Facing Nvidia's 75% profit margin, companies like AMD are under huge pressure!

```

Nvidia announced an adjusted gross margin of 75.2%, hitting a recent high, but whether this industry-leading profitability can be sustained faces multiple pressures, including supply bottlenecks narrowing, accelerated rise of self-developed chips, and customers' AI investment returns yet to materialize.

Nvidia's latest quarterly report shows that the adjusted gross margin for the quarter ending January this year reached 75.2%, the highest since the second half of 2024. The company also expects to maintain a similar level this quarter. On the demand side, large-scale AI companies' capital expenditure is expected to total around $650 billion this year, about 60% higher than 2025. Nvidia will benefit greatly from this—a point already widely anticipated by the market before the earnings release. Therefore, the real highlight of this round of earnings lies in profitability rather than demand.

The competitive landscape is changing. AMD announced this week a "tens of billions of dollars" data center processor supply agreement with Meta, directly impacting Nvidia's core GPU business; Alphabet's TPU chips and Amazon's self-developed chips are also quickly capturing market share at prices far lower than Nvidia's products. The increasingly appealing price-performance ratio is prompting more customers to seek diversified sourcing.

Nvidia CEO Jensen Huang defines "continuous generational technology leaps" as the core lever for maintaining high profit margins, and remains optimistic about agentic AI driving increased computing power demand. However, whether the massive AI hardware expenditure will bring customers a corresponding scale of commercial returns remains unresolved. This is the biggest variable for the sustainability of Nvidia's high profit margins.

Supply Tightness and Cost Pressure: Potential Hidden Risks to Profit Margin

Nvidia's high profit margin does not come without cost. Rising memory prices are an unavoidable reality, even though Nvidia holds priority in the supply queues of critical components.

According to Bloomberg, Nvidia CFO Colette Kress stated the company has "strategically secured inventory and capacity to meet demand for several quarters and beyond," but expects supply to remain "tight" for the time being.

Leading manufacturers of core components warn that shortages may last until 2027 or even longer. The growth rate of AI hardware demand currently still far outpaces the expansion speed of corresponding infrastructure capacity.

This means that while Nvidia faces cost-side pressure, it must continue supplying enough products to customers. Under this dual squeeze, whether the 75% profit margin can remain stable in the coming quarters is quite uncertain.

Rise of Alternatives: Competitors Compete at Lower Prices

Pricing differences are especially pronounced. According to Bloomberg Industry Research, Google TPU's average unit price ranges from $8,000 to $10,000, while Nvidia's H100 chip sells for over $23,000, and the newer Blackwell systems start at $27,000.

The price gap of more than twice makes diversification of computing power procurement economically attractive.

At the transaction level, the competitive situation is also becoming clearer. Meta signed a "tens of billions of dollars" data center processor supply deal with AMD; last October, OpenAI reached a similar arrangement with AMD. In both deals, AMD additionally offered shares to enhance attractiveness.

For Alphabet, its TPU has undertaken a large computing load from Google Cloud customers and its own AI services, such as Gemini, boosting its share price after news broke. Amazon has won heavyweight client Anthropic thanks to its self-developed chips.

The Puzzle of Returns: Hundreds of Billions in Investment Has Not Yet Translated Into Revenue Growth

Nvidia's data center business recorded $62.3 billion in revenue, with slightly more than half coming from large-scale cloud computing companies. This means Nvidia's high profit margin largely depends on these clients' sustained and large-scale procurement willingness.

Jensen Huang expressed optimism for this. "I'm very confident about their cash flow growth, for a simple reason," he said, "We have already seen the inflection point of agentic AI, and the practical value of AI agents in global enterprises is emerging. The massive demand for computing you see comes from this." He further stated, "In the AI era, computing power is revenue."

However, there remains a clear gap between reality and expectation: The massive computing power investment by large-scale cloud enterprises has not yet turned into visible revenue returns sufficient to justify the investment.

If these returns are delayed, the market's willingness to continue paying for high-premium chips will be tested. Nvidia's industry-leading profit margin will then become the first variable to be pressured.

Nvidia's Moat: Arguments of Versatility and Energy Efficiency

Jensen Huang once again explained this logic in Wednesday's analyst call: Compared with Google and Amazon's custom chips, Nvidia GPUs can handle a wider range of AI tasks, not limited to model training or "inference" (i.e., running already built AI models) scenarios.

Against the backdrop of increasingly tight energy supply, Nvidia's progress in power consumption optimization also constitutes a differentiating advantage.

Regarding the profit margin issue, Huang provided his core logic: "Our most important lever for gross margin is actually delivering generational leaps to customers continuously."

Huang also stated, "We also love CPUs," emphasizing that Nvidia's CPU products will surpass competitors in data center scenarios and may become "one of the world's largest CPU manufacturers."

Risk Disclaimer and TermsThe market is risky, and investments must be made with caution. This article does not constitute personal investment advice, nor does it take into account individual users' unique investment objectives, financial circumstances, or needs. Users should consider whether any opinions, viewpoints, or conclusions in this article suit their specific situation. Investing accordingly is at your own responsibility. ```