Foxconn vs Apple Era: How Nvidia AI Servers Are Reshaping the Manufacturing Giant

 

A company known for assembling iPhones saw cloud and networking revenue surpass smart consumer electronics for the first time in the fourth quarter of 2025, during the traditional consumer electronics peak season. Apple's orders did not shrink. Another business line simply grew faster.

 

Per Hon Hai's full-year and fourth quarter 2025 results, revenue from cloud and networking products surpassed smart consumer electronics for the first time even during the traditional peak season for ICT products, becoming the largest product category in the quarter, with the segment expanding from 30% of full-year business the prior year to 40%. By the first quarter of 2026 that share approached half.

 

The market's interest here is not the simplistic notion that Foxconn is abandoning Apple. Second quarter data says the opposite: all four product categories grew year over year, with smart consumer electronics, which includes iPhone assembly, posting significant growth. The real change is that this company's revenue structure went from one engine to two, and that shift has material consequences for its margins, cyclicality and valuation framework.

 

 

 

Cloud and networking products rose from 30% of total business in 2024 to 40% in 2025 and approached half by the first quarter of 2026, becoming the largest product category in the fourth quarter of 2025.

 

Second quarter consolidated revenue was NT$2.5133 trillion, up 18.02% sequentially and 39.83% year over year, with all four product categories growing, cloud and networking, components and others, and computing products showing strong growth while smart consumer electronics showed significant growth.

 

July revenue reached NT$946.512 billion, up 15.18% sequentially and 54.19% year over year, the first month above NT$900 billion, with seven-month cumulative revenue of NT$5.589 trillion, up 37.89%.

 

The company holds roughly 40% market share in AI server racks, with management projecting AI rack shipments to more than double in 2026 and shipments and revenue for 800G-plus switches also potentially doubling.

 

First quarter gross margin was 6.18%, operating margin 3.57% and net margin 2.36%, with the first two improving year over year while net margin declined.

 

Full-year 2025 gross margin was 6.15% and operating margin 3.20%, with the company explaining that high-priced AI products diluted gross margin while operating margin grew steadily above 3%.

 

Capital expenditure rose about 27% in 2025 to NT$174 billion and is expected to rise more than 30% in 2026.

 

Chairman Young Liu targeted 2026 revenue above NT$9 trillion in March, against NT$8.10 trillion in 2025.

 

 

 

Three percentages capture the speed of this transformation. In 2024, cloud and networking products accounted for 30% of total business. For full-year 2025 that rose to 40%. By the first quarter 2026 earnings call, the company said the segment approached half of revenue.

 

Per Hon Hai's first quarter 2026 results, rotating CEO Michael Chiang said that with the ICT industry entering its traditional first quarter off-season alongside rapid growth in AI server demand, the cloud and networking segment now accounts for nearly 50% of revenue, demonstrating that the group's AI strategy has already brought about a structural transformation that helps mitigate the industry's seasonality, while a more diversified AI customer base has made the product portfolio more balanced.

 

Moving from 30% to nearly 50% within two years is an unusually fast migration for a company with annual revenue above NT$8 trillion.

 

 

This is the point most in need of clarification. Second quarter data shows simultaneous growth rather than substitution. Per Yahoo Taiwan citing a Chinese-language wire summary, second quarter cloud and networking, components and others, and computing products all grew strongly year over year, while smart consumer electronics showed significant growth.

 

Cnyes, citing a research provider, put it more directly: second quarter revenue growth came primarily from the cloud and networking business group on strong AI product demand, while the smart consumer electronics business group covering iPhone assembly also delivered significant growth, together lifting overall revenue above market expectations.

 

One caveat is essential. Hon Hai does not break out iPhone or handset revenue in its monthly disclosures, and the smart consumer electronics category spans more than one customer. The safer formulation is that AI servers are the primary growth engine while consumer electronics and other products contribute as well, rather than attributing the overall increase to any single customer or product.

 

 

 

This is the key to reading Hon Hai's financial structure and the part most coverage handles crudely.

 

The company spelled out the mechanism in its full-year 2025 results: gross margin, operating margin and net margin came in at 6.15%, 3.20% and 2.34% respectively, against 6.25%, 2.92% and 2.23% the prior year. This shows that although high-priced AI products diluted gross margin, operating margin grew steadily above 3%, with core business profitability benefiting significantly from AI products.

 

The reason is accounting structure. AI server racks carry very high unit prices, a substantial portion of which is purchased chip cost, lifting both revenue and cost of sales and thereby thinning gross margin. But because individual orders are large and system integration is high, selling and administrative expenses do not scale proportionally, so operating margin rises instead.

