Intel got a rare vote of confidence from Wall Street. Bank of America double-upgraded Intel to Buy from Underperform on June 11, 2026, lifting its price target to $135 from $96 and skipping the usual intermediate Hold rating entirely. Intel stock jumped about 9%, helping push the iShares Semiconductor ETF up more than 8%, and the move was part of a broader bullish call across the entire CPU sector.
What did Bank of America actually say?
A "double upgrade," jumping two rungs from Underperform straight to Buy without stopping at Hold, is unusual and signals a sharp change of mind. The bigger story behind it is a market BofA believes is rapidly forming. Analyst Vivek Arya raised the firm's 2030 estimate for the server-CPU market to more than $170 billion, up from $125 billion, arguing that "agentic AI" is "a powerful demand accelerant that expands the CPU opportunity and lifts both x86 incumbents and ARM challengers." That implies nearly five times growth and a 37% compound annual growth rate from 2025 to 2030.
On that thesis, BofA lifted targets across the group: AMD to $560 from $500 on higher CPU and GPU estimates, Arm to $335 from $245 on greater long-term chiplet potential, and Intel to Buy with the $135 target.
Why would an AI boom need more CPUs, not just GPUs?
This is the counterintuitive core of the call. An AI surge usually means demand for GPUs, the chips that train and run models, not CPUs. The agentic-AI argument flips that. As AI shifts from one-shot chatbot answers to autonomous agents that plan, call tools and chain together many steps, much more of the surrounding work runs on CPUs: orchestrating those steps, preparing and moving data, serving requests and coordinating between models and tools. Each AI agent spins up a cascade of conventional computing around the GPU's narrow job.
That is why server CPUs have become the component most likely to bottleneck an AI deployment, a shortage BofA and others describe as structural rather than cyclical. When a component becomes the constraint on everyone's AI plans, its addressable market expands, which is the mechanical basis for jumping the TAM estimate from $125 billion to $170 billion.
What does it mean for Intel specifically?
For Intel, the upgrade rests on two engines. BofA now sees earnings power above $6 a share by 2030, up from a prior estimate of $3 to $4, with Intel's data-center and AI business reaching $43.7 billion on roughly a 25% share of that $170 billion server-CPU market. The second engine is Intel Foundry, its contract chip-manufacturing arm, where BofA models revenue climbing from $1.1 billion in 2026 to $47.1 billion by 2030, driven by wafers, advanced packaging, Apple-related opportunities, MediaTek TPU work and Terafab engagements. In short, the bank is betting Intel benefits from AI twice, once by selling CPUs and once by building other companies' chips.
How do AMD and Arm fit in?
The call is deliberately not Intel-only, because the thesis is that a bigger CPU pie lifts the whole table. AMD, the resurgent x86 rival whose stock has roughly doubled this year, got its target raised to $560 on stronger CPU and GPU estimates. Arm, whose architecture now powers a large share of hyperscaler servers, was lifted to $335 on its long-term chiplet and licensing potential. BofA's framing is that agentic AI rewards both the x86 incumbents and the Arm challengers at once, rather than picking a single winner.
What is the catch?
The enthusiasm arrives after an enormous run. Intel has more than tripled in value since the start of the year, so a great deal of the turnaround story is already priced in, and double upgrades sometimes land closer to a sentiment peak than to the start of one. The foundry projection in particular assumes years of near-flawless execution against TSMC, an area where Intel has repeatedly stumbled, and Arm's push into the data center with its own CPU complicates the tidy "everyone wins" narrative by turning a supplier into a competitor. The thesis is genuinely compelling, but the bar Intel now has to clear to justify a $135 stock is considerably higher than it was at $96.
This article is not investment advice.
Frequently Asked Questions
Why is Intel stock up today?
Bank of America double-upgraded Intel to Buy from Underperform on June 11, 2026, and raised its price target to $135 from $96, citing Intel's opportunity in both CPUs and its foundry business. The stock rose about 9%, and the broader semiconductor ETF gained more than 8%.
What is a "double upgrade"?
A double upgrade is when an analyst raises a stock two rating levels at once, here from Underperform straight to Buy, skipping the intermediate Hold. It is relatively rare and signals a significant reversal in the analyst's view.
What is the $170 billion AI CPU thesis?
Bank of America raised its estimate of the 2030 server-CPU market to more than $170 billion from $125 billion, arguing that agentic AI dramatically increases CPU demand around GPU workloads. That implies nearly five times growth and a 37% annual growth rate from 2025 to 2030.
Did BofA upgrade other chip stocks?
Yes. On the same thesis, BofA raised AMD's price target to $560 from $500 and Arm's to $335 from $245, alongside the Intel double upgrade, arguing that expanding CPU demand lifts both x86 incumbents and Arm-based challengers.
ⓒ 2026 TECHTIMES.com All rights reserved. Do not reproduce without permission.
