Intel, once the world's most admired chipmaker before innovation failures pushed it to the brink of collapse, is dreaming of a spectacular revival on the back of its own restructuring and full-fledged support from the Donald Trump administration. Riding the artificial intelligence (AI) boom, the company is seeing strong sales of its flagship central processing units (CPUs) while also accelerating efforts to rebuild its foundry business, the contract manufacturing of semiconductors. On Wall Street, skepticism toward Intel is gradually lifting. Intel's rebound is expected to have a significant impact on the market, as it symbolizes the broader revival of the U.S. semiconductor industry that President Trump is pursuing. Still, with the foundry business yet to gain traction and the company not fully out of the red, considerable opinion holds that Intel's progress should be watched cautiously.
Intel Posts Q1 'Earnings Surprise' on Rising AI CPU Demand... Signal of Revival
Intel surprised Wall Street by reporting first-quarter earnings that beat market expectations after the market close on the 23rd local time. Intel announced first-quarter revenue of $13.6 billion (about 20.1 trillion won), up 7% from the same period last year. This was 11% higher than analysts' forecasts. Intel also guided second-quarter revenue of $13.8 billion to $14.8 billion, above the market estimate of $13 billion.
Driving Intel's earnings was the CPU, its traditional flagship product. The company benefited substantially from the spread of CPU demand thanks to the proliferation of AI agents and large language models (LLMs). Wall Street paid particular attention to Intel's first-quarter data center segment revenue of $5.1 billion (about 7.6 trillion won), which surpassed the market estimate of $4.5 billion. This was because it meant demand for enterprise CPUs used in data center servers and the like had risen accordingly. While PC CPUs, a key revenue source for Intel, recently faced difficulties due to surging memory chip prices, data center demand emerged as a positive factor to offset this.
Revenue rose, but profit remained in the red. The market had expected Intel to post net profit of $2.5 million in the first quarter, but the actual figure was a net loss of $3.7 billion. Intel explained that one-time losses occurred, including costs from Mobileye, its autonomous driving technology subsidiary, and settlement costs for derivatives linked to the 10% stake held by the U.S. administration. In a statement that day, Intel CEO Lip-Bu Tan stressed, "The innovation wave in AI is shifting from foundation models toward inference and agents," adding, "This paradigm shift is dramatically driving up demand for Intel's CPUs as well as our wafer and advanced packaging services."
Buoyed by the strong results, Intel showed its mettle by rising 2.31% on the 23rd even in a down market. Intel went on to surge a whopping 23.60% on the 24th as well, leading the broader technology stocks on the New York exchange. Intel's daily share price gain that day was the largest in 26 years since 2000. As expectations for expanding AI CPU demand spread to graphics processing units (GPUs), Nvidia's stock also soared 4.32%. Nvidia's market capitalization recorded $5.075 trillion that day, breaking through the $5 trillion barrier for the first time in six months since October 29 last year. Other technology stocks also rose one after another on the Intel effect, including Microsoft (2.13%), Amazon (3.49%), Google parent Alphabet (1.63%), Broadcom (0.67%), Facebook parent Meta (2.41%), Tesla (0.69%), AMD (13.91%), Qualcomm (11.12%), and Micron (3.11%).
Starting With World's First DRAM to Build a 50-Year CPU Empire... Collapsed by AI, Revived by Trump's Stake Purchase
Wall Street's current interest in Intel was hard to imagine until the first half of last year. Founded on July 18, 1968, by Robert Noyce and Gordon Moore, Intel grew with a focus on memory chips in its early days. In 1970, it also succeeded in commercializing DRAM for the first time in the world.
Using this as a foothold, Intel entered the microprocessor business in the 1970s. Intel rose to become a company symbolizing global information technology (IT) innovation, admired by engineers around the world. Then, unable to overcome the low-price offensive of Japanese companies that entered as latecomers, Intel ultimately withdrew from the memory chip business in 1985. Then-CEO Andy Grove pulled out a strategy of concentrating on the PC CPU business instead of abandoning memory chips, and this became a masterstroke for Intel.
