Mizuho Analyst Sees Four AI Tailwinds for Intel Stock. Can the Altera IPO be the Fifth?

Intel has gained a new IPO catalyst just when its stock needs to prove that its remarkable recovery is more than an AI-driven rerating. 

Altera, the programmable-chip company in which Intel still owns a 49% interest, has confidentially submitted a draft IPO registration statement. That could finally give investors a public valuation benchmark for Intel’s $3.2 billion Altera investment. But the arithmetic also reveals something important: Altera is a useful bonus, not the main reason to justify Intel’s roughly $510 billion market value.

Let’s break down Intel’s four operating tailwinds, why Mizuho still reduced its price target to $92, what an Altera IPO could realistically add, and which numbers will decide the next stage of the Intel stock story. More importantly, we will separate what is improving inside Intel from what its current valuation already assumes.

What’s Covered

• Why agentic AI could increase demand for Intel server CPUs
• How supply shortages and the PC refresh cycle are helping Intel
• Why Mizuho reduced its Intel price target despite four tailwinds
• What Altera’s confidential IPO filing means for Intel
• How much Intel’s 49% Altera stake may be worth
• Why Intel Foundry still needs stronger external customer validation
• An original scenario model for understanding Intel’s valuation
• The financial metrics investors should track next

Intel Stock and Altera IPO: The Story in One Table

Intel’s stock closed at $97.14 on September 15 and was indicated near $102 in September 16 pre-market trading. Its market value stood near $510 billion, while the stock had risen about 303% over the previous 52 weeks. That strong performance is central to the debate. Intel is no longer valued like a deeply troubled turnaround. It is being valued as a company that must successfully participate in the AI infrastructure boom.

Why Mizuho Sees Four Major Tailwinds for Intel

Mizuho reduced its Intel price target from $109 to $92 while maintaining a Neutral view. At first glance, that appears strange because the firm also identified four meaningful growth drivers.

The target reduction was not based on a sudden collapse in Intel’s operating outlook. It reflected a concern that investors may apply lower valuation multiples to companies linked to the agentic AI theme, even if their earnings continue to improve.

Here are the four operating tailwinds.

1. Agentic AI Could Increase Demand for Server CPUs

The AI infrastructure discussion has largely focused on GPUs from Nvidia and AMD. However, GPUs do not operate alone.

CPUs manage operating systems, networking, storage, security, data movement and many of the general-purpose tasks surrounding an AI workload. As AI moves from training large models to running millions of agents and inference requests, the supporting CPU requirement can increase.

Think of GPUs as highly specialised chefs preparing one complex dish at enormous speed. CPUs act more like the kitchen manager, assigning tasks, handling incoming orders, coordinating ingredients and ensuring the entire operation keeps running. More AI agents can mean more work for both.

Intel’s latest results provide evidence that this is already affecting demand.

Intel said server revenue benefited from a 48% increase in average prices and a 9% increase in volume. Higher hyperscaler demand was the main volume driver.

This is the strongest part of Intel’s current financial story. It shows that AI demand is improving the economics of Intel’s existing CPU franchise, even though Intel is not the leading supplier of AI training GPUs.

There is still a major risk. AMD reached an estimated 34.5% share of conventional server CPU units in Q2 2026, according to Mercury Research data reported by industry publications. Intel therefore operates in a rapidly expanding market but continues to face market-share pressure.

Intel does not merely need the server market to grow. It must retain enough of that growth for revenue, margins and earnings per share to keep improving.

2. CPU Supply Tightness Is Supporting Prices

Intel reported that demand exceeded available supply during the first half of 2026. It expects shortages involving substrates, memory and other components to continue into 2027. A shortage can help pricing, but it has two sides.

If Intel can increase production while demand remains strong, it can ship more high-margin processors and protect pricing. If shortages prevent shipments for too long, customers may redesign systems around AMD, Arm or custom processors.

