BlockBeats News, July 9th. A recent report from Bank of America suggested that concerns in the market about NVIDIA facing a cost increase in High Bandwidth Memory (HBM) and ASIC custom chip competition have been amplified. The report underestimated the company's strong pricing power, supply chain advantage, and ecosystem moat. Bank of America believes that the upcoming Rubin AI platform is expected to offset the HBM cost growth with a higher selling price, and NVIDIA's supply chain commitment of around $119 billion will further solidify its cost advantage. The company's gross margin is expected to remain at around the midpoint level of about 75%.

Addressing the market's concerns about ASIC replacement risks, Bank of America pointed out that Google's TPU has been developed for over a decade, but during the same period, NVIDIA's GPU business revenue has grown by about 700 times, demonstrating that custom chips have not weakened the GPU's dominant position in AI training and inference. The bank expects NVIDIA to continue to account for 65% to 70% of global hyperscale cloud providers' AI infrastructure spending in the future.

In terms of valuation, Bank of America believes NVIDIA's current forward P/E ratio is around 18.7 times, only half of the historical average of about 37 times over the past decade, and close to a nearly 11-year low. Pessimistic expectations have been fully priced in. With the advancement of the Rubin platform and the approaching August earnings report, Bank of America expects NVIDIA to once again prove its product competitiveness and profitability, driving the market to reassign a valuation premium.