Thematic Exposure -- 6/10
Intel operates across multiple semiconductor segments with dominant share in client PC CPUs (~65-70%
of x86 by units) and still-leading but eroding share in data center CPUs (~70% units, ~60% revenue).
CCG and DCAI represent ~60% and ~32% of Intel Products revenue respectively, anchoring Intel firmly
above the oligopoly hard gate (>30% share in multiple meaningful segments). However, the strongest
TAM growth themes (AI accelerators, foundry) are areas where Intel is a marginal participant, and its
core x86 franchise faces steady share erosion to AMD and emerging ARM competition.
Weight: 25%
Segment Revenue Breakdown (FY 2025)
| Segment | Q1 2025 | Q2 2025 | Q3 2025 | Q4 2025 | FY 2025 | % of Rev | YoY Chg |
|---|---|---|---|---|---|---|---|
| CCG (Client Computing) | $7,629 | $7,871 | $8,535 | $8,193 | $32,228 | 61% | +7% |
| DCAI (Data Center & AI) | $4,126 | $3,939 | $4,117 | $4,737 | $16,919 | 32% | +32% |
| Intel Foundry | $4,667 | $4,417 | $4,235 | $4,507 | $17,826 | interseg. | +3% |
| Mobileye | $438 | $507 | $504 | $451 | $1,900 | 4% | +15% |
| Altera (deconsolidated Q3) | $368 | $448 | decons. | decons. | ~$816 (H1) | ~2% | -- |
| Total Net Revenue | $12,667 | $12,859 | $13,653 | $13,674 | $52,853 | 100% | ~0% |
Intel Foundry revenue is predominantly intersegment (wafers sold to Intel Products). External foundry
revenue was just $222M in Q4 2025
and ~$307M for FY 2025 -- a negligible fraction of the $320B global foundry TAM. Calendar FY.
Data sourced from Daloopa.
CCG Market Share
~65-70%
x86 client CPU units -- dominant
DCAI Market Share
~70% units
~60% revenue -- eroding toward AMD
External Foundry Rev
~$307M
vs. $320B global foundry TAM
ASIC Run Rate
$1B+
Annualized Q4 2025 -- up 50%+ YoY
Oligopoly Gate: PASS -- x86 Duopoly in Two Major Segments
Intel Exceeds 30% Share in CCG and DCAI -- Score Cap Removed
Client PC CPUs (CCG): ~65-70% unit share of x86, the dominant
PC architecture. This is a meaningful ~$55-60B TAM. Only AMD (>15% share) competes meaningfully.
Server CPUs (DCAI): ~70% unit share (but declining toward 60% revenue share). This is a ~$15-18B TAM growing rapidly from AI infrastructure demand. AMD is the sole competitor with >15% share (~29% units, ~40% revenue). ARM emerging but still <5%.
Mobileye: >50% share in ADAS SoC market, though a smaller segment (~4% of Intel revenue). 19M+ EyeQ6H Surround units committed.
Verdict: The x86 duopoly provides structural relevance and removes the oligopoly score cap of 5/10. However, share erosion is real and accelerating in data center.
Server CPUs (DCAI): ~70% unit share (but declining toward 60% revenue share). This is a ~$15-18B TAM growing rapidly from AI infrastructure demand. AMD is the sole competitor with >15% share (~29% units, ~40% revenue). ARM emerging but still <5%.
Mobileye: >50% share in ADAS SoC market, though a smaller segment (~4% of Intel revenue). 19M+ EyeQ6H Surround units committed.
Verdict: The x86 duopoly provides structural relevance and removes the oligopoly score cap of 5/10. However, share erosion is real and accelerating in data center.
Segment-Level Competitive Landscape
| Segment | Intel Share | Competitors >15% | TAM (2025E) | Trend |
|---|---|---|---|---|
| CCG (PC CPUs) | ~65-70% units | AMD (~25-30%) | ~$55-60B | AI PC refresh cycle; Core Ultra Series 3 (18A) with 200+ OEM designs |
| DCAI (Server CPUs) | ~70% units | AMD (~29% units, ~40% rev) | ~$15-18B | AI-driven demand strong; share losses to EPYC ongoing; ARM <5% |
| Intel Foundry | <1% external | TSMC (~38%), Samsung (~12%) | ~$320B | 18A in production; Microsoft, Apple wins -- but years from scale |
| Mobileye (ADAS) | ~50%+ | NVIDIA approaching ~15% | ~$25-30B | Dominant L2 ADAS; partially divested; competition intensifying |
| Altera (FPGAs) | ~29% | AMD/Xilinx (~51%) | ~$14B | Deconsolidated Q3 2025 -- Silver Lake took control |
| Custom ASIC (NEX) | Emerging | Multiple | ~$26B (edge AI) | $1B run rate in Q4 2025, growing 50%+ YoY |
In core segments (CCG, DCAI), Intel operates in a classic x86 duopoly with AMD. ARM-based server CPUs
(AWS Graviton, NVIDIA Grace, Ampere) are emerging but collectively still <5% of server CPU units.