 

First quarter data continued the pattern: gross margin 6.18%, operating margin 3.57% and net margin 2.36%, with the first two improving from 6.11% and 2.83% a year earlier.

 

 

Notably, first quarter net margin fell from 2.56% a year earlier to 2.36%, moving opposite to gross and operating margins. Over the same period operating profit rose 63% while net profit attributable to the parent rose only 19%, a gap that cannot come from gross profit or operating expenses and must originate in non-operating items, currency, tax or minority interests.

 

The company has not itemised that gap, so its composition cannot be confirmed. For investors, the implication is that operating margin is the cleaner measure when assessing the AI business's contribution, while net margin mixes in variables unrelated to business structure.

 

One analysis frames the practical approach well. A Chinese-language market note suggests putting gross margin, operating margin, the cloud and networking share and consumer electronics pull-in into a single table after each quarterly report and earnings call, warning that if AI server volume grows without those first two ratios holding or improving, the market will suspect high revenue is merely an amplified version of high cost.

 

 

 

Hon Hai is not peripheral in this chain. Per reporting carried by El Paso Inc, the company said in March that shipments of AI server racks could grow exponentially this year after reaching a market share of around 40%. The same report cited Chairman Young Liu saying on the earnings call that 2026 is shaping up to be a promising year as AI investments by global hyperscalers continue to rise.

 

Specific product guidance came from the first quarter call. Management said AI rack shipments were expected to maintain high double-digit growth in the second quarter, with full-year shipments projected to more than double and quarterly volumes rising sequentially as customer projects progress. In high-speed switches above 800G, driven by demand for high-speed networking architectures in AI data centres, related shipments and revenue also have potential to double this year.

 

Co-packaged optics is the next milestone. The rotating CEO said CPO switches were scheduled to enter mass production in the third quarter with annual shipments expected in the tens of thousands of units and multi-fold growth projected next year based on current visibility, while the group continues strengthening capabilities in optical modules, optoelectronic integration, cables, connectors, high-speed transmission and power management, aiming to enhance system performance, delivery efficiency and supply chain control through greater in-house production and integration.

 

 

Customer capital spending underpins that guidance. Per the Taipei Times, Alphabet, Amazon, Meta and Microsoft are setting aside about US$725 billion for AI spending this year, even as warnings about overcapacity abound and questions about monetisation persist. The same report notes that Hon Hai has established itself as a key AI hardware player by assembling servers that house Nvidia accelerators while deriving a significant chunk of sales from assembling Apple's iPhones and MacBooks.

 

For investors following both AI hardware and crypto, the competition between the two for high-end chips and power capacity is worth tracking. On venues offering crypto trading alongside stock-linked products, such as MEXC, relative performance between the two often signals shifts in risk appetite early.

 

 

 

Capital expenditure is the direct price of this structural change. On the first quarter call the company disclosed that 2025 capital expenditure rose about 27% year over year to NT$174 billion and should rise more than 30% in 2026, with investment focused on regional manufacturing deployment, automation implementation and upgrades to core capacity.

 

Two other metrics provide context. First quarter EBITDA rose to NT$102.4 billion, which management said indicates that while the group actively invests in future growth its overall financial health remains robust. First quarter return on equity was 2.88%, a clear improvement over the same period across the previous two years.

 

Annual targets are more explicit. Full-year 2025 ROE reached 11.25%, and the Enterprise Value Enhancement Plan approved by the board incorporates operating margin and ROE into its core commitments with a near-term ROE target of 12%, against roughly 9% in 2023 and 2024.

 

The strategic framework has been updated alongside. In its 2025 results the company unveiled a five-year plan using AI as the core driving force across three transformation strategies: Foxconn 1.0 Operational Excellence, Foxconn 2.0 Intelligence-Driven Growth and Foxconn 3.0 Platform Value Creation, with a stated goal of transforming from the world's most important technology manufacturing partner into the most trusted industrial platform in the AI era.

 

 

The first risk is margin sustainability. High-price, low-gross-margin AI servers keep diluting gross margin as their share rises. Improving operating margin has offset that, but whether the offset persists depends on scale effects and expense control. The first quarter net margin decline shows that even with operating improvement, final profit can be eroded elsewhere.

 

The second is customer concentration. AI growth depends heavily on a handful of hyperscalers whose capital spending decisions are highly correlated, while the consumer electronics side carries its own large-customer dependence. Two engines do not automatically mean diversified risk; each line has its own concentration issue.

