NVIDIA Q1 FY2027 Earnings: Record $81.6B Revenue as Jensen Huang Declares Agentic AI Has Arrived

NVIDIA Q1 FY2027 earnings hit $81.6 billion in revenue, up 85% year-over-year and 20% sequentially. Data center revenue reached $75.2 billion, accounting for 92% of total sales and nearly doubling compared to a year ago. Both figures beat Wall Street consensus estimates of $78.86 billion.

The number that matters most, though, came from Jensen Huang’s mouth during the earnings call: “Agentic AI has arrived, doing productive work, generating real value and scaling rapidly across companies and industries.”

This is not a marketing line. It signals a shift in the computational demand curve. Agentic AI systems, which plan and execute multi-step tasks autonomously, require orders of magnitude more sustained compute than simple question-answering. That ongoing demand is exactly what NVIDIA’s GPU business is built to supply.

Why the Stock Fell Anyway

After-hours trading saw NVIDIA shares drop more than 1.5% despite the record results.

This is the expectations treadmill in action. The stock had been priced for perfection, trading at roughly 40x forward earnings at the start of the year. A record quarter that beats estimates by 3% is no longer surprising enough to move the needle up. The market’s response measures not results but surprise, and NVIDIA has made surprise increasingly difficult to manufacture.

The forward P/E has since compressed to around 25x. That re-rating reflects investors recalibrating their assumptions about how long an 85% annual growth rate can hold.

Vera Rubin: The Platform Built for Agentic AI

The earnings release confirmed shipping timelines for the Vera Rubin platform: volume production in Q3 FY2027, ramp beginning in Q4.

Vera Rubin centers on the Vera CPU, the world’s first processor purpose-built for agentic AI architectures, paired with the BlueField-4 STX accelerated storage infrastructure. Integrated with Google Cloud’s A5X instances, the platform targets enterprise AI agent deployments at scale. Huang stated Vera Rubin will deliver an order-of-magnitude improvement in inference performance over Blackwell, with 10x more throughput per watt.

Blackwell remains the current production workhorse. Q2 FY2027 guidance of $91 billion (plus or minus 2%) well exceeded the $86 billion analyst consensus, suggesting the near-term demand pipeline is still intact.

Concentration Risk Is the Real Story

Major cloud providers plan to collectively spend approximately $725 billion on AI infrastructure this year. NVIDIA captures a disproportionate share of that spend.

The upside and the risk are the same thing. Revenue this concentrated in a handful of hyperscaler customers means any meaningful capex pullback from a single large buyer shows up immediately in NVIDIA’s financials. It is a structural vulnerability that the company’s current scale makes harder, not easier, to diversify away from.

The full financial results are available in NVIDIA’s official Q1 FY2027 press release. For live earnings commentary, CNBC’s earnings coverage tracked analyst reactions as they happened.

---

If this was useful, subscribe to the newsletter for weekly AI PM insights and GenAI case studies.