01 · Equity deep-dive — synthesized analyst desk
INTC
$133.99 ▲ 6× off the Aug ’25 low
NASDAQ · SEMICONDUCTORSMKT CAP ≈ $670B52-WK $18.97 – $135.48AS OF CLOSE JUNE 18, 2026

The silicon is finally working. The market has already priced the cure.

Left for dead near $19 last summer, Intel has been reforged — a $19B government-anchored recapitalization, marquee checks from Nvidia and SoftBank, the first US angstrom-era process node in volume, and six straight earnings beats. The turnaround is real. Yet at an all-time high the entire sell-side sits roughly 30% below the price, the company still posts GAAP losses, and shares trade near 120× forward earnings. Five analyst lenses, three scenarios, four time horizons.

The verdict · TL;DR
One tension decides the stock: has Intel been recapitalized into a structural winner, or re-rated on a story the numbers don’t yet support? The business is genuinely healing — foundry losses narrowing, AI-exposed revenue compounding 40%+, 18A in production. But the parabola has already discounted the good news and then some. The base case treads water; the bull case requires the foundry to become a real second pillar; the bear case round-trips the move. The setup is symmetric — a doubling and a two-thirds drawdown sit on either side of roughly flat.
5-yr · prob-weighted
$133
−1% vs $133.99
52-week playback · where the tape sits ▲ Pinned at the high · Street left behind
$133.99 · close June 18, 2026 consensus $92 · −31%
$18.97 · 52-wk low · Aug ’25 $135.48 · 52-wk high · June ’26
Price history + cone of outcomes · 2024 → 2031
HISTORICALBULLBASEBEARPROB-WTD
$280$210$140 $70$0 202420252026 202720282029 20302031 $19 · 52-wk low · Aug ’25 $133 $118$122$126 $265 $128 $45 TODAY · $133.99

Gray line = Intel’s actual price into today (~$51 in early 2024 → $19 trough in Aug ’25 → $134 now — a roughly 6× round trip and then some); colored paths = synthesized scenario midpoints forward, probability-weighted (base 40% · bull 25% · bear 35%, deliberately bear-tilted at an all-time high). Mid-year marks. Wall Street’s 12-month consensus sits near $92 — roughly 30% below today’s price (range $45–$150).

Re-weight the scenarios

Those probabilities are a judgment call — so make them yours. Drag to set how likely the bear and bull cases are (base takes the remainder); the blended target below, the dotted line on the chart, and the prob-weighted row of the scenario cards all update live. The default leans bear because the stock is parabolic at an all-time high with the entire Street below it.

35% bear 40% base 25% bull
Blended 5-yr expected $133 −1% vs $133.99
+7%
Q1’26 Revenue ($13.6B)
$0.29
Non-GAAP EPS · vs $0.01 est
41%
Non-GAAP Gross Margin
+22%
Data Center (DCAI) revenue YoY
−$2.4B
Intel Foundry operating loss
$174M
External foundry revenue (Q1)
18A
First US angstrom node in HVM
~120×
Forward P/E on 2026E EPS
02 · The panel — five ways to read the same tape

Five analyst lenses, five answers

The same fundamentals support wildly different conclusions depending on which framework you trust. Each lens below is a synthesized analytical perspective — not a real individual or firm rating — with its own 12-month target.

Turnaround Believer

The Reforging

This is a different company than the one that bottomed at $19. A $19B government-anchored recap converted CHIPS grants into a 10% federal stake; Nvidia and SoftBank wrote $7B of checks at ~$23; 18A is in volume and AI-exposed revenue is compounding 40%+. Six straight beats say Lip-Bu Tan’s discipline is working. When foundry crosses into profit, a second earnings engine ignites and estimates re-rate upward.

