India's Adani seeks up to $5 billion investment in Google data center to join AI boom
The logo of the Adani Group is seen on the facade of its Corporate House on the outskirts of Ahmedabad · Reuters
Reuters
Fri, November 28, 2025
How much will Adani Group invest in the Google project?
What is the scale of Google's AI data centre investment in India?
Why is there such high demand for AI data centres?
What does this mean for India's AI infrastructure development?
(Reuters) -India's Adani Group plans to invest up to $5 billion in Alphabet-owned Google's India AI data centre project, an executive said on Friday, as it seeks to cash in on booming demand for data capacity in the world's most populous nation.
In October, Google said it would invest $15 billion over five years to set up an artificial intelligence data centre in the southern state of Andhra Pradesh, its biggest investment in India.
AI requires enormous computing power, pushing demand for specialised data centres that enable thousands of chips to be linked in clusters.
Adani Group CFO Jugeshinder Singh said the Google project could mean an investment of up to $5 billion for Adani Connex - a joint venture between Adani Enterprises and private data centre operator EdgeConneX.
"It's not just Google, there are a lot of parties that would like to work with us, especially when the data centre capacity goes to gigawatt and higher," Singh told reporters on Friday.
Google has committed to spending about $85 billion this year to expand data centre capacity as tech companies invest heavily in infrastructure to meet the booming demand for AI services.
Indian billionaires Gautam Adani and Mukesh Ambani have also unveiled investments in building data centre capacity.
The data centre campus in the port city of Visakhapatnam will have an initial power capacity of 1 gigawatt.
($1 = 89.3660 Indian rupees)
(Reporting by Harshita Meenaktshi and Dhwani Pandya; Editing by Kevin Liffey)
Reliance Industries, JV partners to invest $11 billion in India AI data capacity
A guard walks past the Reliance Industries logo near the entrance of Dhirubhai Ambani Knowledge City in Navi Mumbai · Reuters
Reuters
Wed, November 26, 2025
(Reuters) -A Reliance Industries joint venture will invest $11 billion over five years to develop 1 gigawatt of AI data capacity in the southern Indian state of Andhra Pradesh, the companies and the state's government said on Wednesday.
Canadian multinational company Brookfield Corporation and U.S.-based real estate investment trust Digital Realty are the other partners involved in the joint venture, called Digital Connexion.
The project aims to establish an AI-native data centre campus across 400 acres of land in Andhra Pradesh's Visakhapatnam city.
In October, Google disclosed it will build AI data centre capacity in Visakhapatnam over five years, set to be the tech major's largest-ever AI hub outside of the U.S.
The recent boom in AI, which requires vast amounts of computing data, has fuelled a corporate rush to pour money into the technology globally and has led to an unprecedented growth in data centres across the world.
India's data centre capacity is expected to more than triple to 4.5 gigawatt by 2030 from current levels, according to real estate consultant Colliers.
Last week, Indian IT firm TCS also unveiled a partnership with private equity firm TPG to invest $2 billion in equity to form a joint venture aimed at developing AI data centres.
(Reporting by Abhirami G and Hritam Mukherjee in Bengaluru; Editing by Mrigank Dhaniwala and Janane Venkatraman)
Reliance plans $11bn AI data centre in Visakhapatnam, India
Reliance Industries · Verdict
RanjithKumar Dharma
Fri, November 28, 2025
Reliance Industries, in partnership with Brookfield Asset Management and Digital Realty Trust, has announced plans to invest $11bn by 2030 to develop an AI-focused data centre campus in Visakhapatnam, Andhra Pradesh, India.
Digital Connexion, the JV formed by the three companies, confirmed the signing of a memorandum of understanding (MoU) with the Andhra Pradesh Economic Development Board for this project, reported Bloomberg.
The campus will cover 400 acres (1.6187km²) and is expected to have a capacity of one gigawatt (1GW).
Other technology companies have recently announced similar investments in India.
In October 2025, Google stated its intention to spend approximately $15bn over five years on an AI infrastructure hub, also in Visakhapatnam.
Meanwhile, Amazon has outlined plans to invest $12.7bn towards cloud infrastructure in India by the end of this decade, while OpenAI is exploring the development of a 1GW data centre in the South Asian country.
