AI companies raise more capital
↓
More infrastructure spending
↓
More Nvidia demand
Investors have questioned whether such relationships could make Nvidia’s reported demand less independent than it appears.
Nvidia Is More Than a Chip Supplier
The company’s role in the AI economy has expanded substantially.
Nvidia supplies GPUs, provides networking technology and software, invests in AI startups and increasingly participates in infrastructure financing.
This gives the company exposure to several layers of the AI ecosystem.
Nvidia’s Expanding Role
Chips
+
Networking
+
Software
+
Startup investments
+
Infrastructure financing
↓
AI ecosystem
This strategy can strengthen Nvidia’s position, but it also increases the amount of financial risk connected to the success of its customers.
OpenAI Is One of Nvidia’s Most Important Customers
OpenAI requires enormous amounts of computing power to train and operate its AI models.
As ChatGPT usage expands and the company develops increasingly sophisticated models and AI agents, its computing requirements are expected to grow dramatically.
This makes OpenAI a strategically important customer for Nvidia.
However, it also creates concentration risk.
OpenAI’s Compute Demand
More users
↓
More AI queries
↓
More inference
↓
Larger models
↓
More data centers
↓
More GPUs
↓
Greater Nvidia demand
Nvidia therefore has a strong incentive to support OpenAI’s infrastructure expansion.
The Ohio Data Center Is Part of a Much Larger Buildout
OpenAI’s planned Ohio infrastructure is part of the enormous expansion of computing capacity required to support its AI ambitions.
The company has been pursuing multiple data-center partnerships and infrastructure projects.
These projects require billions of dollars in capital and enormous amounts of electricity.
The scale of the investment has intensified debate over whether AI companies can generate enough revenue to justify the infrastructure being built around them.
Nvidia Wants AI Infrastructure to Become an Investable Asset
Nvidia has increasingly worked with financial institutions to bring outside capital into AI infrastructure.
The company recently announced partnerships with major financial institutions around a potential $500 billion AI infrastructure financing effort.
The goal is to mobilize capital for data centers, chips, power infrastructure and related systems while reducing the amount of capital Nvidia itself has to provide.
Recent reporting said Nvidia could contribute up to 25% of investment on a case-by-case basis, leaving much of the financing burden with institutional investors.
New Financing Model
Wall Street capital
↓
AI infrastructure
↓
Data centers
+
Power
+
Nvidia systems
↓
Cloud and AI companies
↓
Compute revenue
↓
Investor returns
The model could allow the AI industry to expand without placing all the financial burden on technology companies.
Investor Concerns Are About Circular Financing
The concern is not that Nvidia’s chips are unnecessary.
Demand for AI computing is clearly enormous.
The question is whether companies are financing one another in ways that make the ecosystem appear stronger than it would be under independent market demand.
For example, if Nvidia invests in an AI company and that company uses the funding to purchase Nvidia hardware, Nvidia benefits both from the investment and the resulting chip sales.
Potential Circular Structure
Nvidia investment
↓
AI company receives capital
↓
AI company buys GPUs
↓
Nvidia records revenue
↓
Nvidia’s investment supports customer growth
↓
More GPU purchases
The arrangement can be commercially rational, but investors want to understand the underlying economics.
Anthropic’s Revenue Provides a Counterargument to Bubble Fears
While investors are questioning some of Nvidia’s financial relationships, Anthropic’s latest revenue growth provides evidence that AI companies are generating significant real-world demand.
Anthropic’s revenue reportedly rose from approximately $4.7 billion to more than $11.5 billion in a single quarter.
That represents growth of more than 2.4 times over the period.
The numbers are significant because they show that at least one leading AI company is rapidly converting demand for AI products into revenue.
Anthropic Has Become a Major AI Business
Anthropic develops the Claude family of AI models and has become one of OpenAI’s most important competitors.
Its products are increasingly used by businesses and developers for coding, research, customer service and other enterprise applications.
The company has also experienced rapid growth in annualized revenue.
Recent reporting indicates that Anthropic’s revenue has increased more than tenfold in each of the past three years.
Anthropic’s Growth
Earlier revenue
↓
$4.7 billion
↓
Latest reported revenue
↓
More than $11.5 billion
↓
Rapid enterprise adoption
↓
Higher AI compute demand
The growth suggests that demand for advanced AI services is becoming commercially meaningful.
Anthropic’s Numbers Challenge the Simplest Bubble Argument
The AI bubble argument generally centers on the idea that companies are spending enormous amounts of money on infrastructure before generating enough revenue to justify those investments.