 

The third is component supply. In its June revenue commentary the company noted that AI pull-in demand for cloud and networking products remained strong while materials were tight, indicating capacity constraints are already appearing at the supply chain level.

 

The fourth is the external environment. In its March results the company identified the Middle East conflict as this year's biggest external risk while stating that diversified capacity and localised manufacturing have enhanced its resilience.

 

On scenarios, the base case is the cloud and networking share continuing to rise and stabilising above half, gross margin holding near 6% while operating margin keeps improving, and the company reaching its NT$9 trillion annual revenue target. A second is rapid volume growth in higher value-added products such as CPO and high-speed switches, stabilising or lifting gross margin, at which point profit growth could outpace revenue growth and the valuation framework migrates from contract manufacturer toward system integrator. A third is hyperscaler capital spending slowing, where the combination of high revenue growth and thin margins amplifies earnings volatility while rigid capital expenditure compounds the effect.

 

Status distinctions: complete first quarter and 2025 financial data come from official company releases and are confirmed. Second quarter and monthly revenue figures come from regular disclosures. Product shipment guidance reflects management's view at a specific time. Third-party analyst views are those institutions' judgements rather than statements of fact.

 

 

What actually matters in this transformation is not that cloud and networking rose from 30% to nearly 50% of revenue. It is that the migration changed margins in two opposite directions simultaneously. The company itself states in its results that high-priced AI products diluted gross margin while operating margin grew steadily above 3%. This was not a simple business substitution but a restructuring of the financial profile: gross margin down, operating margin up, and net margin dragged by non-operating items. Three lines moving in three directions means any single metric produces a partial conclusion about how the transition is going.

 

Three misreadings look likely. The first is reading cloud and networking overtaking consumer electronics as evidence that the Apple business is shrinking. All four product categories grew year over year in the second quarter, with smart consumer electronics including iPhone assembly posting significant growth. The mix shift stems from differing growth rates rather than absolute contraction. The second is treating declining gross margin as evidence of a failing transition. In a business model that passes through chip or commodity costs, high-priced products naturally dilute gross margin, and the real test is whether operating margin improves in parallel, which it did in both 2025 and the first quarter of 2026. The third is ignoring the limits of monthly disclosure. The company does not break out iPhone or handset revenue monthly, and the smart consumer electronics category spans multiple customers and product lines, so attributing the overall increase to a single customer lacks basis.

 

Three verifiable metrics deserve tracking, and they should sit in the same table. First, the parallel movement of the cloud and networking share against the consumer electronics share, since only one rising while the other falls in absolute terms constitutes genuine substitution. Second, the combined direction of gross margin and operating margin, where both deteriorating alongside AI volume growth would confirm that high revenue is merely amplified cost. Third, actual shipment progress on CPO and 800G-plus switches, guided at third quarter mass production, tens of thousands of units this year and multi-fold growth next year, since such components should carry markedly better margin structures than system assembly and are the only hard evidence of genuine mix upgrading.

 

For cross-asset investors, the lesson is that migration within a supply chain determines long-run returns more than sector conditions do. Foxconn is extending from system assembly toward system integration and in-house components, which is fundamentally a move upstream along the same chain, and margin improvement only materialises after that move completes, with a transitional period of diluted gross margin in between. The same path recurs in crypto, where mining hardware manufacturing, mining operations and chip design sit on one chain with margin structures differing several times over, and where markets frequently price new businesses using the old framework during the transition. When evaluating names like this, establish which direction the company is moving along the value chain first, then judge where current margins sit within that move. From a risk management standpoint, during the transition the direction of operating margin carries more information than the absolute level of gross margin.

 

This analysis rests on official company releases, monthly revenue disclosures and credible reporting available now. Complete financial detail, subsequent operating results and customer capital spending cadence could each change the conclusion, and no single scenario should be treated as a fixed expectation.

 

 

 

Not deliberately. Another business line is simply growing faster. All four product categories grew year over year in the second quarter, with cloud and networking, components and others, and computing products showing strong growth while smart consumer electronics, which includes iPhone assembly, showed significant growth. The mix shift reflects differing growth rates rather than absolute contraction. Note also that the company does not break out iPhone or handset revenue in monthly disclosures, and the consumer electronics category spans multiple customers.

 

 

The share has moved quickly. Cloud and networking products rose from 30% of total business in 2024 to 40% for full-year 2025, and by the first quarter 2026 earnings call the company said it approached half of revenue. In the fourth quarter of 2025, even during the traditional peak season for ICT products, the category's revenue surpassed smart consumer electronics for the first time to become the largest product category in the quarter.