12-MO TARGET $165 · foundry inflection begins
Sum-of-the-Parts Strategist

The Break-Up Math

Value the pieces: Intel Products (CCG + DCAI) is a real, cash-generative ~$50B franchise worth a market multiple; Intel Foundry is an option — mostly losses today, but with sovereign backing, advanced-packaging backlog, and angstrom nodes that could be worth a great deal if external customers sign. Add the strategic stakes and net cash. The parts justify a premium to the old Intel — but not an unbounded one.

12-MO TARGET $105 · parts > whole, with limits
Foundry Skeptic / Short

The Priced-In Cure

External foundry revenue was $174M in Q1 against a −$2.4B segment loss — the customer base that justifies the story does not yet exist. Intel’s own 10-K warns it may pause or discontinue 14A and shift to TSMC absent a major external customer. Marquee “deals” (Apple, Nvidia) remain non-binding; a Truth Social post is not a contract. At 120× forward earnings with GAAP losses, the cure is fully priced and then some.

12-MO TARGET $70 · the premium evaporates
Process / Moat Analyst

The Angstrom Edge

The technology is real and arguably ahead. 18A brought RibbonFET and backside power to volume first; 18A-P entered risk production in June; 14A — with High-NA EUV — is reportedly yielding ahead of schedule and is “the real deal” per process analysts. Intel is one of only three firms on the planet with the most advanced packaging, a multi-billion-dollar backlog. The moat is the node roadmap — if yields hold above 90%, customers must come.

12-MO TARGET $115 · roadmap credit, execution risk
Quant / Technical

The Mean-Reversion

Six-fold in twelve months, beta north of 2, an RSI pinned in overbought territory, and insiders selling into the rip. Every analytical model that anchors to fundamentals — DCF, EV/EBITDA, the entire sell-side — sits 30%+ below spot. Parabolas this steep historically give back the final, fastest leg. The signal isn’t that the business is bad; it’s that the price has detached from every disciplined anchor.

12-MO TARGET $90 · gravity reasserts
03 · Wall Street's read

Wall Street 12-month price targets

What the sell-side expects over the next year. Bars are sorted low to high; the dashed line is today’s $133.99 — note how nearly every target sits below it. The stock has lapped the Street.

Consensus ≈ $92 (−31%) · selected names, range $45–$150
BUYHOLDSELL
HSBC $45 UBS $65 Morgan Stanley $80 Mizuho $90 Wells Fargo $95 Barclays $100 Citi $105 Goldman Sachs $112 BofA $135 Benchmark $140 Wedbush $150 TODAY · $133.99

Sell-side 12-month targets — a representative selection of the desks covering Intel; the full consensus is ≈ $92, roughly 30% below today, a Hold/Neutral skew. The dashed line marks today’s $133.99: only a handful of the most bullish desks (BofA double-upgraded to $135 in June; Benchmark $140; Wedbush $150) reach the current price, and even the highest target is just +12%. The median desk sees meaningful downside — the bear’s core observation that the stock has detached from every fundamental anchor. Firms, ratings, and targets illustrative, consistent with a ~$92 consensus and $45–$150 range.

04 · Price scenarios — 1 / 2 / 3 / 5 years

Where the road leads

Synthesized scenario midpoints (mid-year). Returns shown vs. today’s $133.99. These are illustrative frameworks, not predictions with certainty — the outcomes hinge almost entirely on whether Intel Foundry attracts external customers and turns the corner on profitability.