Anthropic’s growth does not eliminate that concern.
However, it demonstrates that AI companies can generate substantial revenue while simultaneously expanding their infrastructure spending.
The Two Sides of the AI Debate
BUBBLE CONCERNS
↓
Huge infrastructure spending
+
High valuations
+
Large capital requirements
+
Uncertain profitability
VS
REAL DEMAND
↓
Rapid revenue growth
+
Enterprise adoption
+
Growing AI usage
+
Higher willingness to pay
The reality may ultimately fall somewhere between these two extremes.
Revenue Growth Does Not Equal Profitability
Anthropic’s revenue surge is impressive, but revenue alone cannot prove that the AI business model is sustainable.
AI companies face enormous expenses.
They must pay for GPUs, data centers, electricity, networking, engineering talent and research.
The cost of serving AI queries can also remain high as usage increases.
AI Revenue Equation
Customer payments
↓
AI revenue
MINUS
Compute costs
+
Data centers
+
Energy
+
Employees
+
Research
+
Sales and marketing
↓
Profit
The industry must eventually demonstrate that revenue can grow faster than these costs.
Anthropic Is Spending Heavily on Infrastructure
Anthropic is among the companies making enormous commitments to computing capacity.
The company relies on infrastructure from major cloud providers and chip suppliers.
Its rapid revenue growth therefore contributes directly to broader demand for AI infrastructure.
This is one reason Anthropic’s financial performance matters beyond the company itself.
AI Revenue Is Becoming More Diverse
AI companies are increasingly generating revenue from multiple customer categories.
These include individual subscriptions, enterprise software, API usage and coding tools.
Enterprise customers are particularly important because they can generate much larger and more predictable revenue streams than individual users.
AI Revenue Sources
Consumers
+
Developers
+
Enterprise customers
+
API usage
+
AI agents
↓
AI company revenue
Diversification could make AI businesses more resilient as the market matures.
Coding Is a Major AI Growth Area
AI coding tools have become one of the strongest commercial applications for frontier models.
Developers can use AI to write, review, debug and maintain software.
This creates repeated demand for model inference rather than one-time usage.
Anthropic has benefited significantly from demand for Claude in coding-related applications.
AI Coding Cycle
Developer
↓
AI coding assistant
↓
Code generation
↓
Testing
↓
Debugging
↓
More AI requests
↓
Recurring compute demand
This creates a direct connection between AI software revenue and GPU demand.
Nvidia Benefits From Both OpenAI and Anthropic
Nvidia supplies hardware to a large portion of the AI industry.
The company therefore benefits when OpenAI, Anthropic and other AI companies expand their computing capacity.
This gives Nvidia exposure to the broader growth of AI rather than dependence on a single model provider.
However, its investment relationships with customers have made the financial structure more complicated.
The AI Infrastructure Boom Is Getting More Expensive
The amount of capital required to build frontier AI infrastructure is rising rapidly.
Modern data centers require huge amounts of electricity, advanced cooling and high-speed networking.
AI companies therefore need access to enormous pools of capital.
AI Infrastructure Stack
AI model
↓
GPUs
↓
Servers
↓
Networking
↓
Data center
↓
Power
↓
Cooling
↓
Real estate
↓
Financing
The economics of the AI boom therefore extend well beyond the price of individual chips.
Power Is Becoming a Major Constraint
AI data centers consume enormous amounts of electricity.
In many markets, securing power capacity can take years.
This means that even if a company has enough money to buy GPUs, it may not be able to deploy them immediately.
Power availability could therefore become one of the biggest constraints on AI expansion.
The AI Boom Is Becoming a Financing Story
The industry is increasingly moving from a technology investment story to a financing story.
AI companies need capital.
Data-center developers need capital.
Chip suppliers are participating in financing.
Banks and private-equity firms are increasingly providing capital.
AI Financing Ecosystem
AI startups
+
Cloud providers
+
Chip companies
+
Data centers
+
Banks
+
Private equity
↓
AI infrastructure financing
This interconnected system makes the financial health of each participant increasingly important.
Nvidia’s Reduced OpenAI Guarantee May Limit Risk
Reducing the OpenAI guarantee could allow Nvidia to maintain its strategic relationship with OpenAI while limiting potential financial exposure.
That could reassure investors who worry that Nvidia is taking too much risk outside its core semiconductor business.
It also signals that the company is paying attention to market concerns.