Intel’s Q2 figures show that premium product mix was more important than simple price increases. Client processor average prices rose 27%, while server processor average prices rose 48%. Intel said demand-based pricing also contributed, but to a lesser extent.

Premium mix is generally more sustainable than charging more for the same product. However, it also means investors must watch unit growth. Intel’s client processor volume fell 8% even as client revenue increased.

The most favourable outcome is therefore not permanent scarcity. It is a controlled expansion of supply while premium demand remains strong.

3. The PC Refresh Cycle Is Better Than Expected

Intel’s Client Computing and Physical AI Group generated $8.9 billion in Q2 revenue, 13% higher than a year earlier. Corporate device upgrades, AI-enabled PCs and the transition to newer operating systems have supported the refresh cycle.

However, the quality of the growth deserves attention.

Revenue growth came from a richer mix of premium processors and higher prices, not broad unit growth. That is profitable while it lasts, but it is different from a healthy PC market where both prices and shipments are expanding.

Memory and component shortages could extend the refresh cycle by delaying some purchases. They could also make PCs more expensive and reduce overall demand. Investors should therefore avoid treating every supply shortage as automatically favourable.

4. Advanced Packaging and Intel Foundry Offer Long-Term Optionality

Mizuho estimates that Intel’s advanced-packaging revenue could reach $3.5 billion by 2029. It also sees the possibility of roughly $3.5 billion in external foundry revenue, supported by Intel’s 14A manufacturing process.

Advanced packaging connects multiple chips inside one high-performance system. It is becoming increasingly important because AI processors now combine CPUs, GPUs, memory and specialised accelerators.

Intel’s EMIB and Foveros technologies could become valuable even when Intel does not design every chip inside the final product. This is similar to a construction company earning revenue not only by building an entire house, but also by providing the specialised foundation required by several other builders.

The foundry opportunity is real, but current numbers show that it remains an option rather than a fully proven business.

Approximately 95% of Intel Foundry’s Q2 revenue came from other parts of Intel. External revenue represented only about 5% of the segment total.

The operating loss improved from $3.17 billion a year earlier, but Intel still lost $2.09 billion on $5.77 billion of foundry revenue. Intel also acknowledged that substantially all of its foundry activity currently supports internal manufacturing.

That is why external customer announcements matter more than another quarter of higher internal foundry revenue. Intel needs unaffiliated customers to validate manufacturing yields, pricing, service quality and the economic value of its factories.

Why Four Tailwinds Can Still Produce a Lower Intel Price Target

A stock price can be simplified into two major components:

Stock price = earnings per share × the valuation multiple

Intel’s earnings can improve while its stock struggles if investors become less willing to pay a high multiple for those earnings. This is called multiple compression.

Consensus estimates collected by S&P Global indicate adjusted EPS of approximately $1.52 for 2026 and $2.06 for 2027. That represents expected growth of about 36%.

At a price of $97.14, Intel trades at roughly 64 times the 2026 EPS estimate.

Now assume Intel reaches the $2.06 estimate in 2027, but its valuation falls to 45 times earnings:

$2.06 × 45 = $92.70

Intel could therefore grow adjusted EPS by approximately 36%, yet its indicated value would remain near $93 because the earnings multiple fell from roughly 64 to 45.

This simple calculation almost recreates Mizuho’s $92 target. That is the red flag. Intel’s operating momentum may remain positive, but investors have already paid for a large part of the expected recovery.

What Is Altera and Why Is It Going Public?

Altera produces field-programmable gate arrays, commonly called FPGAs. A traditional application-specific chip is designed for a fixed task. An FPGA can be reconfigured after manufacturing. Think of a fixed chip as a completed toy model and an FPGA as a box of electronic building blocks that can be rearranged for different jobs.

This flexibility makes FPGAs useful in data centres, communications equipment, industrial systems, aerospace, defence, robotics and edge AI.

Intel acquired Altera for approximately $16.7 billion in 2015. In 2025, it transferred a 51% controlling interest to Silver Lake for $4.46 billion, valuing Altera at $8.75 billion. Intel retained the remaining 49%.