1 Year

Mid-2027
Bull$160+19%
Base$118−12%
Bear$85−37%
Prob-wtd$117−13%

2 Years

Mid-2028
Bull$195+46%
Base$122−9%
Bear$72−46%
Prob-wtd$123−8%

3 Years

Mid-2029
Bull$225+68%
Base$126−6%
Bear$60−55%
Prob-wtd$128−5%

5 Years

Mid-2031
Bull$265+98%
Base$128−4%
Bear$45−66%
Prob-wtd$133−1%
Bull case — show the assumptions & math
Intel Foundry becomes a genuine second pillar: marquee evaluations convert to binding volume, 14A yields hold above 90%, and external foundry revenue scales from a rounding error into the billions. Products (CCG + DCAI) keep their AI tailwind. BofA’s bull frame has Products near $86B and Foundry near $47B of revenue by 2030, with the segment crossing into profit.
EPS ≈ $7–8 by 2031 × ~32× exit multiple → ≈ $265 · 5-yr price CAGR ≈ +15%/yr
Base case — show the assumptions & math
The turnaround is real but the foundry only grinds toward breakeven. Products grow modestly, restructuring keeps margins improving, and Intel earns into — but does not exceed — the multiple it already carries. The stock treads water for years as earnings catch up to the price.
EPS ≈ $4.5 by 2031 × ~28× exit multiple → ≈ $128 · 5-yr price CAGR ≈ −1%/yr
Bear case — show the assumptions & math
No anchor external customer signs; Intel pauses or discontinues 14A and shifts leading-edge production to TSMC (the scenario its own 10-K names). The strategic premium and AI halo evaporate, sentiment reverses, and the parabola round-trips much of its 2026 gain as the multiple de-rates toward a no-growth chipmaker.
EPS ≈ $2 with a ~20× de-rated multiple → ≈ $45 · 5-yr price CAGR ≈ −20%/yr
05 · Follow the cash

Revenue, capex, free cash flow & debt ($B)

Where the money actually goes — and the single hardest fact in the bull case. Intel has burned free cash flow for three straight years; the entire turnaround rests on that bottom bar climbing back above zero.

Annual revenue, capex, FCF & total debt · 2023 → 2026E
REVENUECAPEXFREE CASH FLOWTOTAL DEBT
$60$40$20$0−$20 2023202420252026E −11.9−15.7−4.9+1.0

Revenue has been essentially flat near $53B for three years — this is a margin-and-mix turnaround, not a growth one. The story the bulls love: capex slashed from ~$26B to ~$15B under Lip-Bu Tan’s “no more blank checks” discipline, dragging free cash flow from roughly −$16B back toward breakeven, with management guiding to positive FCF in 2026 (olive). Total debt (slate) is heavy in absolute terms (~$45B) but offset by ~$33B of cash, leaving net debt near $12B. The bear’s rebuttal: 18A and 14A are not free — capex must re-accelerate (tool orders already up ~25%), and the FCF bar could dip again before it durably clears zero. Figures illustrative; debt is gross, FCF is adjusted/free.

06 · Earnings power

EPS path underpinning the targets ($)

The price targets aren’t pulled from the air — each is an EPS estimate times an exit multiple. Here’s the earnings ladder the scenarios are built on, and the uncomfortable fact it reveals: at $134, the stock is paying today for rungs that don’t arrive until the back half of the decade.

Non-GAAP EPS · reported vs. estimated, 2024 → 2031E
REPORTEDESTIMATE
$0$1$2$3$4$5 202420252026E2027E2028E2029E2030E2031E −$0.13 $0.42 $1.09 $1.56 $2.35 $3.10 $3.98 $4.60

Non-GAAP EPS — the clean view; reported GAAP earnings remain negative and swing on restructuring and mark-to-market on the escrowed strategic shares (a ~$1.1B item in Q1’26 alone). Gray = reported, olive = consensus estimates climbing from a 2024 loss toward ~$4.60 by 2031. The catch: at $134 the stock trades at roughly 120× the 2026E rung and over 300× trailing — the base case’s ~$4.50 of 2031 EPS at a ~28× multiple lands near the $128 base-case target, which means years of earnings growth are needed simply to stand still. The whole valuation rests on the upper rungs actually arriving.

07 · Growth scorecard

The turnaround is real

Q1 FY26, year-over-year — read these against a stock that has already run roughly 6× and sits at an all-time high. The business is genuinely healing; the question the chart can’t answer is whether the price already reflects it.