Also in October 2025, Reliance Industries, through Reliance Intelligence, and Google have announced a strategic partnership to promote the adoption of AI across India.
The collaboration will combine Reliance Industries’ scale and ecosystem with Google’s AI technology to increase AI accessibility and support the development of India’s digital infrastructure.
Reliance Industries is partnering with Google Cloud to provide broader access to the latter’s AI hardware accelerators, including Tensor Processing Units (TPUs), as part of its plan to develop large-scale, clean energy-powered computing infrastructure.
The collaboration also designates Reliance Intelligence as a strategic partner to support the adoption of Gemini Enterprise by Indian organisations.
In a separate announcement, Tata Consultancy Services reported it will work with private equity firm TPG to establish a JV named HyperVault AI Data Centre, with a combined equity investment of Rs180bn ($2.03bn).
The capital will be committed in stages over several years, and the companies are seeking to raise an additional $4.5bn to $5bn through debt financing.
TCS did not provide further information regarding the number or locations of the new data centres.
"Reliance plans $11bn AI data centre in Visakhapatnam, India" was originally created and published by Verdict, a GlobalData owned brand.
The information on this site has been included in good faith for general informational purposes only. It is not intended to amount to advice on which you should rely, and we give no representation, warranty or guarantee, whether express or implied as to its accuracy or completeness. You must obtain professional or specialist advice before taking, or refraining from, any action on the basis of the content on our site.
Illustration shows Reliance logo · Reuters
Reuters
November 14, 2025
BENGALURU (Reuters) -Reliance Industries plans to set up a 1-gigawatt AI data center in India's Andhra Pradesh, the state's chief minister said on Friday, adding to infrastructure capacity in India where the likes of Google and Microsoft have made huge AI investments.
Chief Minister Chandrababu Naidu did not disclose financial details of the investment. Reliance did not immediately respond to a request for comment.
Globally, companies are investing heavily to build new infrastructure to meet booming demand for AI services.
India is a critical growth market where nearly a billion users access the internet. Google last month committed to a $15 billion investment over five years to create an AI data center in Andhra Pradesh, its biggest ever investment in India. Microsoft and Amazon have also poured billions into building data centers in India.
Reliance's planned data center will operate as a twin to its gigawatt-scale AI data center in Jamnagar city in Gujarat state, "together forming one of Asia's strongest AI infrastructure networks," Naidu said.
AI requires enormous computing power, pushing demand for specialised data centers that enable tech companies to link thousands of chips together in clusters.
The Reliance group, led by Indian billionaire Mukesh Ambani, includes Jio, India's leading telecoms carrier, Reliance Retail Ltd, Network18 Media & Investments Ltd and Jamnagar, India's largest oil complex.
(Reporting by Munsif Vengattil in Bengaluru; Editing by Joe Bavier and Susan Fenton)
Reuters
November 14, 2025
BENGALURU (Reuters) -Reliance Industries plans to set up a 1-gigawatt AI data center in India's Andhra Pradesh, the state's chief minister said on Friday, adding to infrastructure capacity in India where the likes of Google and Microsoft have made huge AI investments.
Chief Minister Chandrababu Naidu did not disclose financial details of the investment. Reliance did not immediately respond to a request for comment.
Globally, companies are investing heavily to build new infrastructure to meet booming demand for AI services.
India is a critical growth market where nearly a billion users access the internet. Google last month committed to a $15 billion investment over five years to create an AI data center in Andhra Pradesh, its biggest ever investment in India. Microsoft and Amazon have also poured billions into building data centers in India.
Reliance's planned data center will operate as a twin to its gigawatt-scale AI data center in Jamnagar city in Gujarat state, "together forming one of Asia's strongest AI infrastructure networks," Naidu said.
AI requires enormous computing power, pushing demand for specialised data centers that enable tech companies to link thousands of chips together in clusters.
The Reliance group, led by Indian billionaire Mukesh Ambani, includes Jio, India's leading telecoms carrier, Reliance Retail Ltd, Network18 Media & Investments Ltd and Jamnagar, India's largest oil complex.
(Reporting by Munsif Vengattil in Bengaluru; Editing by Joe Bavier and Susan Fenton)
Google's $15 Billion Bet Sparks India's New Tech Gold Rush
Khac Phu Nguyen
November 17, 2025
This article first appeared on GuruFocus.