But Nvidia Still Has Strong AI Exposure
The reduction does not mean Nvidia is abandoning OpenAI.
Nvidia remains one of the most important suppliers of computing infrastructure for AI companies.
The company can benefit from OpenAI’s expansion simply by selling hardware and related technology.
Reducing financial guarantees therefore does not necessarily reduce Nvidia’s exposure to AI growth.
Anthropic’s IPO Could Provide a Major Test
Anthropic is preparing for a potentially enormous initial public offering.
Recent reports suggest investors could value the company at more than $2 trillion if it reaches ambitious future revenue targets.
Reuters reported that the valuation case is tied to expectations of approximately $190 billion to $200 billion in revenue by 2028.
That is an extraordinary forecast.
Investors Will Demand Evidence of Sustainable Growth
If Anthropic goes public at a multitrillion-dollar valuation, public-market investors will be able to examine its finances more closely.
They will likely focus on:
- Revenue growth
- Gross margins
- Compute costs
- Customer concentration
- Enterprise retention
- Cash burn
- Capital expenditure
- Infrastructure commitments
- Future profitability
The IPO could therefore provide one of the clearest tests yet of the economics behind frontier AI.
Anthropic’s Growth Could Support Nvidia’s Thesis
If Anthropic and OpenAI continue generating rapidly increasing revenue, the demand for Nvidia GPUs could remain strong.
AI companies would have more financial resources to purchase computing capacity.
That would strengthen Nvidia’s argument that the AI infrastructure boom is supported by genuine demand.
But AI Revenue Growth Must Eventually Slow Into Profit
No business can grow at extraordinary rates indefinitely.
Eventually, AI companies will need to transition from explosive revenue growth to sustainable margins.
The key question is whether model costs will fall quickly enough as hardware and software become more efficient.
Future AI Economics
More capable models
↓
More customers
↓
Higher revenue
PLUS
Better hardware
+
Better algorithms
+
More efficient inference
↓
Lower cost per query
↓
Higher margins
This could create a sustainable long-term business model.
Competition Could Put Pressure on Prices
OpenAI and Anthropic are not operating in isolation.
They compete with Google, Meta, xAI, Chinese AI companies and a growing number of open-source model developers.
Competition could push model prices lower.
Lower prices could accelerate adoption but reduce revenue per unit of compute.
Competitive Pressure
More AI models
↓
More competition
↓
Lower prices
↓
More usage
↓
Higher volume
↓
Uncertain profitability
The balance between pricing and usage will be crucial.
Open-Weight Models Add Another Challenge
Open-weight AI models can be downloaded and deployed by companies without paying per-use fees to a model provider.
If these models become sufficiently capable, they could reduce demand for proprietary APIs.
However, operating open models still requires substantial computing infrastructure.
This means the impact on Nvidia could be mixed.
Nvidia Could Remain the Biggest Infrastructure Winner
Regardless of which AI company ultimately dominates, most leading AI systems require substantial computing resources.
Nvidia therefore has the potential to benefit from competition between AI companies.
OpenAI wins
↓
GPU demand
Anthropic wins
↓
GPU demand
Google wins
↓
GPU demand
Open-source models grow
↓
GPU demand
↓
Nvidia can benefit across scenarios
This is one of the strongest elements of Nvidia’s AI investment thesis.
The Risk Is Overbuilding
The biggest threat to Nvidia’s long-term growth may not be a single AI company failing.
It could be excessive infrastructure investment.
If companies build more data-center capacity than customers ultimately need, GPU utilization could fall.
That could reduce rental prices and delay future hardware purchases.
Overbuilding Scenario
Huge capital spending
↓
Too much compute capacity
↓
Lower utilization
↓
Lower GPU rental prices
↓
Lower infrastructure returns
↓
Reduced future spending
This is the risk investors are increasingly monitoring.
What It Means for Nvidia
The reduction in its OpenAI guarantee could be interpreted as a move toward more disciplined capital allocation.
Nvidia can continue benefiting from AI infrastructure without taking on as much direct customer financing risk.
Its core business remains the sale of GPUs, networking equipment and software.
What It Means for OpenAI
OpenAI may need to find additional sources of financing for its enormous infrastructure plans.
The company has been raising large amounts of capital and forming partnerships with cloud providers, chip companies and infrastructure developers.
The reduction in Nvidia’s guarantee therefore increases the importance of outside financing.
What It Means for Anthropic
Anthropic’s revenue growth strengthens its case that enterprise AI demand is real.