Theme 1: x86 Duopoly -- Structural Relevance With Eroding Moat
Dominant Franchise Under Competitive Pressure -- Share Losses Persistent in Data Center
Client (CCG): AI PC refresh cycle is a real tailwind.
Core Ultra Series 3 (Panther Lake) on 18A has strong OEM adoption with 200+ designs.
Share erosion to AMD ongoing but slowing in mobile. Intel maintains ~75% revenue share in x86 client.
Server (DCAI): AI-driven infrastructure demand is surging -- Intel is supply-constrained (demand exceeds supply). However, AMD EPYC is gaining rapidly, heading toward ~50% share by 2026. Server ASPs down 8% YoY in Q2 2025 due to competitive pricing pressure. Diamond Rapids (16-channel) and Coral Rapids (multi-thread) are key upcoming products.
Pricing power: Intel is increasingly a price-taker in its core markets. Server ASP declines are a clear sign of competitive weakness. Only Mobileye genuinely sets prices among Intel-owned businesses.
Server (DCAI): AI-driven infrastructure demand is surging -- Intel is supply-constrained (demand exceeds supply). However, AMD EPYC is gaining rapidly, heading toward ~50% share by 2026. Server ASPs down 8% YoY in Q2 2025 due to competitive pricing pressure. Diamond Rapids (16-channel) and Coral Rapids (multi-thread) are key upcoming products.
Pricing power: Intel is increasingly a price-taker in its core markets. Server ASP declines are a clear sign of competitive weakness. Only Mobileye genuinely sets prices among Intel-owned businesses.
12-Month Replacement?
Partial Only
x86 ecosystem locks; AMD capacity-limited
Server ASP Trend
-8% YoY
Q2 2025 -- price-taker dynamics
ARM Server Share
<5%
Graviton, Grace emerging -- still small
Theme 2: Intel Foundry Services -- Ambitious But Early-Innings
18A Technically Impressive -- First Gate-All-Around + Backside Power -- But External Revenue Trivial
Technology milestone: 18A is the first process in the world to
combine gate-all-around transistors with backside power delivery. This is a genuine technical
achievement that has attracted marquee early customers.
Customer wins: Microsoft (18A), Apple (18A, 2027 production). 14A engagements active; customer commitments expected H2 2026. Advanced packaging (EMIB-T) generating customer prepayments with potential >$1B opportunity.
Reality check: External foundry revenue was just ~$307M for FY 2025 against a $320B global foundry TAM. Intel has effectively <0.1% external share. TSMC controls ~38% of the market. The gap between the technology narrative and the revenue reality is enormous. Meaningful external revenue is years away.
CHIPS Act anchor: $8.5B+ in federal subsidies supports US sovereign manufacturing capacity. This is a structural advantage for Intel as one of very few domestic options for leading-edge fabrication.
Customer wins: Microsoft (18A), Apple (18A, 2027 production). 14A engagements active; customer commitments expected H2 2026. Advanced packaging (EMIB-T) generating customer prepayments with potential >$1B opportunity.
Reality check: External foundry revenue was just ~$307M for FY 2025 against a $320B global foundry TAM. Intel has effectively <0.1% external share. TSMC controls ~38% of the market. The gap between the technology narrative and the revenue reality is enormous. Meaningful external revenue is years away.
CHIPS Act anchor: $8.5B+ in federal subsidies supports US sovereign manufacturing capacity. This is a structural advantage for Intel as one of very few domestic options for leading-edge fabrication.