For investors who remember the first wave of India's tech rise, N. Chandrababu Naidu's latest pitch may feel like deja vu with a sharper edge. Three decades after he fought off skepticism to lure Bill Gates (Trades, Portfolio) and later host the first visit by a sitting US president to southern India, Naidu is back selling a new story: a rebuilt Andhra Pradesh, aiming to grow faster than the rest of the country even as the state absorbs the loss of Hyderabad and President Donald Trump's 50% tariffs. He described the US-India tension as a temporary setback and suggested a deal between Trump and Prime Minister Narendra Modi could emerge very soon. The spark that has investors paying attention again is Google's (NASDAQ:GOOG) decision to commit $15 billion to data centers over the next five years an amount Naidu noted already exceeds the state's total investment inflows from the past five years.
Naidu is now telling investors that Andhra Pradesh could double India's broader growth pace, targeting a 15% expansion rate and eyeing $1 trillion in investments over the next decade. That ambition sits at the center of a larger redevelopment push: remaking Amaravati as a capital, rebuilding industrial momentum, and widening the state's footprint beyond agriculture. He pointed to Dubai's reinvention and Singapore's ascent as proof that policy, labor reform, and tax clarity could reshape a region's trajectory. His pitch spans data centers, green energy, quantum computing, drones, biotechnology, pharmaceuticals, and even commercial space launches tied to Sriharikota's long-standing role. When asked whether he wants Elon Musk through Tesla (NASDAQ:TSLA) or Chinese automaker BYD (BYDDF) to invest, Naidu said he would welcome them as long as national security rules are followed.
Still, the state's hurdles could be material for investors. Andhra Pradesh is carrying one of India's highest fiscal deficits, grappling with uneven infrastructure, and navigating years of political uncertainty including legal cases against Naidu that he described as political. Yet his strengthened alliance with Modi, reinforced by a recent ruling-bloc win in Bihar, could possibly bring a steadier policy runway through 2029. Naidu said the central government is doing their best, but he emphasized that the state must generate its own revenues and job opportunities as youth unemployment and AI-related anxiety climb. With US-trained talent returning and his Stanford-educated son Nara Lokesh now serving as technology minister, Naidu insisted the fundamentals land, water, people, coastline are already in place. As he put it, he must perform, or otherwise I will perish.
November 17, 2025
This article first appeared on GuruFocus.
For investors who remember the first wave of India's tech rise, N. Chandrababu Naidu's latest pitch may feel like deja vu with a sharper edge. Three decades after he fought off skepticism to lure Bill Gates (Trades, Portfolio) and later host the first visit by a sitting US president to southern India, Naidu is back selling a new story: a rebuilt Andhra Pradesh, aiming to grow faster than the rest of the country even as the state absorbs the loss of Hyderabad and President Donald Trump's 50% tariffs. He described the US-India tension as a temporary setback and suggested a deal between Trump and Prime Minister Narendra Modi could emerge very soon. The spark that has investors paying attention again is Google's (NASDAQ:GOOG) decision to commit $15 billion to data centers over the next five years an amount Naidu noted already exceeds the state's total investment inflows from the past five years.
Naidu is now telling investors that Andhra Pradesh could double India's broader growth pace, targeting a 15% expansion rate and eyeing $1 trillion in investments over the next decade. That ambition sits at the center of a larger redevelopment push: remaking Amaravati as a capital, rebuilding industrial momentum, and widening the state's footprint beyond agriculture. He pointed to Dubai's reinvention and Singapore's ascent as proof that policy, labor reform, and tax clarity could reshape a region's trajectory. His pitch spans data centers, green energy, quantum computing, drones, biotechnology, pharmaceuticals, and even commercial space launches tied to Sriharikota's long-standing role. When asked whether he wants Elon Musk through Tesla (NASDAQ:TSLA) or Chinese automaker BYD (BYDDF) to invest, Naidu said he would welcome them as long as national security rules are followed.