Year-over-year growth by metric · Q1 FY26
COREFRONTIER
Client / CCG +1% Total revenue +7% Intel Products +9% Intel Foundry +16% Data center / DCAI +22% AI-driven revenue +40% ASIC / custom silicon +90% Non-GAAP EPS +123%

The healing is broad: data center +22%, foundry revenue +16%, and the AI-exposed lines — AI-driven revenue, custom ASIC silicon, non-GAAP EPS — compounding far faster off smaller bases (clay). Six straight earnings beats sit behind these bars. The inversion of the usual setup: here the business is improving and the stock has already soared. That gap is the bear’s entire case in one chart — the fundamentals are fine; the multiple has run ahead of them. Off small bases; segment and frontier figures illustrative.

08 · The debate

Bull vs. Bear

The entire valuation argument compresses into one disagreement: has Intel been recapitalized into a structural winner, or re-rated into a story stock that has already lapped its own fundamentals?

▲ THE BULL CASE

  • Backed by the most powerful checkbooks on earth. A 10% US federal stake makes Washington the largest shareholder; Nvidia put in $5B plus a product collaboration; SoftBank added $2B — all in near $20–23. A “Washington Put” under the floor that simply didn’t exist a year ago.
  • The technology is finally working. 18A is in volume — the first US angstrom-era node, with RibbonFET and backside power; 18A-P entered risk production in June; 14A, on High-NA EUV, is reportedly yielding ahead of schedule and called “the real deal” by process analysts.
  • Six straight earnings beats. Q1’26 non-GAAP EPS of $0.29 crushed the $0.01 estimate; gross margin came in at 41% (+650bps vs. guide); data center revenue +22% and AI-exposed revenue +40%.
  • Foundry losses are narrowing toward an inflection. The −$2.4B segment loss shrank quarter-on-quarter; BofA’s bull frame sees Products near $86B and Foundry near $47B of revenue by 2030, with the segment crossing into profit and igniting a second engine.
  • Advanced-packaging scarcity + a real pipeline. Intel is one of only three firms on the planet with the most advanced packaging — a multi-billion-dollar backlog — and a roster of evaluators spanning Apple, Nvidia, Amazon, Google and MediaTek.
  • Capital discipline restored. Lip-Bu Tan’s “no more blank checks” cut capex from ~$26B to ~$15B, dragging free cash flow back toward positive in 2026 after years of heavy burn.

▼ THE BEAR CASE

  • The cure is fully priced — and then some. ~120× forward earnings, >300× trailing, GAAP still in the red, and the entire sell-side sits roughly 30% below the price. Even the most bullish target on the Street implies only ~12% upside.
  • External foundry revenue is a rounding error. $174M in Q1 against a −$2.4B segment loss. The third-party customer base that the whole “second pillar” thesis depends on does not yet meaningfully exist.
  • Intel’s own 10-K names the bear case. It may pause or discontinue 14A and shift leading-edge production to TSMC absent a major external customer — an existential risk disclosed by management itself.
  • Marquee “deals” are non-binding. The Apple arrangement is low-end chips only — TSMC keeps >90% — and surfaced via a Truth Social post. As one analyst put it, a Truth Social post is not a contract.
  • Government ownership cuts both ways. The 10% federal stake is dilutive (shares +9.5% YoY), reduces governance rights, and politicizes capital allocation; one analyst called converting grants into equity “worse” than the grants themselves.
  • The parabola itself is the risk. Up ~6× in a year, beta north of 2, insiders selling into the rip, momentum overbought — with PC demand guided down low-double-digits and memory-cost margin pressure looming in the second half.
09 · Risk map

Risk map — likelihood × impact

Where each risk sits, not just how big it is. The inversion of the usual setup: the hot upper-right corner here isn’t a broken business — it’s the price itself. The structural foundry questions cluster one row down, in “possible.”

Low impact
Medium impact
High impact
Likely
  • Noisy GAAP losses
  • PC / client softening
  • Memory-cost margin pressure
  • Valuation de-rating / momentum unwind
Possible
  • Govt-stake dilution
  • Capex re-acceleration
  • AMD / Arm share
  • Foundry customer shortfall / 14A pause
  • Marquee deals stay non-binding
Tail
  • AI-capex air-pocket / macro shock

Valuation de-rating / momentum unwind

Likely × High

At ~120× forward and >300× trailing with the whole Street below the price, any stumble — or simply fading momentum — can compress the multiple hard and give back the parabola’s final, fastest leg.