External Foundry Rev
~$307M
FY 2025 -- vs. TSMC at ~$90B+
Q4 2025 External Rev
$222M
Negligible share of $320B TAM
CHIPS Act Subsidies
$8.5B+
US sovereign manufacturing anchor
Key Customers
MSFT, AAPL
18A early wins -- 2027 production
Theme 3: AI Positioning -- Missing the Largest Growth Theme
No Meaningful Position in AI Accelerators -- NVIDIA Dominates -- Intel Benefits Indirectly via CPU Demand
The gap: AI accelerators (GPUs, custom ASICs) represent the
largest and fastest-growing semiconductor TAM. NVIDIA dominates with 80%+ share. Intel has no
meaningful discrete GPU/accelerator position after Habana (Gaudi) failed to gain traction.
Indirect benefit: CPUs are essential for AI infrastructure orchestration and inference workloads. The AI build-out is driving strong server CPU demand -- Intel is supply-constrained in data center. This is a real but secondary tailwind.
Custom ASIC opportunity: The custom ASIC business reached $1B annualized run rate in Q4 2025, growing 50%+ YoY. Management is targeting a $100B TAM opportunity. This is early but represents a credible path into AI-adjacent silicon.
AI PC: The AI PC refresh cycle (on-device inference via NPU) is a legitimate CCG tailwind. Core Ultra Series 3 embeds AI processing capabilities. TAM growing mid-single-digit. This is more defensive than offensive -- preserving the PC CPU franchise rather than capturing new AI revenue.
Indirect benefit: CPUs are essential for AI infrastructure orchestration and inference workloads. The AI build-out is driving strong server CPU demand -- Intel is supply-constrained in data center. This is a real but secondary tailwind.
Custom ASIC opportunity: The custom ASIC business reached $1B annualized run rate in Q4 2025, growing 50%+ YoY. Management is targeting a $100B TAM opportunity. This is early but represents a credible path into AI-adjacent silicon.
AI PC: The AI PC refresh cycle (on-device inference via NPU) is a legitimate CCG tailwind. Core Ultra Series 3 embeds AI processing capabilities. TAM growing mid-single-digit. This is more defensive than offensive -- preserving the PC CPU franchise rather than capturing new AI revenue.
Total Addressable Markets
| Market | Est. TAM | Growth Rate | Intel Position |
|---|---|---|---|
| x86 Client CPU | ~$55-60B | Mid-single-digit | Dominant (~65-70% share) -- AI PC refresh tailwind |
| Server CPU | ~$15-18B | Strong (AI-driven) | Leading but eroding (~70% units to AMD) |
| Global Foundry | ~$320B | 10-15% | <0.1% external share -- years from scale |
| AI Accelerators (GPU/ASIC) | ~$200B+ | 40-60%+ | No meaningful position -- NVIDIA dominates |
| ADAS / Autonomous (Mobileye) | ~$25-30B | 20-25% (to $50B+ by 2030) | Dominant >50% share -- partially divested |
| Edge AI / Custom ASIC | ~$26B | 25-30% | Emerging -- $1B run rate, 50%+ YoY growth |
Intel is well-positioned in mature but large TAMs (client CPU, server CPU). It is a marginal
participant in the fastest-growing markets (AI accelerators, external foundry).
Portfolio Simplification Under Lip-Bu Tan
Shrinking the Portfolio to Focus on Core x86 + Foundry -- Mobileye Partially Divested, Altera Deconsolidated
Under new CEO Lip-Bu Tan (replaced Pat Gelsinger in December 2024), Intel is simplifying its
portfolio to focus on core competencies:
Altera (FPGAs): Silver Lake took control in Q3 2025. Intel retains an equity stake but no longer consolidates. ~29% of FPGA market vs. AMD/Xilinx at ~51%.
Mobileye (ADAS): Intel sold down its stake. Mobileye remains dominant in L2 ADAS with 19M+ EyeQ6H Surround units committed, but competition is intensifying from NVIDIA, Qualcomm, and Chinese players (Horizon Robotics, Huawei).
Net effect: Intel is becoming a more focused company (x86 products + foundry services), but the divested/deconsolidated businesses were among the few where Intel had genuine competitive advantages and growing TAMs.
Altera (FPGAs): Silver Lake took control in Q3 2025. Intel retains an equity stake but no longer consolidates. ~29% of FPGA market vs. AMD/Xilinx at ~51%.
Mobileye (ADAS): Intel sold down its stake. Mobileye remains dominant in L2 ADAS with 19M+ EyeQ6H Surround units committed, but competition is intensifying from NVIDIA, Qualcomm, and Chinese players (Horizon Robotics, Huawei).
Net effect: Intel is becoming a more focused company (x86 products + foundry services), but the divested/deconsolidated businesses were among the few where Intel had genuine competitive advantages and growing TAMs.
