NVIDIA Corp. (NASDAQ: NVDA) CEO Jensen Huang stated that demand for artificial intelligence computing is accelerating as companies race to build new infrastructure. He described the current cycle as a flywheel that is "really, really flying."

Speaking with CNBC’s Jim Cramer at Dreamforce in San Francisco, Huang highlighted the rapid construction of NVIDIA-powered "AI factories" over the past six months.

AI Factory Economics

Huang estimated that a 1-gigawatt NVIDIA AI factory costs about $50 billion to $60 billion to build. He said this infrastructure can generate about $50 billion in annual rental revenue. This implies a return on invested capital of roughly one year under current market conditions.

He noted that NVIDIA infrastructure can remain useful for more than five or six years. Meanwhile, AI usage continues to climb. The rate of token generation has increased 25-fold in less than a year. Open models now account for nearly 70% of generated tokens, up from about 30%.

New Demand Drivers

Huang pushed back against concerns that AI safety efforts could derail industry growth. He said thousands of companies worldwide are working on AI safety, security and guardrails.

He expects testing to become a third major category of AI infrastructure alongside training and inference. Large testing environments will require additional data centers, creating another source of compute demand.

Rental Price Surge

Huang pointed to rising rental prices as evidence of tight demand. NVIDIA Grace Blackwell systems can now rent for about $16 per GPU hour, compared with roughly $5 under some contracts signed a year ago.

As older contracts expire, cloud providers are raising prices. This boosts revenue forecasts and returns on invested capital, prompting customers to buy more equipment and build more infrastructure.

What the Numbers Show

The tripling of GPU rental prices from $5 to $16 per hour directly supports Huang’s claim of an accelerating spending cycle. Higher rental costs improve the return on invested capital for existing infrastructure, which in turn incentivizes further capital expenditure on new data centers. This creates a self-reinforcing loop where higher utilization rates drive higher revenues, which justify additional infrastructure builds.

Stock Performance

NVIDIA stock rose in Wednesday premarket trading as technology stocks gained alongside U.S. equity futures. Nasdaq futures rose 0.39%, while S&P 500 futures gained 0.16%.

NVIDIA is trading 2.6% below its 20-day simple moving average of $218.81. The stock is roughly in line with its 50-day SMA of $213.24. It trades 7.7% above its 200-day SMA of $197.79. The 20-day SMA remains above the 50-day average, while the 50-day remains above the 200-day.

Momentum is neutral. NVIDIA’s relative strength index stands at 45.60. Resistance sits near $214, close to the 50-day SMA. Support is at $190. The stock’s 52-week range is $164.27 to $236.54.

Analyst Outlook

NVIDIA carries a Buy consensus rating with an average price forecast of $348.22. Piper Sandler initiated coverage with an Overweight rating and a $300 price forecast on Sept. 10. Rosenblatt maintained a Buy rating and $390 forecast on Sept. 4. Needham also maintained a Buy rating and $300 forecast that day.

ETF Exposure

NVIDIA accounts for 9.94% of the Xtrackers Net Zero Pathway Paris Aligned U.S. Equity ETF (NYSE: USNZ). It holds 9.79% of the First Trust Innovation Leaders ETF (NYSE: ILDR) and 9.73% of the Franklin Focused Dynamic Growth ETF (NASDAQ: FFOG).

Large inflows or outflows from these funds can translate into additional buying or selling of NVIDIA shares.

NVIDIA Corp (NASDAQ: NVDA) demonstrates a distinct valuation profile compared to its semiconductor peers, trading at a Price-to-Earnings (P/E) multiple of 26.82, which is below the industry average of 54.06. Despite this relatively lower earnings multiple, the company commands premium multiples on book and sales value, reflecting strong market sentiment around its artificial intelligence and data center positioning.

Financial Performance vs Industry

NVIDIA’s financial metrics indicate robust profitability and growth relative to the broader Semiconductors & Semiconductor Equipment sector. The company reported an EBITDA of $72.86 billion, substantially higher than the industry average of $4.38 billion. Similarly, gross profit reached $72.14 billion, outpacing the peer average of $5.02 billion.

The company’s Return on Equity (ROE) of 28.12% exceeds the sector average of 7.19% by more than three times, highlighting efficient capital utilization. Revenue growth accelerated to 105.85%, nearly double the industry’s average expansion rate of 54.18%.

Balance Sheet Strength

NVIDIA maintains a conservative leverage profile with a debt-to-equity ratio of 0.17. This figure is lower than that of its top four peers, indicating a stronger balance sheet position and reduced financial risk relative to competitors.

What the Numbers Show

NVIDIA’s valuation divergence is notable: it trades at a significant discount to peers on an earnings basis (P/E 26.82 vs 54.06) yet carries a premium on sales (P/S 17.09 vs 11.83) and book value (P/B 22.37 vs 8.65). This suggests investors are pricing in high future margin expansion rather than current earnings yield, supported by the company’s 105.85% revenue growth and 28.12% ROE, both of which far outstrip industry averages.