Altera has now announced that its parent submitted a confidential draft S-1 registration statement. The company has not disclosed the number of shares, price range, updated financial statements or proposed valuation.

Reuters previously reported that Altera could raise more than $2 billion and that CEO Raghib Hussain expected revenue growth in the mid-20% range in 2026. Those figures remain reports and management expectations, not final offering terms.

What Do We Know About Altera’s Financials?

Altera generated $1.54 billion in revenue during 2024.

The $8.75 billion transaction valuation equalled approximately 5.7 times Altera’s 2024 revenue.

The large difference between GAAP and adjusted profitability is one reason the public S-1 will matter. Investors will need updated information on revenue growth, product margins, research spending, customer concentration, cash generation and acquisition-related accounting adjustments.

Until those figures become available, assigning a precise valuation to Altera would involve more speculation than analysis.

How Much Could the Altera IPO Add to Intel’s Value?

Intel reported the carrying value of its 49% Altera interest at $3.2 billion as of June 27, 2026. That implies an accounting equity value of approximately $6.53 billion for the whole company.

The following table shows what Intel’s interest could be worth under different illustrative public equity valuations.

The calculation uses approximately 5.25 billion Intel shares. It ignores taxes, listing expenses, lock-up discounts and any dilution caused by Altera issuing new shares.

It also treats the figures as equity values. Intel’s earlier $8.75 billion transaction figure should not be compared directly without considering Altera’s net debt and the final IPO capital structure.

The main conclusion is clear. Even if Altera receives a $15 billion public equity valuation, Intel’s theoretical uplift over its current carrying value would be about $4.15 billion, or roughly $0.79 per Intel share.

That is useful, but not transformative for a company valued near $510 billion.

Intel’s approximately $5.30 premarket price increase on September 16 represented nearly $28 billion of market value. That was almost nine times the entire carrying value of Intel’s Altera investment. The move therefore cannot be explained by the direct value of the stake alone. It also reflects sentiment, signalling value and broader expectations around Intel’s turnaround.

The Overlooked Altera Connection: It Is Both an Asset and a Customer

Altera’s relationship with Intel does not end with the 49% investment.

Intel provides wafer-manufacturing services to Altera. Intel recognised $181 million of revenue from Altera in Q2 2026 and $320 million during the first half.

This produces an interesting financial loop:

• Intel owns 49% of Altera.
• Altera purchases manufacturing services from Intel.
• A stronger Altera can increase the value of Intel’s investment.
• It can also support Intel Foundry’s external revenue.
• However, dependence on Altera makes Intel Foundry’s external growth look stronger than its fully independent customer traction.

Intel Foundry reported $293 million of external revenue in Q2. Altera contributed $181 million, or approximately 62%, of that amount. After removing Altera, other external customers contributed only around $112 million. That equals less than 2% of total foundry revenue.

This is one of the most important numbers for analysing Intel. The foundry story will become more credible when unrelated customers generate a materially larger share of revenue.

Intel Stock Valuation: Three Possible Scenarios

The following model is not a formal price target. It is a framework showing how earnings and valuation assumptions can produce very different outcomes.

The pressure case does not require Intel’s business to collapse. It only assumes earnings fall below current expectations while the market applies a more ordinary semiconductor valuation.

The execution case assumes Intel reaches the current 2027 consensus estimate. At 45 times earnings, the indicated value is approximately $93, close to Mizuho’s target.

The breakout case requires more than strong CPU demand. Intel would need faster margin improvement, reduced foundry losses, larger advanced-packaging commitments and visible external manufacturing customers.

Altera could add modest asset value in each scenario, but it does not change the central earnings debate.

The Three-Ledger Framework for Understanding Intel

A useful way to analyse Intel is to divide it into three separate ledgers.

