Stay updated with the latest stock market news and insights
Weekly insights + SMS alerts
Bank of America says the race to build artificial-intelligence agents is the next big driver of chip demand. The bank now expects the market for AI data-center systems to reach $2.2 trillion by 2030, up from a prior forecast of $1.8 trillion.
That implies growth of about 40% a year from 2026, compared with 33% in the old model.
The question for investors is which chip stocks stand to capture the most of it as the year enters its final quarter.
In a note shared to clients on Wednesday, Bank of America's analyst Vivek Arya named five top semiconductor picks with near-term catalysts:
The bank also likes Advanced Micro Devices Inc. (AMD), Applied Materials Inc. (AMAT), Analog Devices Inc. (ADI) and ON Semiconductor Corp. (ON).
Timing is part of the call.
From 2010 to 2025, the fourth and first quarters were the two best seasons to own chip stocks.
In those quarters, chip stocks beat the S&P 500 by a median of 3 to 5 percentage points.
Valuation is the other part.
The PHLX Semiconductor Index, a basket of the largest U.S.-listed chipmakers, trades at 21 times forward earnings. In other words, investors pay $21 for every dollar of profit expected over the next 12 months.
The index's median multiple since ChatGPT launched is 24 times. As a result, the index trades about 12% below that level, while the bank projects earnings growth of 40% or more.
AI agents are programs that carry out multi-step tasks on their own, such as writing and testing code.
According to the bank, a single agent workflow can generate 10 to 1,000 times more tokens, the units of text an AI model processes, than a chat exchange.
More tokens means more computing.
Arya said agent adoption, competition between AI labs and tight chip supply "will likely keep AI spending robust for the next several years."
Meanwhile, Bank of America expects the biggest U.S. and Chinese cloud providers, plus newer AI cloud firms, to spend about $1 trillion on capital projects in 2026, roughly double last year.
That figure rises to $1.4 trillion in 2027.
Nvidia's catalysts are larger share buybacks and several GTC developer conferences. The bank expects the company to keep roughly 70% or more of the AI accelerator market over time.
Intel benefits because AI agents are reviving demand for central processing units, the general-purpose chips that coordinate AI workloads.
The bank sees the server CPU market growing from about $61 billion in 2026 to $210 billion in 2030. It also flags a possible customer win for Intel's contract manufacturing business.
Micron's catalyst is the start of share buybacks in December. The memory maker also reports fiscal fourth-quarter results after the close on Wednesday.
Marvell holds its Analyst Day on Oct. 6. Bank of America is watching the ramp of its custom AI processors, chips designed for individual cloud customers.
Lam Research makes the equipment that etches and deposits material on silicon wafers.
The bank expects it to gain market share as industry spending on wafer equipment rises from $156 billion in 2026 to $360 billion in 2030.
Real-time alerts on price moves, news, and trading opportunities.
Join 20,000+ investors. No spam, ever.