Still, the state's hurdles could be material for investors. Andhra Pradesh is carrying one of India's highest fiscal deficits, grappling with uneven infrastructure, and navigating years of political uncertainty including legal cases against Naidu that he described as political. Yet his strengthened alliance with Modi, reinforced by a recent ruling-bloc win in Bihar, could possibly bring a steadier policy runway through 2029. Naidu said the central government is doing their best, but he emphasized that the state must generate its own revenues and job opportunities as youth unemployment and AI-related anxiety climb. With US-trained talent returning and his Stanford-educated son Nara Lokesh now serving as technology minister, Naidu insisted the fundamentals land, water, people, coastline are already in place. As he put it, he must perform, or otherwise I will perish.
The Google logo is seen outside the company's office in London · Reuters
Reuters
November 14, 2025 2 min read
(Reuters) -Alphabet's Google said on Friday it would invest $40 billion in three new data centers in Texas, as part of its push to expand capacity for artificial intelligence initiatives.
The investment, which will be made through 2027, underscores the intensifying competition among AI and cloud service providers to build infrastructure capable of supporting advanced AI models.
OpenAI, Microsoft, Meta Platforms and Amazon are among companies spending billions in new AI-focused data centers.
Google said one of the new data centers will be in Armstrong County, in the Texas Panhandle, and the other two in Haskell County, a stretch of West Texas near Abilene.
"This investment will create thousands of jobs, provide skills training to college students and electrical apprentices, and accelerate energy affordability initiatives throughout Texas," Alphabet CEO Sundar Pichai said in a statement.
The company will also invest in its existing Midlothian campus and Dallas cloud region, part of its global network of 42 cloud regions.
"Google's $40 billion investment makes Texas Google's largest investment in any state in the country and supports energy efficiency and workforce development in our state," Texas Governor Greg Abbott said in the same statement.
Tech companies have announced massive spending plans this year, with many focusing on expanding their U.S. footprint, as President Donald Trump pushes for investments to maintain the country's edge in the AI sector.
Earlier this week, Anthropic said it would invest $50 billion in data centers across the U.S., including New York and Texas.
Google on Tuesday announced it would invest 5.5 billion euros ($6.41 billion) in Germany in the coming years in a push to expand its infrastructure and data center capacity in Europe's largest economy.
The latest AI investment surge echoes past tech bubbles, with valuations and spending outpacing near-term returns, some analysts and investors have warned. They say demand projections may prove overly optimistic if AI adoption does not grow at a similar pace as capital expenditure.
($1 = 0.8575 euros)
(Reporting by Juby Babu in Mexico City; Editing by Shilpi Majumdar)
Japan’s Ruling Party Eyes $6.5 Billion a Year for Chips, AI
Yoshiaki Nohara
November 25, 2025
BLOOMBERG
Yoshihiro Seki
(Bloomberg) -- Japan’s ruling party aims to secure roughly ¥1 trillion ($6.5 billion) per year to keep supporting the nation’s semiconductor and artificial intelligence sectors, according to a lawmaker who leads such efforts.
Most Read from Bloomberg
Trump’s 90,000-Square-Foot Ballroom Plan Puzzles the Experts
Can Anyone Save Gary, Indiana?
Boston’s South Station Gets a Towering Makeover
In Vancouver, A Court Ruling on Indigenous Land Triggers Real Estate Angst
NYC Gets 12-Minute Helicopter for Suburbanites Avoiding Gridlock
Most of the funding will be secured in a regular budget for the year starting in April rather than in an extra budget for the current fiscal year, according to Yoshihiro Seki, secretary general of the Liberal Democratic Party’s group of lawmakers that supports chip making in Japan.
That will be a change in funding method from the past few years, when the government relied on supplementary budgets to fund its chip revival strategy. The shift is expected to make it easier for the government to secure funding in a stable manner, according to Seki, who spoke to reporters on Thursday after the LDP group met.
Japan has set aside roughly ¥5.7 trillion to support Japan’s semiconductor and AI sectors since 2021, when it created a new strategy to revive domestic chip making, according to the Ministry of Economy, Trade and Industry. Most of the funding has come from extra budgets rather than from regular ones.
“Since we were uncertain whether this approach would succeed, we had been moving ahead only with extra budgets,” Seki said. “But from now on, METI’s share in the regular budget will really jump up. So the idea is that allocations from supplementary budgets will fall, leading to more stable operations.”
In last year’s supplementary budget, about ¥1.5 trillion was earmarked for the efforts, as the first round of funding for former Prime Minister Shigeru Ishiba’s pledge to provide more than ¥10 trillion of fresh public support for the sectors. The ¥1 trillion that the LDP aims to secure will be part of the ¥10 trillion pledge, too, according to Seki.
