Why Microsoft Could Surprise and Hit $4 Trillion Before NVIDIA Does

AI & Compute

NVIDIA posts $115.84B as Microsoft’s contracted backlog hits $677B

NVIDIA’s $115.84 billion revenue and $130 billion outlook give it the edge in a near-even $4 trillion race with Microsoft, whose contracted backlog reached $677 billion.

By
Nathan Brooks
Filed
Channel
AI & Compute
Read
4 min read

By Nathan Brooks

NVIDIA reported quarterly revenue of $115.84 billion, up 103.2%, while Microsoft reported $92.28 billion, up 18.8%. Their market values stood at roughly $3.75 trillion and $3.76 trillion, leaving the two companies nearly tied in the race to $4 trillion.

Microsoft earns from AI through subscriptions and contracts; NVIDIA sells the hardware that runs it. The contrast sets up two different tests: whether Azure can keep converting demand into recognized revenue, and whether accelerator demand can retain its current momentum.

What do the earnings show?

Azure grew 48%, and Microsoft’s commercial remaining performance obligation rose 73% to $677 billion. That figure represents contracted revenue not yet recognized. Because Microsoft has already signed those contracts, the backlog does not depend on next quarter’s chip orders, according to the market analysis.

Management said customer demand still exceeds supply. The counterweight was Windows OEM and Devices, whose revenue fell 22% as PC demand weakened.

NVIDIA’s growth came from Vera Rubin systems, which generated nearly $20 billion in the first quarter of production shipments. Interim CEO Colette Kress said Rubin was “off to an explosive start.” The quote was colorful; the shipment figure did the harder work.

What must each company deliver next quarter?

  • Microsoft guides to $96 billion to $97.1 billion in revenue and Azure growth of 49% to 50% next quarter. The company plans roughly $170 billion of calendar-year capital expenditure, putting that spending behind the Azure target.
  • NVIDIA guides to quarterly revenue of $130 billion, plus or minus 3%. Management expects gross margin to bottom at 71% to 72%, and its outlook contains no China data-center compute revenue.

Microsoft also designs its own Maia accelerators, creating a route to pressure NVIDIA’s pricing. Foundry now has more than 120,000 customers, which the analysis treats as evidence of broad enterprise adoption. Its Maia 300 targets 40% more tokens per watt than leading GPU systems, although that remains a product target rather than a reported benchmark result.

CEO Satya Nadella said Microsoft’s advantage comes from “adding value at every layer of the stack.” Azure and Copilot remain the named growth engines.

How concentrated is NVIDIA’s growth?

NVIDIA also carries sharper company-specific risks. Jensen Huang resigned, and the board hired an outside firm to search for a permanent CEO. Three direct customers contributed 14%, 11% and 11% of revenue, concentrating a large share of sales in a small buyer group.

Management nevertheless raised its fiscal 2028 growth outlook to about 75%. That figure is forward guidance, not a completed delivery result. The Vera Rubin number reflects production shipments already reported; the 75% forecast does not.

Can Microsoft overtake NVIDIA?

To reach $4 trillion, Microsoft needs an increase of about 6.4%; NVIDIA needs about 6.7%. Those narrow gaps make volatility a plausible decider. NVIDIA’s beta is 2.247, compared with 1.045 for Microsoft, and its stock fell 27.4% in one month. A rebound could arrive nearly as quickly.

The underlying analysis gives NVIDIA the edge, citing momentum and the $130 billion revenue outlook. Microsoft remains a lower-volatility sleeper: its $677 billion backlog could carry the shares past NVIDIA if chip stocks stay shaky, while another sharp decline in NVIDIA would open the path. A strong permanent CEO appointment at NVIDIA would reinforce the contrary case.

Which commercial risks matter most?

For Microsoft, the test is whether roughly $170 billion of calendar-year capital expenditure can produce the guided Azure growth of 49% to 50%. For NVIDIA, investors must check whether gross margin actually bottoms at 71% to 72%, as management expects, rather than treating the forecast as settled.

Commerce Department export rules add another variable. NVIDIA’s outlook already contains no China data-center compute, leaving that market outside the near-term forecast. Microsoft’s custom-chip strategy offers a different form of protection by reducing reliance on leading GPUs, if Maia 300 meets its stated efficiency target.

Next quarter will not settle the entire AI market, but it should show whether Azure demand can absorb Microsoft’s spending and whether Vera Rubin can defend NVIDIA’s margin against custom silicon, export controls and concentrated customers. One of the two market values will move first; the more useful signal will be what revenue and margins do afterward.

Source: 247wallst.com

Share this article:

More from Nathan Brooks

Nathan Brooks

Show full bio

Senior reporter covering people, launches and day-to-day practice in Semiconductors, chip manufacturing, hardware, AI compute at Chip Dispatch.

1,092 articles

Related articles

« Previous articleNext article »