Foundry customer shortfall / 14A pause

Possible × High

If no anchor external customer signs, Intel may pause or discontinue 14A and shift leading-edge work to TSMC — the scenario its own 10-K names. The “second pillar” thesis collapses.

Marquee deals stay non-binding

Possible × High

The Apple, Nvidia and other “wins” remain evaluations or low-volume arrangements rather than committed volume. A Truth Social post is not a contract.

PC / client demand softening

Likely × Medium

Management guided the PC TAM down low-double-digits in the second half; CCG is Intel’s largest revenue base and barely growing (+1%).

Memory-cost margin pressure

Likely × Medium

Rising memory and component costs are flagged as a gross-margin headwind into the second half, threatening the hard-won 41% non-GAAP margin.

AI-capex air-pocket / macro shock

Tail × High

A broad pullback in AI infrastructure spend or a macro shock would hit data-center demand and sentiment at once — low odds, but it reprices a high-beta, high-multiple name overnight.

Government-stake dilution

Possible × Medium

The 10% federal stake diluted holders (~9.5% more shares YoY), trims governance rights, and politicizes capital allocation — an overhang even if the “Washington Put” cuts the other way.

Capex re-acceleration

Possible × Medium

18A and 14A are not free; tool orders are already up ~25%. Capex climbing again could push free cash flow back below zero before it durably clears.

AMD / Arm competitive share

Possible × Medium

AMD holds ~36% of x86 server share and is still gaining; Arm-based designs keep encroaching in both data center and client.

Noisy GAAP losses

Likely × Low

Restructuring charges and mark-to-market swings on the escrowed strategic shares keep reported GAAP earnings volatile and below the non-GAAP headline.

10 · Plain-language glossary

The jargon, decoded

Hover the dotted terms in the metrics, or scan the desk’s working definitions here.

Foundry
A contract chip factory that manufactures designs for outside customers. Intel Foundry aims to do for others what TSMC does — the heart of the turnaround thesis.
Process node
A generation of manufacturing technology (e.g. 18A, 14A). Smaller / more advanced nodes pack more performance and efficiency into the same silicon.
Angstrom era
Nodes measured in angstroms (tenths of a nanometer). 18A ≈ 1.8 angstrom-class — Intel’s leap past the “nanometer” naming and back toward the leading edge.
RibbonFET / gate-all-around
A new transistor design that wraps the gate fully around the channel for better control and speed — Intel’s replacement for the older FinFET, debuting on 18A.
Backside power (PowerVia)
Routing power delivery underneath the transistors instead of above, freeing the top for signals — cleaner, faster chips. Intel brought it to volume first.
High-NA EUV
The newest, highest-resolution extreme-ultraviolet lithography machines — essential for 14A. Intel was first to take delivery; central to its claimed edge.
Advanced packaging
Stitching multiple chiplets into one high-performance package. Intel is one of only three firms with the most advanced capability — a multi-billion-dollar backlog.
Yield
The share of chips on a wafer that come out good. Above ~90% is the threshold at which a node becomes profitable and customers commit volume.
External foundry revenue
Sales to third-party customers (not Intel’s own products). At just $174M in Q1, it’s the single number the bull case most needs to grow.
GAAP vs non-GAAP EPS
GAAP is the official, all-in figure (still negative for Intel); non-GAAP strips out items like restructuring and stock comp ($0.29 in Q1). The gap is unusually wide right now.
Forward P/E
Price ÷ next-twelve-months expected EPS. At ~120× on 2026E earnings, Intel is priced for a great deal of future growth to arrive.
Net debt
Total debt minus cash. Intel’s ~$45B of gross debt is offset by ~$33B of cash, leaving roughly $12B net — modest for its size.