New York, April 24, 2026, 11:48 EDT
- Nvidia shares rose about 4.8% Friday morning to $209.15, putting the AI chipmaker back near its 52-week high as investors looked toward next month’s earnings.
- The company reports fiscal first-quarter results on May 20 after guiding for $78 billion in revenue, plus or minus 2%, with no China data-center compute revenue assumed.
- The main question is no longer whether demand is strong. It is how much of the next AI spending wave stays with Nvidia as Google, Amazon and Broadcom push deeper into custom chips.
Nvidia shares climbed toward a record on Friday, reviving the $300 stock debate just weeks before the company’s May earnings report and as investors look for clearer signs that its next-generation Vera Rubin systems can keep the AI trade centered on one company.
The stock rose $9.51 to $209.15 in late morning New York trading, after touching $209.67. Nvidia’s market value stood at about $5.12 trillion, leaving it within reach of the stock’s 52-week high of $212.19 but still far from the $300 level that would require another steep move in an already giant company.
That matters now because the chip rally is broadening. A 24/7 Wall St. analysis published Friday said Nvidia was up only 7% for the year through Thursday’s close, while Marvell, Micron, Advanced Micro Devices, Taiwan Semiconductor Manufacturing and Broadcom had posted much stronger 2026 gains as investors spread their AI exposure across memory, networking, foundry and custom silicon suppliers.
The earnings bar is still high. Nvidia reported record fiscal fourth-quarter revenue of $68.1 billion, up 73% from a year earlier, while data-center revenue rose 75% to $62.3 billion. Chief Executive Jensen Huang said then that “enterprise adoption of agents is skyrocketing,” referring to AI systems that can plan and carry out tasks, not just answer prompts. NVIDIA Newsroom
For the April quarter, Nvidia forecast revenue of $78 billion, plus or minus 2%, and expected non-GAAP gross margin of 75%. The same outlook said Nvidia was not assuming any China data-center compute revenue, a reminder that export controls remain a live risk even as demand elsewhere stays heavy.
Vera Rubin is the next test. Nvidia says the platform can cut inference token cost by up to 10 times versus Blackwell; inference is the process of running a trained AI model to produce answers, code, images or other outputs. Barron’s framed the near-term stock question this week around whether large cloud customers show enough demand for Vera Rubin to push the shares past prior highs.
Google is both a customer and a rival. At Google Cloud Next, the company announced TPU 8t for training and TPU 8i for inference; TPUs, or tensor processing units, are custom chips built for AI work. Google also said it would be among the first cloud providers to offer instances based on Nvidia’s Vera Rubin platform when it becomes available later this year.
Nvidia and Google described that split as cooperation rather than a clean fight. Google Cloud executive Mark Lohmeyer said combining Google’s infrastructure with Nvidia’s platforms would give customers flexibility to train, tune and serve models while optimizing for performance, cost and sustainability.
Amazon is pressing from another direction. In his 2025 shareholder letter, CEO Andy Jassy said Amazon’s chips business, including Graviton, Trainium and Nitro, had an annual revenue run rate above $20 billion and was growing at triple-digit rates year over year. He also said Trainium3, which began shipping at the start of 2026, was nearly fully subscribed.
Broadcom, meanwhile, has turned custom AI chips into a larger part of the competitive story. CEO Hock Tan said Broadcom’s AI revenue reached $8.4 billion in its fiscal first quarter, up 106% from a year earlier, driven by custom AI accelerators and AI networking. The company expects AI semiconductor revenue of $10.7 billion in the second quarter.
That is the risk paragraph for Nvidia bulls. The stock can still rise on another clean earnings report, but the margin and multiple story could get harder if hyperscalers shift more workloads to in-house chips or use custom silicon suppliers to lower costs. Nvidia’s size also works against it: adding 50% to the share price would require trillions of dollars of additional market value, not just another strong quarter.
Still, the ecosystem has not moved away from Nvidia. Futurum analyst Brendan Burke wrote Friday that Google’s plan to offer both its own TPUs and Nvidia’s Vera Rubin GPUs turns the market from a “zero-sum chip war” into a platform contest for AI workload share. That is the cleaner read for now: Nvidia remains the default platform, but it is no longer the only way investors are trying to own the AI buildout. Futurumgroup