 

 

Because unit prices are high and purchased content is large. A substantial portion of an AI server rack's cost consists of purchased chips and key components, which lifts revenue and cost of sales together and thins gross margin. The company explained in its 2025 results that while high-priced AI products diluted gross margin, operating margin grew steadily above 3%, with core business profitability benefiting significantly from AI products.

 

 

The direction differs by layer. Full-year 2025 gross margin, operating margin and net margin were 6.15%, 3.20% and 2.34% against 6.25%, 2.92% and 2.23% the prior year, so gross margin fell while operating and net margins rose. In the first quarter of 2026, gross margin of 6.18% and operating margin of 3.57% both improved year over year while net margin fell from 2.56% to 2.36%. Operating margin is the cleanest measure of the mix shift's effect.

 

 

At the centre. The company said in March that AI server rack shipments could grow exponentially this year after reaching market share of around 40%. Management projects 2026 rack shipments to more than double, with 800G-plus switch shipments and revenue also potentially doubling, and CPO switches entering mass production in the third quarter at tens of thousands of units this year and multi-fold growth next year. Reporting also notes the company established itself as a key AI hardware player by assembling servers housing Nvidia accelerators.

 

 

Mostly in capital expenditure. The company disclosed 2025 capital expenditure rising about 27% to NT$174 billion with more than 30% growth expected in 2026, focused on regional manufacturing deployment, automation and core capacity upgrades. For context, first quarter EBITDA was NT$102.4 billion, which management said reflects robust financial health. Annual ROE rose from roughly 9% in 2023 and 2024 to 11.25% in 2025, with a near-term target of 12%.

 

 

First, margin sustainability, since AI products' gross margin dilution must keep being offset by scale effects. Second, customer concentration, with AI growth depending on a few hyperscalers while consumer electronics carries its own large-customer dependence, so two engines do not equal diversified risk. Third, component supply, with the company noting in June that AI pull-in demand remained strong while materials were tight. Fourth, the external environment, with the Middle East conflict identified as this year's biggest external risk.

 

 

Three items, ideally in one table. First, the parallel movement of the cloud and networking share against the consumer electronics share, distinguishing substitution from joint growth. Second, the combined direction of gross margin and operating margin, since both deteriorating would confirm that high revenue is merely amplified cost. Third, actual shipment progress on CPO and 800G-plus switches, the only hard evidence that the product mix is genuinely upgrading.

 

 

This article is provided for informational and research purposes only and does not constitute investment advice, financial advice, legal advice, tax advice or any recommendation to transact. The financial figures, monthly revenue data and business guidance referenced here come from official company releases, regular disclosures and credible reporting; complete financial detail should be taken from the company's formal financial reports; product shipment guidance reflects management's view at a specific point in time and actual results may differ; third-party analyst views and interpretations are those institutions' judgements rather than statements of fact; and every ratio comparison from public data in this article is a simple arithmetic illustration. Prices of crypto assets, equities and other related financial instruments can move sharply over short periods, investors may lose their entire principal, and exposure to New Taiwan dollar denominated assets carries additional currency risk. Historical performance, technical indicators and on-chain data cannot guarantee future outcomes and should not be read as a promise or forecast regarding any asset. Readers should conduct their own independent research, verify official information directly, and evaluate any decision against their own financial circumstances, investment objectives, experience and risk tolerance, consulting a qualified professional adviser where appropriate. The MEXC Crypto Pulse team accepts no liability for any direct or indirect loss arising from use of or reliance on the information in this article.

 

 

James Mitchell specializes in technical analysis, market trends, and trading strategies for both Bitcoin and altcoins. Based in London, he has over 10 years of experience in financial markets. Before joining MEXC Learn, James worked as a senior analyst at a leading European investment firm, where he developed expertise in risk management and quantitative trading. His transition to cryptocurrency markets began in 2017, and he has since become recognized for his data-driven approach. He holds a Master's degree in Financial Economics from the London School of Economics. His analytical approach combines traditional technical analysis with on-chain metrics to provide readers with actionable insights.

 

Areas of Expertise:

 

 

Want the fastest access to MEXC's latest updates? Join our official Telegram group now!

MEXC Platform: Download App | Zero Fee Trading | Buy USDT Now

Join MEXC Community: X (Twitter) | Telegram | Discord

Get Insights: MEXC Learn | Terms of Service | MEXC Blog | MEXC News

Account Verification: Understand KYC | How to Complete KYC

External Content Platforms: Substack | Medium | Paragraph | LinkedIn | X(News)