Of the total funding earmarked, roughly ¥1.7 trillion has been allocated to Rapidus Corp., which aims to mass produce cutting-edge chips by 2027. Micron Technology Inc.’s Hiroshima factory has been awarded ¥774.5 billion.
The overwhelming majority of the world’s advanced artificial intelligence chips are manufactured by Taiwan Semiconductor Manufacturing Co., sparking fears about reliance on an island that China claims as its own.
(Updates with comments from Seki, more details.)
Most Read from Bloomberg Businessweek
Reuters
Sat, November 29, 2025
What strategic benefits does Hiroshima location provide Micron?
What drives the surging demand for HBM chips?
Why are HBM chips in high demand?
How does this investment fit Japan's semiconductor strategy?
Nov 29 (Reuters) - Micron Technology will invest 1.5 trillion yen ($9.6 billion) to build a new plant in Hiroshima in western Japan to produce advanced high-bandwidth memory (HBM) chips, the Nikkei reported on Saturday, citing people familiar with the matter.
The U.S. chipmaker aims to start construction at an existing site in May next year and begin shipments around 2028, with Japan's Ministry of Economy, Trade and Industry providing up to 500 billion yen for the project, the Nikkei said.
Reuters could not immediately verify the report.
To revive its aging semiconductor industry, Japan's government is offering generous subsidies to lure investment from foreign chip makers such as Micron and Taiwan Semiconductor Manufacturing Co (TSMC). It is also funding the construction of a plant that will mass produce advanced logic chips using IBM technology.
Demand for HBM chips is being driven by the growth of artificial intelligence and data centre investment.
The expansion of its plant in Hiroshima will help Micron diversify production away from Taiwan and compete with market leader SK Hynix, the Nikkei said.
($1 = 156.1500 yen)
(Reporting by Rajveer Singh Pardesi in Bengaluru and Tim Kelly in Tokyo; Editing by William
Mizuho Sees Micron (MU) Benefiting From AI Storage Needs and Persistent HDD Shortages
Sheryar Siddiq
November 13, 2025
Micron Technology Inc. (NASDAQ:MU) ranks among the best long-term stocks to buy according to D. E. Shaw. Following virtual investor meetings with company executives, including CFO Mark Murphy, Mizuho analyst Vijay Rakesh maintained an Outperform rating and a $265 price target for Micron Technology Inc. (NASDAQ:MU) shares on November 10. The analyst noted many significant trends from the conversations, including the fact that demand for high bandwidth memory (HBM) is expected to remain strong through 2026-2027.
According to the analyst, memory prices could be supported over the coming years by China’s restricted capacity. Chipmaking regulations are delaying between 30% and 35% of Samsung and Hynix’s memory production, which could keep supply tight and enable Micron Technology Inc. (NASDAQ:MU) to profit from higher prices into 2027 and 2028.
Additionally, he pointed out that Micron’s upcoming U.S. plants in Idaho and New York, which are expected to begin production between 2027 and 2029, may increase its U.S. DRAM share over the next ten years from mid-single digits to about 20%.
Rakesh noted that while AI server demand for QLC enterprise solid-state drives (eSSDs) is rising, capital investment in the NAND flash memory segment continues to be regulated, while hard disk drives (HDDs) remain in short supply.
Micron Technology Inc. (NASDAQ:MU) designs, develops, manufactures, and sells memory and storage products across the world.
While we acknowledge the potential of MU as an investment, we believe certain AI stocks offer greater upside potential and carry less downside risk. If you're looking for an extremely undervalued AI stock that also stands to benefit significantly from Trump-era tariffs and the onshoring trend, see our free report on the best short-term AI stock.
Disclosure: None. This article is originally published at Insider Monkey.
Why Micron and SK Hynix Could Quietly Become the Real AI Winners
Manali Pradhan, CFA, The Motley Fool
Mon, November 24, 2025
Key Points
Memory is becoming a major bottleneck in the AI infrastructure buildout.
Both Micron and SK Hynix enjoy pricing power and multiyear revenue visibility for their HBM and advanced DRAM offerings.
SK Hynix is already the leader in the HBM market, while Micron is focused on expanding its market share.