Intel, which had dominated the CPU market for nearly 50 years, began to falter with the arrival of the mobile era. Then, as the AI era emerged, it handed over much of the market to Nvidia's GPUs. The foundry division, where it made a strategic gamble, was completely pushed aside by Taiwan's TSMC. Intel, which posted its worst-ever loss of about $2.8 billion in the first quarter of 2023, recorded net losses in succession in 2024 and last year. Over two years it laid off about 20,000 people, close to roughly 20% of its entire workforce, and was reduced to a company barely surviving on subsidies from the former Joe Biden administration's CHIPS Act. The U.S. Department of Commerce announced in November 2024 that it would provide up to $7.865 billion in direct funding to Intel. Including this, Intel agreed to receive a total of $10.9 billion in subsidies for commercial and military semiconductor production.
The turning point came with the Trump administration's intervention. In August last year, the Trump administration purchased a 9.9% stake in Intel, which had fallen into management difficulties, using subsidies promised by the former Biden administration, directly seizing the position of largest shareholder.
The federal government's acquisition of an Intel stake gained momentum when President Trump and CEO Tan met on August 11 last year. Until the 7th of the same month, before meeting CEO Tan, President Trump had raised concerns that Tan was linked to China and called for his resignation. But on the 11th, on his social networking service Truth Social, he praised Tan, saying, "I met with CEO Tan together with Commerce Secretary Howard Lutnick and Treasury Secretary Scott Bessent, and it was very interesting."
After Intense Restructuring, Wins Investment From Nvidia and SoftBank... Musk's 'Terafab' Project Also Joins
Intel, having been embraced by the Trump administration, launched intense restructuring in the second half of last year. After cutting costs through massive layoffs and the sale of business units, it attracted large investments from Nvidia, SoftBank and others.
Afterward, Intel gradually began to change the perception of a Wall Street that had been ambivalent about its prospects for revival. Intel particularly heightened expectations for securing future growth engines by deciding this month to join "Terafab" (a mega-scale semiconductor production facility), the AI chip production project of Tesla CEO Elon Musk. On the 7th of this month, Intel said on its official account on X (formerly Twitter), "We are proud to participate in the 'Terafab' project together with SpaceX, xAI, and Tesla to refactor silicon fab technology." It also posted a photo of CEO Tan and CEO Musk shaking hands and said, "It was a pleasure to welcome CEO Musk at Intel last weekend." Refactoring refers to the development process of improving a semiconductor's performance or reliability.
Intel went on to stress, "Intel's capability to design, manufacture, and package ultra-high-performance semiconductors at scale will contribute to accelerating the goal of producing 1 terawatt (TW) of computing power annually to support advances in AI and robotics." CEO Tan also posted on his X account, "CEO Musk has a proven track record of reimagining entire industries, which is exactly what semiconductor manufacturing needs today," expressing hope that "Terafab will bring major changes to the way silicon logic, memory, and packaging are manufactured going forward."
Terafab is a mega-scale semiconductor production base that CEO Musk is pursuing to produce his own chips for use in AI, robotics, space data centers and more. CEO Musk has repeatedly stressed the need for Terafab production, saying that chip output has not met his company's demand. Last month, he announced he would start construction of an advanced technology fab in Austin, Texas.
CEO Musk first revealed his plan to build Terafab at Tesla's shareholder meeting in November last year. At the time, CEO Musk stressed, "Considering the geopolitical uncertainty that will emerge three years from now, there is always a risk that the chips we expect will not arrive, and we need to secure more production capacity within the U.S. in case semiconductor supply is cut off." The Terafab that CEO Musk envisions is estimated to be on a scale comparable to TSMC. Currently, Tesla's AI chips are produced by Samsung Electronics (005930.KS) and TSMC, among others, dividing the work.