Ledger 1: The Earnings Engine

This includes client processors, server CPUs, ASICs and related products. The earnings engine is performing well. Q2 Intel Products revenue increased 28% to $15.1 billion, while operating income reached $4.8 billion. Server pricing and hyperscaler demand were particularly strong. This part of Intel currently supports the valuation.

Ledger 2: The Manufacturing Option

Intel Foundry and advanced packaging create large strategic possibilities. Successful external adoption could turn Intel into a major alternative to TSMC and Samsung, particularly for customers seeking manufacturing capacity in the United States. However, the segment still loses billions of dollars and relies heavily on internal revenue. The option has value, but investors should not treat projected 2029 revenue as if it already exists.

Ledger 3: The Portfolio Side Pocket

Altera, Mobileye and other investments sit outside Intel’s main CPU earnings engine. An Altera listing can make one of these assets easier to value and potentially easier to monetise over time. Still, the direct valuation effect is much smaller than the market value tied to Intel’s CPU and foundry expectations.

This framework helps prevent a common mistake: using one positive development in a small asset to justify the valuation of the entire company.

What Should Intel Investors Track Next?

1. Server Volume, Not Just Pricing: A 48% increase in server average prices is powerful, but sustainable growth should eventually include strong unit expansion. Investors should compare server revenue growth with volume growth every quarter.

2. AMD and Arm Market Share: Intel can grow while losing share if the market expands quickly. That may work temporarily, but long-term valuation requires competitive products and stable customer relationships.

3. Foundry Revenue From Unrelated Customers: The most useful foundry metric is not total segment revenue. It is external revenue after removing related-party business from Altera.

4. Foundry Operating Loss: The Q2 loss improved to $2.09 billion from $3.17 billion. Continued improvement is necessary because high factory losses can absorb the profits produced by Intel’s successful CPU divisions.

5. Gross Margin: Intel’s GAAP gross margin reached 40.4% in Q2, compared with 27.5% a year earlier. The Q3 forecast calls for 41% GAAP and 42% adjusted gross margin. A sustainable move into the mid-40% range would strengthen the earnings case. A reversal would suggest that premium pricing and favourable mix are not translating into durable economics.

6. Altera’s Public S-1: The document should reveal:

• Updated revenue and growth
• GAAP and adjusted margins
• Cash flow and capital expenditure
• Customer concentration
• Exposure to data centres and AI
• Use of IPO proceeds
• Primary versus secondary shares
• Intel’s ownership after the offering
• Manufacturing commitments with Intel
• Restrictions on Intel reducing its stake

These details will matter far more than the confidential submission itself.

Author’s View on Intel Stock

Intel’s operating recovery is real.

Q2 revenue increased 25%, Data Center and AI revenue increased 59%, operating margins improved and demand remains above available supply. Intel also has credible opportunities in advanced packaging, domestic semiconductor manufacturing and AI inference infrastructure.

But the stock price already reflects substantial progress.

At approximately $97, Intel trades near 64 times expected 2026 adjusted earnings, around 9 times trailing revenue and roughly 180 times trailing free cash flow. These are demanding levels for a company whose foundry division still loses more than $2 billion per quarter and whose external foundry revenue remains limited.

The Altera IPO improves visibility and could reveal additional asset value. It also provides a second benefit because Altera is an Intel Foundry customer. However, the direct value of Intel’s 49% interest is too small to meaningfully alter the overall valuation by itself.

Our conclusion is therefore positive on Intel’s business direction but cautious on the price required to participate in it. The current valuation needs Intel to execute across several fronts at the same time: protect CPU demand, improve unit growth, reduce foundry losses, attract independent manufacturing customers and convert higher revenue into per-share earnings after dilution.

That does not make the optimistic case impossible. It makes the margin for error much smaller.

For investors evaluating fresh exposure, the key question is no longer whether Intel is recovering. The evidence suggests that it is. The more important question is whether future earnings can grow faster than the valuation multiple contracts.

At present, Intel looks like a better company than it did a year ago, but also a far more demanding stock.