| Ticker | Status | Jurisdiction | Filing Date | CP Start | CP End | CP Loss | Deadline |
|---|
| Ticker | Case Name | Status | CP Start | CP End | Deadline | Settlement Amt |
|---|
| Ticker | Name | Date | Analyst Firm | Up/Down | Target ($) | Rating Change | Rating Current |
|---|
BofA Securities analyst Vivek Arya highlighted key considerations for the U.S. semiconductor sector amid the artificial intelligence frenzy and potential concerns about a bubble.
The analyst noted that large-scale AI infrastructure deployments carry inherent risks of overbuilding, as companies compete to protect existing market positions or establish new revenue streams, making it challenging to perfectly match capacity to future demand.
• NVDA is surging to new heights today. Check it out here.
He identified four key metrics to monitor, namely:
• AI capacity utilization
• Cloud capital expenditure intentions
• Valuations of leading AI stocks
• Financing environment, particularly the direction of the U.S. Federal Reserve rates.
Arya remains cautiously optimistic on AI chip leaders, including NVIDIA Corp (NASDAQ:NVDA), Broadcom Inc (NASDAQ:AVGO), Advanced Micro Devices Inc (NASDAQ:AMD), and Credo Technology Group (NASDAQ:CRDO), along with their semiconductor capital equipment (semicap), memory, optical and foundry peers.
Also Read: Big Tech Earnings To Validate AI Spending Surge
The analyst argued that the current AI buildout differs fundamentally from the dot-com era. AI computing power sees high utilization, unlike the underused “dark fiber” of 2000, and adoption is frictionless with OpenAI projected to reach one billion users in roughly three years, compared to the eight to 13 years Meta Platforms Inc.’s (NASDAQ:META) Facebook and Alphabet Inc.’s (NASDAQ:GOOGL)(NASDAQ:GOOG) Google required, he noted.
The capex intentions of top cloud service providers remain well-supported by operating cash flow rather than debt, and U.S. interest rates are more likely to decline than rise, contrasting with prior market crashes, Arya noted. Valuations also differ sharply: Nvidia trades at a 29 times calendar 2026 PE, well below its EPS growth rate, unlike dot-com leaders that exceeded 100 times PE, the analyst said.
He emphasized that practical limits — such as power, data center space and water — will likely constrain AI infrastructure buildouts more than the ambitions of disruptive AI companies.
AI chip valuations remain far below the theoretical total addressable market implied by OpenAI’s long-term rollout, reflecting only 10%–20% of Nvidia’s projected calendar 2030 total addressable market (TAM) of $3–$4 trillion, Arya noted. Multiple cloud service provider ecosystems will compete for limited resources, which should pace AI expansion, the analyst noted.
Regarding geopolitical risk, he assessed exposure to China as lowest for AI vendors like Nvidia, medium for analog vendors, and highest for Electronic Design Automation (EDA)/Intellectual Property (IP) and semicap vendors, with tariffs and rare-earth supply threats potentially impacting operations.
Read Next:
Photo: Shutterstock