At Tesla's earnings call on January 28 this year, CEO Musk also argued, "Even in the best-case scenario, when looking at the output of major suppliers and considering supply chains beyond strategic partners like Samsung Electronics, TSMC, and Micron, the amount they can produce is not enough," adding, "We must build the Tesla Terafab to remove the constraints that are highly likely to occur within the next three to four years." CEO Musk also expressed hope at Tesla's earnings call on the 22nd of this month, "We plan to use Intel's advanced process '14A' for Terafab," adding, "It is not yet fully complete, but by the time Terafab fully scales up, 14A will be sufficiently mature or ready for actual deployment." 14A is a 1.4-nanometer (nm) ultra-fine foundry process that Intel is developing with the goal of mass production next year.
Buys Back Ireland Fab, Joins Hands With Google... Foundry Rebuild Accelerates Amid 124% Stock Surge This Year
Intel further announced on the 9th that it had signed a multi-year contract to supply its latest CPU "Xeon 6" and other key semiconductors to Google Cloud servers. Google extended a contract under which it had previously used Intel chips. Intel also agreed to expand cooperation with Google on the joint development of custom infrastructure processing units (IPUs). An IPU is a chip that improves data center efficiency, handling roles such as internal communication, data storage, and security. When IPUs take charge of these tasks, the CPU can concentrate solely on its core function of data processing. CEO Tan argued, "Scaling AI requires not just accelerators but a balanced system," adding, "CPUs and IPUs play a central role in delivering the performance, efficiency, and flexibility of modern AI demand."
News that Intel decided on the 1st of this month to repurchase its stake in a joint venture (JV) related to a semiconductor manufacturing plant in Ireland after two years also caught the eye of Wall Street investors. The joint venture is a company that holds the rights to the semiconductor wafers produced at Fab 34. Fab 34 is considered a key facility producing process products such as Intel's major CPU lineups "Intel Core Ultra" and "Intel Xeon 6." Intel, which was experiencing a cash crunch in 2024, sold a 49% stake in Fab 34 to alternative investment manager Apollo Global Management for $11.2 billion. The cost for Intel to buy back this stake was set at $14.2 billion. The funds needed for the stake repurchase will be raised through cash on hand and about $6.5 billion in new debt. Intel CFO David Zinsner offered a self-assessment, saying, "We now have a stronger financial structure, improved financial health, and an advanced business strategy."
Intel's foundry division also announced on the 15th that it would recruit Samsung Electronics Vice President Han Seung-hoon (Sean Han) as senior vice president and general manager starting next month. This amounts to stepping on the accelerator in rebuilding the foundry business. Intel entered the foundry business in 2012 and poured in massive investment, but withdrew after six years due to poor performance. It then declared its re-entry into the business in 2021, but the gap with competitors had already widened considerably. According to market research firm TrendForce, last year's annual market share in the foundry market was 69.9% for TSMC and 7.2% for Samsung Electronics. Intel does not even rank within the world's top 10 in the foundry market.
As the importance of CPUs is being re-examined in an AI era that had been dominated by GPUs, Intel appears to have, for now, escaped the period of worrying about survival. Forecasts are also emerging in various quarters that Intel's business opportunities will increase further as the AI market shifts to agent competition. Intel's stock price, which stood at just $36.90 at the end of last year, also surged 123.69% through the 24th this year, rising to $82.54.
Morgan Stanley also issued a report on the 19th, analyzing that as AI increases demand for CPUs, the investment domain could expand one step further from existing GPUs. Morgan Stanley said, "As AI shifts from the generation phase to the autonomous action phase, the computing bottleneck is moving toward CPUs and memory chips," estimating that "AI agents could expand the data center CPU market—previously expected to surpass $100 billion by 2030—by an additional $32.5 billion to $60 billion (about 48 trillion to 88 trillion won)."
How far Intel's advance will go appears to depend on the form AI industry evolution takes. This is because, as CPUs—once assessed as outdated technology—come back into the spotlight, the company can rapidly expand its territory into integrated semiconductor businesses such as foundry, based on the cash secured here. Above all, the joint resolve of the U.S. political and business circles to revive the nation's semiconductor business is expected to be Intel's greatest ally.
※ "Trump Stocker" is a column that delivers on-the-ground stories and current-issue analysis related to the U.S. market, companies, policy, politics, and diplomacy that may be helpful for investment in the era of President Donald Trump. Subscribe to receive useful news from the United States.
