Get the latest crypto news, updates, and reports by subscribing to our free newsletter.
Giấy phép số 4978/GP-TTĐT do Sở Thông tin và Truyền thông Hà Nội cấp ngày 14 tháng 10 năm 2019 / Giấy phép SĐ, BS GP ICP số 2107/GP-TTĐT do Sở TTTT Hà Nội cấp ngày 13/7/2022.
© 2026 Index.vn
Most Wall Street analysts view Nvidia and Broadcom as deeply undervalued. Nvidia’s median target price of $265 per share implies 50% upside from its $177 share price, while Broadcom’s median target price of $472.50 implies 50% upside from its $314 share price.
Jay Goldberg at Seaport Research takes a different approach. Goldberg has a sell rating on Nvidia, with a target price of $140 per share, implying 21% downside from the current $177 level. At the same time, he recommends buying Broadcom, though his $430 target price is below the Wall Street median.
Nvidia is described as the leading supplier of AI infrastructure, with GPUs that are widely used as AI accelerator chips and a CUDA software platform that is positioned as the industry standard for AI application development. The article notes that more than half of Nvidia’s engineers work on software.
Goldberg’s concerns focus on the structure of Nvidia’s investments and demand dynamics. The company has signed cloud service agreements totaling $27 billion over the next six years to support research and development, which Goldberg characterizes as a form of rebate because Nvidia is effectively renting its own technology back from customers.
The article also cites equity investments totaling $40 billion in customers including Anthropic, CoreWeave, and OpenAI, arguing that this provides those firms with cash flow to purchase AI infrastructure. Goldberg’s view is that this could be interpreted as artificially inflating demand for Nvidia’s products by paying customers to buy its GPUs.
Finally, Goldberg highlights competition from custom silicon, including tensor processing units (TPUs) developed by Alphabet’s Google and by Broadcom. While the article says TPUs have a less mature software ecosystem, it notes they can be more cost-efficient than Nvidia GPUs for certain AI workloads because they remove unnecessary parts of the chip. It adds that several major customers—including Anthropic, OpenAI, and Meta Platforms—use TPUs.
Despite Goldberg’s caution, the article points to recent results and expectations. In the fourth quarter, Nvidia’s adjusted earnings rose 82%, accelerating from the prior quarter. Wall Street expects adjusted earnings to increase 53% annually through the fiscal year ending in January 2028, and the article says the current valuation of 37 times earnings appears cheap.
Broadcom is presented as a supplier of data center networking solutions and application-specific integrated circuits (ASICs) designed to support artificial intelligence. The article also notes that Broadcom earns substantial revenue from legacy product lines, including connectivity, storage, and broadband chips, as well as infrastructure software.
Goldberg’s view is supported by Broadcom’s position in high-end networking and custom AI accelerators. The article states that Broadcom’s Tomahawk switches are the industry standard for scale-out data center networking. It also says Broadcom holds about 60% market share in custom AI accelerators (XPUs), described as an alternative to Nvidia GPUs.
In custom silicon, the article highlights Broadcom’s tensor processing unit for Alphabet and custom solutions for Meta Platforms, ByteDance, OpenAI, and Anthropic. It reports that AI semiconductor sales increased 106% in the first quarter, and CEO Hock Tan expects momentum to accelerate as Broadcom’s custom AI XPUs move into the next phase of deployment among its five customers.
The article notes that Broadcom is growing more slowly than Nvidia due to drag from legacy products. In the first quarter, total revenue increased 29% to $19.3 billion, and earnings increased 28% to $2.05 per diluted share. Management expects revenue growth to accelerate to 46% in the second quarter.
Looking ahead, Wall Street estimates Broadcom’s earnings will increase 66% annually through the fiscal year ending in November 2027. The article concludes that the current valuation of 43 times earnings looks relatively cheap, adding that both AI stocks are attractive at current prices, even though the author prefers Nvidia for its dominant market position.

In brief\n\nBitcoin dropped to about $93,000, falling back below the EMA50 and putting its recent golden cross at risk of invalidation. The global crypto market cap stands at $3.15 trillion, down 2.38% in 24 hours. On Myriad Markets, 82% of the money is betting on Bitcoin pumping to $100K before…