Nvidia Reports Record Quarter, Huang Concedes China Market

Nvidia delivered an 85% year-over-year revenue increase to $81.62B, beat estimates, and announced an $80B share buyback plus a 2400% dividend hike, according to Seeking Alpha. The company guided fiscal Q2 revenue to $91B, plus or minus 2%, and told investors it does not expect any revenue from China for the period, per Seeking Alpha. CEO Jensen Huang told CNBC's Sara Eisen that Nvidia has "largely conceded" China's AI chip market to Huawei, as reported by Seeking Alpha. Seeking Alpha also reports SpaceX's S-1 shows $11.4B in Starlink revenue and a $6.4B operating loss for its AI segment, and that PepsiCo plans selective price increases on single-serve chips.
What happened
Nvidia delivered an 85% year-over-year revenue surge to $81.62B, beat analyst estimates, and unveiled an $80B share buyback plus a 2400% dividend increase, according to Seeking Alpha. The company guided fiscal second-quarter revenue to $91B, plus or minus 2%, and, per Seeking Alpha, told investors it does not expect any revenue from China in the period. CEO Jensen Huang told CNBC's Sara Eisen, "We've really largely conceded that market to them," referring to Huawei, as reported by Seeking Alpha.
Seeking Alpha's coverage of SpaceX's S-1 shows Starlink generated $11.4B in revenue and $4.4B in operating income, while the filing reported a $6.4B operating loss for the company's AI segment. Seeking Alpha also reports that PepsiCo plans selective price increases on single-serve chips after holding prices steady for 15 years.
Editorial analysis - technical context
Companies that build or procure AI infrastructure rely on access to advanced accelerators and compatible ecosystems. Export controls and market bifurcation reduce available vendor options in some regions, complicating capacity planning and hardware refresh cycles for organizations that need the latest-generation accelerators.
Industry context
The combination of Nvidia's sizable buyback and dividend actions alongside an 85% revenue jump signals strong near-term monetization of accelerator demand, per Seeking Alpha's earnings summary. Reporting on Jensen Huang's CNBC comments frames a geographic split where local suppliers, led by Huawei in China, are capturing workloads that Nvidia is currently unable to serve because of export restrictions. Industry observers have previously noted that regional supplier strength often accelerates local software and hardware co-design, which can widen compatibility gaps between ecosystems.
SpaceX's S-1, as covered by Seeking Alpha, recasts the company as a broader constellation of businesses where Starlink is a material revenue engine. The filing's numbers-$11.4B revenue and $4.4B operating income for Starlink-contrast with a multi-billion-dollar operating loss in the AI segment, highlighting divergent unit economics inside a single corporate filing.
What to watch
- •Nvidia's commentary and regulatory filings for updates on export approvals and any change to China revenue guidance.
- •Huawei and other Chinese AI-accelerator vendors' reported shipments and ecosystem partnerships for indicators of local market traction.
- •SpaceX's S-1 follow-ons and investor presentations for clarity on how Starlink profits map to any public valuation and how the AI segment's losses are being addressed.
For practitioners
Organizations planning capacity purchases should treat regional supply constraints as an operational variable. Observers buying or benchmarking accelerators should monitor vendor compatibility, software stack maturity, and regulatory developments because these factors materially affect procurement timelines and total cost of ownership.
Scoring Rationale
Nvidia's blockbuster quarter and Huang's concession about China alter the supply landscape for AI infrastructure, a notable development for practitioners. SpaceX's S-1 adds secondary importance for cloud and connectivity economics.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems


