TAIPEI, Taiwan — October 31, 2026
Foxconn, the Taiwanese manufacturing giant best known for assembling Apple’s iPhones, is pursuing an ambitious plan to capture half of the global artificial intelligence server market as demand for advanced computing infrastructure continues to reshape the technology industry.
Hon Hai Technology Group Chairman Young Liu revealed that the company already estimates its global AI server market share at approximately 40% and believes it could reach 45% in 2027 and potentially 50% within the next two years.
In an October 30 interview with CNBC’s Christine Tan, Liu said Foxconn’s production capacity was fully booked for at least the next 18 months.
The company also expects AI infrastructure to generate as much as 60% to 65% of its revenue within three to five years, excluding devices that perform AI computing locally.
The projection signals a major transformation for a manufacturer long associated with smartphones, computers and consumer electronics.
But while Foxconn’s chairman believes demand for artificial intelligence remains exceptionally strong, he also delivered a stark warning about the dangers of spending too aggressively.
As technology giants pour hundreds of billions of dollars into artificial intelligence, can Foxconn transform its early manufacturing advantage into lasting profits—or could an industry-wide spending slowdown leave the company exposed to a costly overexpansion?
Foxconn Wants to Control Half of the Global AI Server Market
Foxconn is one of the world’s largest electronics contract manufacturers and a major supplier of AI computing infrastructure.
The company produces systems used in data centers that support advanced artificial intelligence applications.
These include server racks, networking equipment and integrated computing systems used by major technology customers.
According to Liu, Foxconn currently accounts for approximately 40% of the global AI server market.
He expects its share could grow to about 45% next year before potentially reaching 50% within a couple of years.
That target would represent an extraordinary competitive position.
However, the figures should be interpreted carefully.
The current 40% share is a company executive’s estimate, while the 45% and 50% figures are projections.
They are not independently audited market-share results or guaranteed outcomes.
The precise market definition also matters because AI servers, complete racks and broader data-center systems are related but different product categories.
Foxconn’s AI Server Factories Are Booked for 18 Months
One of the strongest indicators cited by Liu was the company’s manufacturing capacity.
Foxconn’s available production capacity is reportedly fully booked for at least the next year and a half.
The demand comes primarily from companies developing advanced AI models and cloud providers building large computing facilities.
This suggests that customers continue committing to infrastructure despite growing concern among investors about the industry’s financial sustainability.
For manufacturers, strong orders can provide greater production visibility.
But booked capacity is not identical to guaranteed future profit.
Companies must still deliver equipment, manage costs and recognize revenue under applicable accounting rules.
Contract terms, component availability and customer spending decisions may also affect final results.
Nevertheless, the reported backlog gives Foxconn a substantial operational advantage as competitors seek a larger share of AI infrastructure demand.
AI Infrastructure Could Become Foxconn’s Biggest Business
Liu expects AI infrastructure to account for approximately 60% to 65% of Foxconn’s revenue within three to five years.
The projection excludes edge AI devices, which perform certain computing tasks locally rather than relying entirely on centralized data centers.
The forecast would represent a dramatic change in Foxconn’s business mix.
For decades, the company has been closely associated with manufacturing products for Apple and other consumer electronics companies.
Now, AI server and data-center systems are becoming increasingly important.
During the second quarter of 2026, Foxconn’s cloud and networking products segment accounted for 51% of consolidated revenue.
That was the first time this business category exceeded half of the company’s quarterly revenue.
However, cloud and networking products include more than AI servers.
The 51% reported segment share should therefore not be described as proof that AI infrastructure alone already generates more than half of Foxconn’s revenue.
The distinction is important when evaluating the company’s changing business model.
Foxconn Revenue Jumps 47% as AI Demand Accelerates
The company’s latest revenue figures support management’s description of strong demand.
Foxconn reported approximately NT$3.03 trillion in third-quarter revenue, equivalent to roughly US$95.4 billion at the exchange rate cited by Reuters.
That represented a 47% increase from a year earlier.
The result exceeded the approximately NT$2.83 trillion estimate from analysts surveyed by LSEG.
September revenue alone reached NT$1.16 trillion, increasing 38% year over year.
Foxconn attributed much of the growth to strong demand for cloud and networking products, including AI infrastructure.
The company also reported increased activity in smart consumer electronics.
But the results do not yet establish how much additional net profit the company generated.
Foxconn’s full third-quarter earnings report is scheduled for November 12.
Until then, investors should distinguish the confirmed revenue increase from the company’s profitability outlook.
Nvidia Is Central to Foxconn’s AI Manufacturing Strategy
Foxconn is one of Nvidia’s most important AI server manufacturing partners.
Nvidia designs advanced graphics processors and computing platforms used to train and run AI models.
Foxconn helps manufacture and integrate systems built around those technologies.
At its October 30 technology event, Foxconn showcased technologies associated with Nvidia’s next-generation Vera Rubin platform.
The demonstration included components, rack-level integration and other infrastructure used in advanced AI data centers.
The relationship highlights the division of responsibilities across the AI supply chain.
Chip designers develop processors and computing architectures.
Manufacturing partners produce, assemble and integrate equipment.
Cloud providers and AI developers then deploy that equipment in computing facilities.
Foxconn’s ambition is to capture more economic value across those stages rather than remaining focused exclusively on assembly.
Six Houston Facilities Support Foxconn’s US Expansion
Foxconn is also expanding its American manufacturing operations.
According to Reuters, Liu said the company currently has six factory buildings in Houston, Texas, where it manufactures AI servers for Nvidia.
He indicated that further expansion could be necessary.
The company is seeking to move further up the supply chain, including manufacturing higher-value equipment for data centers.
US production can bring manufacturing closer to major technology customers.
It may also help address concerns about supply-chain resilience and geopolitical risks.
However, manufacturing in the United States can be more expensive than production in some Asian locations.
Foxconn therefore plans to prioritize products with stronger margins, including advanced server racks and other data-center equipment.
The strategy depends on maintaining enough demand and pricing power to justify the additional operating costs.
Foxconn and OpenAI Are Exploring a New Collaboration
Another important development emerged during Foxconn’s annual technology event.
Liu said he had met OpenAI chief executive Sam Altman and that the companies were working toward a new collaboration.
He indicated that further details would be announced once arrangements were finalized.
The potential partnership would build upon a 2025 agreement involving the design and engineering of data-center racks, components and related AI hardware.
OpenAI is among the companies requiring substantial computing resources to develop and operate advanced artificial intelligence systems.
A deeper partnership could create additional opportunities for Foxconn.
But the latest discussions should not be described as a completed new agreement.
Neither the final commercial terms nor the value of any new arrangement has been publicly established in the reporting reviewed.
For investors, the relationship remains a potentially significant development rather than confirmed new revenue.
Big Tech’s $750-Billion Spending Race Creates Enormous Opportunities
Foxconn’s growth is connected to the world’s largest technology companies.
Amazon, Microsoft, Alphabet, Meta and other businesses are spending heavily on AI infrastructure.
Their investments involve specialized chips, data centers, cooling systems, networking equipment and electricity supply.
WWC ONE MEDIA examined the broader investment cycle in Big Tech’s $750-Billion AI Gamble Shakes Wall Street as Core Businesses Slow — But Will the Spending Ever Pay Off?.
The report examined the risk that infrastructure spending could outpace the financial benefits generated by AI services.
For Foxconn, such investment creates demand for the hardware it manufactures.
But the relationship also introduces concentration risks.
If major cloud providers reduce spending simultaneously, equipment manufacturers could face fewer orders and lower factory utilization.
The industry’s capital expenditure boom therefore creates both tremendous opportunities and substantial financial uncertainty.
Amazon’s Sold-Out AI Capacity Highlights the Supply Problem
Foxconn’s reported manufacturing backlog is consistent with broader infrastructure shortages.
WWC ONE MEDIA recently reported that Amazon’s AI cloud capacity was already sold out through 2026, with 2027 nearing full booking.
The report examined Amazon’s efforts to expand computing capacity amid heavy demand.
Such shortages can benefit hardware manufacturers because customers may place additional orders to secure infrastructure.
However, demand for computing services and orders for physical equipment are not identical.
Building more data centers also requires sufficient electricity, real estate, cooling systems and networking capacity.
A shortage in any one of those areas can delay deployment even when servers are ready.
For Foxconn, the challenge is aligning manufacturing expansion with the pace at which customers can build and operate new facilities.
Young Liu Warns That Overinvestment Could Destroy Value
Despite his optimism, Liu emphasized that companies must manage capital expenditure carefully.
In his CNBC interview, he warned against allowing investment to outpace sustainable demand.
Foxconn’s anticipated annual AI spending is approximately $3 billion to $5 billion, although that amount could rise if the company increases its vertical integration.
Expanding production requires new facilities, machinery, labor and supply-chain arrangements.
Those investments can be worthwhile when demand is strong and margins are sufficient.
But they can become costly if the market changes.
A company that builds too much manufacturing capacity may be left with underutilized assets.
The risk is especially important in industries where technology changes rapidly.
Equipment purchased for one generation of hardware may become less valuable as customers move toward newer architectures.
Liu’s comments suggest Foxconn is attempting to benefit from AI growth without assuming that demand will remain unlimited.
Is the AI Industry Becoming a Financial Bubble?
Investors are increasingly debating whether artificial intelligence infrastructure spending is sustainable.
Some companies have reported rapid growth in cloud computing and AI services.
Others have faced questions about financing costs, profitability and the timing of returns.
WWC ONE MEDIA explored these concerns in Asian Chip Stocks Plunge as Amazon’s AI Spending Sparks Funding Fears — But Could the $2.1-Trillion Data Center Boom Be at Risk?.
That report examined how investor concerns about the cost of AI infrastructure affected Asian technology shares.
Liu rejected the idea that the entire AI industry is experiencing a bubble.
He expects Foxconn to maintain double-digit growth in 2027.
But his optimism is a management outlook rather than proof that AI infrastructure spending cannot slow.
There is a difference between strong current demand and the long-term profitability of every company participating in the industry.
Financing Costs Could Become a Major Constraint
The infrastructure boom is also influencing corporate financing.
Some leading technology companies are raising substantial amounts through debt markets.
WWC ONE MEDIA recently examined this in Microsoft Raises $25 Billion in Massive Bond Sale as AI Spending Explodes — But Is Big Tech’s Debt Gamble Getting Dangerous?.
Borrowing allows companies to finance expansion without immediately using all available cash.
But debt creates repayment obligations and interest costs.
If customers become more cautious about investing, those obligations could affect future spending plans.
A reduction in AI infrastructure orders would potentially affect manufacturers such as Foxconn.
The risk is not that all spending is financed by borrowing.
Rather, the pace of investment increasingly depends on whether companies can continue generating sufficient cash and securing capital on acceptable terms.
Anthropic’s AI Chip Financing Plans Show How Large the Commitments Have Become
The financial scale of the AI industry is also visible in arrangements involving major model developers.
WWC ONE MEDIA covered one example in Anthropic Offers Blackstone More Than 10% Returns in $60 Billion AI Chip Deal — But the Rising Cost of AI Financing Signals a Bigger Problem.
The report discussed a proposed financing structure for AI computing infrastructure.
Such arrangements can support hardware demand, but they also highlight how much capital is needed to operate advanced AI systems.
For Foxconn, its ability to benefit from that demand depends on winning contracts and delivering equipment profitably.
An increase in industry investment does not automatically translate into the same percentage increase in Foxconn earnings.
Margins, financing costs, competition and production efficiency all matter.
Could Humanoid Robots Transform Foxconn’s Factories?
Foxconn is also investing in industrial automation.
Reuters reported that the company has deployed more than 10 humanoid robots at facilities in Texas for internal use.
It plans to expand deployment to other locations.
Foxconn Industrial Internet has also introduced wheeled robotic systems designed for industrial tasks such as lifting, assembly and material handling.
The company is working with Nvidia’s robotics technology, including its GR00T foundation model.
These developments could eventually improve productivity and help manufacturers respond to labor shortages or rising operating costs.
However, early robot deployment should not be confused with fully autonomous factories.
The commercial benefits of humanoid robots will depend on reliability, safety, maintenance costs and the range of tasks they can perform.
For now, Foxconn is using automation as part of a broader effort to improve industrial efficiency.
What Foxconn’s Expansion Means for Asia and the Philippines
Foxconn’s transformation matters to the Asia-Pacific manufacturing economy.
Taiwan remains an important center for electronics production, semiconductor supply chains and advanced computing technology.
Expansion into AI infrastructure can increase demand for specialized components, engineering services and manufacturing equipment.
Other Asian economies may benefit through their participation in those supply chains.
For the Philippines, opportunities could arise in electronics manufacturing, engineering services and related business activities.
But Foxconn’s announced US expansion does not establish that additional investments or jobs are coming to the Philippines.
Any such development would require specific project announcements.
The wider implication is that manufacturing competitiveness is increasingly tied to advanced computing systems, reliable electricity, skilled workers and resilient supply chains.
The Bigger Question: Can Foxconn Dominate AI Servers Without Taking Excessive Risk?
Foxconn’s ambitions are extraordinary.
The company wants to control approximately half of the global AI server market.
It expects AI infrastructure to account for most of its revenue within several years.
Its production capacity is reportedly booked for 18 months.
And its latest reported quarterly revenue increased 47%.
These are encouraging indicators of demand.
But they are not guarantees of future profitability.
Competition can intensify.
Customers can delay spending.
Manufacturing costs can rise.
And the next generation of computing technology may require substantial new investments.
The company’s ability to control capital expenditure while meeting customer demand will be central to its long-term strategy.
The Bottom Line
Foxconn Chairman Young Liu estimates the company currently controls approximately 40% of the global AI server market and could reach 45% in 2027 and 50% within a couple of years.
Production capacity is reportedly fully booked for at least 18 months.
AI infrastructure could eventually account for 60% to 65% of company revenue, according to management’s forecast.
Foxconn is expanding its US manufacturing operations, strengthening its Nvidia relationship and exploring further collaboration with OpenAI.
Third-quarter revenue increased 47% to NT$3.03 trillion, although full quarterly earnings remain scheduled for November 12.
Despite strong demand, Liu has emphasized the need to avoid excessive investment.
As Foxconn races to transform itself from Apple’s manufacturing powerhouse into one of the world’s dominant AI infrastructure suppliers, the biggest question remains: Can it capture half the global AI server market while protecting profitability—or will the next slowdown in technology spending expose the hidden risks of its unprecedented expansion?