Score Rationale
| Factor | Impact | Notes |
|---|---|---|
| x86 duopoly position (CCG + DCAI) | Strong positive | Oligopoly gate PASS -- >30% share in two major segments; removes 5.0 cap |
| AI PC refresh cycle | Positive | Core Ultra Series 3 on 18A with 200+ OEM designs; mid-single-digit TAM growth |
| Server CPU demand (AI infra) | Positive | Supply-constrained; AI build-out drives CPU demand for orchestration/inference |
| Custom ASIC business | Positive | $1B run rate, 50%+ YoY growth -- early but credible AI-adjacent entry |
| 18A process technology | Neutral / TBD | Technically impressive; MSFT + AAPL wins -- but external revenue years away |
| Server CPU share erosion | Strong negative | AMD heading toward 50% by 2026; ASPs down 8% YoY -- price-taker dynamics |
| No AI accelerator position | Strong negative | Missing the $200B+ fastest-growing semiconductor TAM entirely |
| Foundry reality gap | Negative | ~$307M external revenue vs. $320B TAM; narrative far ahead of results |
| Price-taker in core markets | Negative | Only Mobileye sets prices; server and foundry are price-taker businesses |
| Portfolio shrinkage | Mixed | Altera deconsolidated, Mobileye partially divested -- focus vs. lost advantages |
6/10 — Intel scores a 6 reflecting
structural relevance in a critical duopoly, partially offset by share erosion and absent positioning
in the largest growth themes.
The score is shaped by the tension between a dominant installed base and a deteriorating competitive position:
(a) Duopoly passes the gate. Intel holds >30% share in two major segments (CCG at ~65-70%, DCAI at ~70% units). The x86 software ecosystem creates 12-24 month switching costs. This is real structural relevance -- not easily displaced even as share erodes.
(b) Defensive, not offensive. Intel benefits from AI indirectly (CPU demand for orchestration, AI PC refresh, custom ASICs). But it has no position in the $200B+ AI accelerator market where NVIDIA dominates. The thematic exposure is about protecting the existing franchise, not capturing new high-growth TAMs.
(c) Foundry is a narrative, not yet a business. 18A is technically impressive and the Microsoft/Apple wins are meaningful milestones. But ~$307M in external revenue against a $320B TAM means the business is years from materiality. The margin compression (34.5% guided Q1 2026) reflects the cost of this transition.
(d) Share erosion is the core risk. AMD EPYC is heading toward 50% server revenue share by 2026. Server ASPs are declining 8% YoY. ARM is emerging (<5% today but growing). Intel is a price-taker where it used to be a price-setter. The competitive trajectory is clearly unfavorable.
Why 6 and not higher: No AI accelerator position, persistent server share losses, price-taker dynamics, and a foundry business years from meaningful revenue. The turnaround under Lip-Bu Tan is credible, but thematic exposure is more about defending existing franchise than capturing new growth.
The score is shaped by the tension between a dominant installed base and a deteriorating competitive position:
(a) Duopoly passes the gate. Intel holds >30% share in two major segments (CCG at ~65-70%, DCAI at ~70% units). The x86 software ecosystem creates 12-24 month switching costs. This is real structural relevance -- not easily displaced even as share erodes.
(b) Defensive, not offensive. Intel benefits from AI indirectly (CPU demand for orchestration, AI PC refresh, custom ASICs). But it has no position in the $200B+ AI accelerator market where NVIDIA dominates. The thematic exposure is about protecting the existing franchise, not capturing new high-growth TAMs.
(c) Foundry is a narrative, not yet a business. 18A is technically impressive and the Microsoft/Apple wins are meaningful milestones. But ~$307M in external revenue against a $320B TAM means the business is years from materiality. The margin compression (34.5% guided Q1 2026) reflects the cost of this transition.
(d) Share erosion is the core risk. AMD EPYC is heading toward 50% server revenue share by 2026. Server ASPs are declining 8% YoY. ARM is emerging (<5% today but growing). Intel is a price-taker where it used to be a price-setter. The competitive trajectory is clearly unfavorable.
Why 6 and not higher: No AI accelerator position, persistent server share losses, price-taker dynamics, and a foundry business years from meaningful revenue. The turnaround under Lip-Bu Tan is credible, but thematic exposure is more about defending existing franchise than capturing new growth.
Data sourced from Daloopa, Intel public filings and earnings calls, and third-party market research as of April 2026.