Investors who are looking for high-potential artificial intelligence (AI) stocks often pick well-known semiconductor players such as Nvidia and Advanced Micro Devices, hyperscalers such as Amazon and Microsoft, or software players like Palantir and SoundHound AI.
But many investors are missing one real AI opportunity that's staring us in the face.
Image source: Getty Images.
AI models continue to get larger and more complex. Additionally, as the systems mature, hyperscalers and enterprises are transitioning from infrequent AI training workloads to more frequent AI inferencing workloads. And now, memory bandwidth and capacity are proving to be bigger bottlenecks than AI computing capacity in large AI clusters.
Within that dynamic, leading memory players such as Micron Technology (NASDAQ: MU) and SK Hynix (OTC: HXSCL) -- already crucial players in the global buildout of AI infrastructure -- are well-positioned to prosper.
Favorable memory market trends
The server clusters built around cutting-edge AI processing chips increasingly require higher-performance, energy-efficient, high-bandwidth memory (HBM) and advanced DRAM in order to operate efficiently. In fact, every new generation of AI GPUs uses increasingly more HBM and advanced DRAM.
So naturally, data center companies are increasing their spending on memory as part of their ongoing AI infrastructure buildouts. The global HBM market is forecast to grow from $17 billion in 2024 to $98 billion in 2030, while HBM's revenue share in the DRAM market is estimated to grow from 18% to 50% in that same time frame. This trend favors Micron and SK Hynix, which stand to benefit from the higher memory content of systems designed to handle large, scalable AI workloads.
Micron and South Korea-based SK Hynix can expect to enjoy solid pricing power and higher margins, as there is currently a chronic shortage of advanced DRAM. According to DigiTimes, many major hyperscalers are securing only 70% of their server DRAM orders, despite accepting 50% price hikes in the fourth quarter.
Micron is capitalizing on this opportunity
U.S.-based Micron has gradually transformed itself from a cyclical DRAM and NAND producer into a full-stack, AI-optimized memory and storage player. The company's performance in its fiscal 2025 (which ended Aug. 28) was impressive, with revenues soaring 49% to $37.4 billion, and non-GAAP diluted earnings per share surging nearly 538% to $8.29. Increasing memory demand from data center operators has been the key growth catalyst. In fiscal 2025, its data center business accounted for 56% of Micron's total revenues.
Micron's DRAM market share was close to 22.5% in September. Management also forecast that its HBM market share would be similar to its DRAM share in the third quarter of calendar 2025, as it ramps its third-generation extended (HBM3E) and fourth-generation (HBM4) HBM products. That was an achievable goal, considering that Micron was already the second-largest player in the HBM market in the second quarter of calendar 2025, with a 21% share, according to market research firm Counterpoint Research.
Micron's HBM strategy seems to have been quite successful, as its HBM revenues reached nearly $2 billion in the fourth quarter of fiscal 2025, which translates to an annual run rate of close to $8 billion. The company has already entered into pricing agreements for a significant chunk of its HBM3E supply in 2026, and expects to sell out its entire 2026 supply in the next few months.
The company is also advancing its technology leadership and has ramped up production of its 1-gamma DRAM node to achieve mature yields 50% faster than the previous-generation process node. The company was the first to ship a 1-gamma DRAM chip and plans to leverage its superior process technology across its entire DRAM portfolio.
SK Hynix remains the leader
While Micron is slowly expanding its market share, SK Hynix is already the leader in the HBM market, with a 62% share as of the second quarter of 2025. SK Hynix has also completed development of its next-generation HBM4 and plans to begin shipping that product in the fourth quarter, followed by a rapid scale-up of production in 2026.
SK Hynix has also finalized its 2026 HBM supply plan for major clients, and management expects HBM supply to remain tight even in 2027. As memory companies increasingly allocate production capacity to HBM, the marketplace is now experiencing shortages of some traditional memory products, too. SK Hynix is also a leader in the DRAM market, with a 35% share in the third quarter. Hence, some of its customers have started issuing pre-purchase orders for DRAM and NAND products for 2026.
The company also delivered exceptional performance in the third quarter, as revenues soared 39% year over year to 24.4 trillion Korean won (about $16.6 billion). Operating profit also rose 62% to an all-time high of 11.4 trillion Korean won (about $7.7 billion).
