Nvidia fuels the AI boom: Record growth comes with rising risks

Nvidia fuels the AI boom: Record growth comes with rising risks
Where is Nvidia headed?

​Nvidia’s latest report came in stronger than Wall Street expected. Revenue is hitting records, and the company is promising even faster growth. But the longer this rally continues, the more it starts to resemble a bubble.

Records are becoming the norm

In the second quarter of fiscal 2027, Nvidia posted record revenue of $96.2 billion, according to its website. That was 18% higher than in the previous quarter and 106% above the year-ago result. Wall Street had expected roughly $92.2–92.5 billion.

Data centers were once again the main source of growth. Revenue from the segment reached $89 billion, up 117% year over year. The business now accounts for more than 90% of Nvidia’s total sales.

The company does not expect a slowdown in the current quarter either. Nvidia is forecasting revenue of around $108 billion, plus or minus 2%, while analysts had expected roughly $104–105 billion. The outlook does not include any Data Center compute revenue from China.

The outlook for the next fiscal year surprised the market even more. CFO Colette Kress said Nvidia expects revenue to grow by around 70%, compared with analysts’ forecasts of roughly 44–45%.

After the earnings release, Nvidia shares initially fell but then reversed course and climbed to around $222 in after-hours trading, gaining about 6% from the closing price.

Money is creating its own demand

Nvidia is increasingly acting not only as a chip supplier but also as a participant in the projects that buy those chips. The company provides financial support to data center operators and AI developers, helping them build new infrastructure faster.

In August, Nvidia teamed up with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR to establish platforms for financing AI infrastructure. The initiative aims to mobilize more than $500 billion in third-party capital over time. Some of that money is expected to go toward data centers that will use Nvidia hardware.

Another example is the OpenAI and SB Energy project in Ohio. The plan involves a data center with capacity of up to 8 GW and potential investment of as much as $150 billion. Nvidia agreed to support the project and supply it with its computing systems.

At the same time, the company continues to take stakes in AI projects. Through the end of fiscal 2027, Nvidia has already committed around $18 billion to equity investments in other companies. Against that backdrop, reports emerged of a possible $12.9 billion acquisition of Hugging Face, a platform that hosts open-source AI models and datasets.

The price of endless growth

The more money Nvidia and its partners pour into AI infrastructure, the harder it becomes to maintain the same pace of growth. Demand is already running into component shortages. CFO Colette Kress said directly that the company could sell more if it had access to additional supply.

The biggest problem is memory used in AI accelerators. In just one quarter, Nvidia’s commitments to suppliers jumped from $119 billion to $279 billion, mainly because of memory purchases. Rising costs are already putting pressure on profitability: gross margin could fall from the current 75% to 71–72% in the fourth quarter.

At the same time, Nvidia’s largest customers are trying to reduce their dependence on its chips. Google, Amazon and other technology giants are developing their own accelerators, while OpenAI has already introduced its own processor. Over the long term, part of the multibillion-dollar AI budgets could bypass Nvidia.

Demand remains enormous, but sustaining it requires the industry to keep investing more and more money. Nvidia’s latest report shows that the AI bubble is not deflating yet — on the contrary, revenue, spending, investment and the bets on further expansion are all growing together.

This material may contain third-party opinions, none of the data and information on this webpage constitutes investment advice according to our Disclaimer. While we adhere to strict Editorial Integrity, this post may contain references to products from our partners.