However, its future valuation will depend on whether it can translate rapid growth into sustainable margins.
Its potential IPO will provide investors with much greater visibility into the economics of frontier AI.
What It Means for AI Investors
The latest developments suggest that investors should distinguish between two separate questions.
First, is demand for AI products real?
Anthropic’s growth suggests that it is.
Second, are AI infrastructure investments generating sufficient returns?
That question remains unanswered.
What Investors Should Watch
Investors should monitor:
- Nvidia’s financial commitments to AI customers
- OpenAI’s infrastructure spending
- Anthropic’s revenue growth
- Anthropic’s potential IPO valuation
- AI inference costs
- GPU utilization rates
- Data-center power availability
- Cloud-provider capital expenditure
- AI model pricing
- Enterprise AI adoption
- AI infrastructure financing
- Nvidia’s future investment strategy
These indicators will help determine whether the AI boom is developing into a sustainable technology cycle or an infrastructure oversupply.
Key Facts at a Glance
| Metric | Detail |
|---|---|
| Nvidia’s original reported OpenAI guarantee | About $250 billion |
| Revised guarantee | Just under $120 billion |
| Reduction | More than 50% |
| Anthropic earlier revenue | About $4.7 billion |
| Anthropic latest reported revenue | More than $11.5 billion |
| Anthropic potential 2028 revenue forecast | $190–$200 billion |
| Anthropic potential IPO valuation | More than $2 trillion |
| Broader Nvidia AI financing initiative | More than $500 billion |
| Main market debate | AI growth vs. infrastructure bubble |
Infographic: Two Sides of the AI Boom
NVIDIA
↓
OPENAI INFRASTRUCTURE
↓
ORIGINAL GUARANTEE
$250 BILLION
↓
REVISED
↓
JUST UNDER $120 BILLION
↓
INVESTOR PRESSURE
+
CONCERN OVER RISK
BUT
↓
ANTHROPIC
↓
REVENUE
$4.7 BILLION
↓
MORE THAN
$11.5 BILLION
↓
RAPID AI DEMAND
↓
THE CENTRAL QUESTION
↓
IS AI DEMAND REAL?
YES
↓
REVENUE IS SURGING
BUT
↓
IS INFRASTRUCTURE SPENDING SUSTAINABLE?
STILL UNCERTAIN
↓
FUTURE DEPENDS ON
REVENUE
+
MARGINS
+
COMPUTE COSTS
+
UTILIZATION
+
AI PRICING
The Bigger Picture
Nvidia’s decision to reduce its planned financial exposure to OpenAI highlights a new phase in the AI boom. The industry’s largest chip supplier is increasingly being asked to balance aggressive investment in AI infrastructure with investor demands for disciplined capital allocation. Cutting the reported OpenAI guarantee from about $250 billion to just under $120 billion allows Nvidia to maintain a major strategic relationship while limiting some of the financial risk associated with funding its customers. The development also comes as Nvidia seeks to bring more banks and institutional investors into AI infrastructure financing.
At the same time, Anthropic’s rapid revenue growth complicates the argument that the AI boom is purely speculative. Revenue reportedly increased from about $4.7 billion to more than $11.5 billion in a quarter, demonstrating that businesses are willing to spend heavily on advanced AI services. Yet revenue growth alone does not settle the bubble debate. Frontier AI companies continue to face enormous compute, infrastructure and talent costs, while competition could push prices lower. The next stage of the AI market will therefore depend on whether revenue growth can eventually translate into strong and sustainable margins.
Looking Ahead
The biggest test will come as OpenAI and Anthropic move toward increasingly ambitious infrastructure plans and potentially public-market scrutiny. Anthropic’s prospective IPO could offer investors an unusually detailed look at the economics of a frontier AI company, while Nvidia’s future financial disclosures will show how much capital it continues to commit to customers and infrastructure projects. If AI companies can maintain rapid revenue growth while reducing the cost of inference, the current infrastructure buildout could prove justified.
If growth slows while data-center capacity continues expanding rapidly, however, the industry could face an extended period of overcapacity and weaker returns. Nvidia’s decision to reduce its OpenAI exposure suggests that even the biggest beneficiaries of the AI boom are becoming more conscious of these risks. The central story for the next phase of AI will therefore be less about whether companies can build bigger models and more about whether the enormous capital being invested in AI infrastructure can ultimately generate durable economic returns.
Get the day’s top stories in your inbox
One concise email. No spam, unsubscribe anytime.
