Nvidia Forecasts Continued Rapid Growth as AI Spending Boom Accelerates

SANTA CLARA, Calif. — Nvidia reported second-quarter fiscal 2027 revenue of $96.2 billion, up 106% from a year earlier and 18% from the prior quarter, the company said in its earnings release Wednesday. Data Center revenue reached $89.0 billion, up 117% year over year, driven largely by Blackwell Ultra deployments at hyperscale customers. GAAP and non-GAAP gross margins were both 75.0%, and non-GAAP earnings per share were $2.22, above the $2.09 analysts had expected, according to LSEG.

Q3 Guidance and Long-Term Forecast

For the third quarter, Nvidia guided to revenue of $108.0 billion, plus or minus 2%, with non-GAAP gross margin near 74%. The company said this outlook assumes no Data Center compute revenue from China, citing government export restrictions; Hopper chip shipments accounted for less than 1% of Data Center revenue in the quarter just reported.

Beyond that near-term guidance, Nvidia offered an unusual longer-term projection, telling investors it expects revenue to grow roughly 70% in fiscal 2028. That figure is a company forecast rather than a formal quarterly guide, and executives framed it as reflecting visibility into customer commitments rather than a guarantee.

CEO Jensen Huang attributed the growth to broadening AI demand beyond a handful of early customers. "AI has reached its inflection point. It's doing useful work. Its tokens are productive and profitable. Now, compute is revenue," Huang said, adding that "today, we have a golden age of new AI labs and startups, multiple frontier labs scaling in parallel, a thriving open-model ecosystem and physical AI coming online."

Vera Rubin Platform and Supply Commitments

Nvidia said its next-generation Vera Rubin platform is now in full production, with racks running at hyperscalers and neocloud providers. Amazon Web Services said it would purchase 2 million Nvidia GPUs alongside Nvidia's new Vera CPU, with some units integrated with Rubin. The company said its supply commitments, largely tied to memory for the Rubin ramp, have grown to $279 billion.

Margins, Competition, and Market Reaction

Chief Financial Officer Colette Kress said gross margin is expected to decline through the year, bottoming at 71% to 72% in the fiscal fourth quarter, partly due to memory costs. "We want to be direct about this, rather than let it linger as an open question," Kress said. "Memory scarcity today is being driven in large part by the AI buildout itself."

Nvidia continues to face competition from AMD's MI350 and MI400 accelerators and from custom AI chips developed in-house by major cloud providers, though analysts have said it remains too early to gauge how much share those alternatives will capture.

Nvidia shares fell in Wednesday's regular session before recovering in after-hours trading and rising further in premarket trading Thursday, gaining as much as 6% to 7.6% on the guidance, according to CNBC and Fortune. The results suggest hyperscaler and enterprise AI infrastructure spending remains strong, though Nvidia's own forecasts are not independent confirmation that industry-wide capital spending will continue at its current pace.