AI winners
SK Hynix expects the AI memory market to grow by almost 30% annually through 2030. The company is also a key HBM supplier to Nvidia; Micron supplies the GPU leader with smaller HBM volumes.
With a favorable demand environment for their offerings and strong technological advantages, both Micron and SK Hynix are well positioned to keep capturing a significant share of this growing and profitable market.
Manali Pradhan, CFA has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Advanced Micro Devices, Amazon, Microsoft, Nvidia, and Palantir Technologies. The Motley Fool recommends the following options: long January 2026 $395 calls on Microsoft and short January 2026 $405 calls on Microsoft. The Motley Fool has a disclosure policy.
Why Micron and SK Hynix Could Quietly Become the Real AI Winners was originally published by The Motley Fool
Is This the Most Underrated AI Infrastructure Play of the Decade?
AI models continue to get larger and more complex. Additionally, as the systems mature, hyperscalers and enterprises are transitioning from infrequent AI training workloads to more frequent AI inferencing workloads. And now, memory bandwidth and capacity are proving to be bigger bottlenecks than AI computing capacity in large AI clusters.
Within that dynamic, leading memory players such as Micron Technology (NASDAQ: MU) and SK Hynix (OTC: HXSCL) -- already crucial players in the global buildout of AI infrastructure -- are well-positioned to prosper.
Favorable memory market trends
The server clusters built around cutting-edge AI processing chips increasingly require higher-performance, energy-efficient, high-bandwidth memory (HBM) and advanced DRAM in order to operate efficiently. In fact, every new generation of AI GPUs uses increasingly more HBM and advanced DRAM.
So naturally, data center companies are increasing their spending on memory as part of their ongoing AI infrastructure buildouts. The global HBM market is forecast to grow from $17 billion in 2024 to $98 billion in 2030, while HBM's revenue share in the DRAM market is estimated to grow from 18% to 50% in that same time frame. This trend favors Micron and SK Hynix, which stand to benefit from the higher memory content of systems designed to handle large, scalable AI workloads.
Micron and South Korea-based SK Hynix can expect to enjoy solid pricing power and higher margins, as there is currently a chronic shortage of advanced DRAM. According to DigiTimes, many major hyperscalers are securing only 70% of their server DRAM orders, despite accepting 50% price hikes in the fourth quarter.
Micron is capitalizing on this opportunity
U.S.-based Micron has gradually transformed itself from a cyclical DRAM and NAND producer into a full-stack, AI-optimized memory and storage player. The company's performance in its fiscal 2025 (which ended Aug. 28) was impressive, with revenues soaring 49% to $37.4 billion, and non-GAAP diluted earnings per share surging nearly 538% to $8.29. Increasing memory demand from data center operators has been the key growth catalyst. In fiscal 2025, its data center business accounted for 56% of Micron's total revenues.
Micron's DRAM market share was close to 22.5% in September. Management also forecast that its HBM market share would be similar to its DRAM share in the third quarter of calendar 2025, as it ramps its third-generation extended (HBM3E) and fourth-generation (HBM4) HBM products. That was an achievable goal, considering that Micron was already the second-largest player in the HBM market in the second quarter of calendar 2025, with a 21% share, according to market research firm Counterpoint Research.
Micron's HBM strategy seems to have been quite successful, as its HBM revenues reached nearly $2 billion in the fourth quarter of fiscal 2025, which translates to an annual run rate of close to $8 billion. The company has already entered into pricing agreements for a significant chunk of its HBM3E supply in 2026, and expects to sell out its entire 2026 supply in the next few months.
The company is also advancing its technology leadership and has ramped up production of its 1-gamma DRAM node to achieve mature yields 50% faster than the previous-generation process node. The company was the first to ship a 1-gamma DRAM chip and plans to leverage its superior process technology across its entire DRAM portfolio.
SK Hynix remains the leader
While Micron is slowly expanding its market share, SK Hynix is already the leader in the HBM market, with a 62% share as of the second quarter of 2025. SK Hynix has also completed development of its next-generation HBM4 and plans to begin shipping that product in the fourth quarter, followed by a rapid scale-up of production in 2026.
SK Hynix has also finalized its 2026 HBM supply plan for major clients, and management expects HBM supply to remain tight even in 2027. As memory companies increasingly allocate production capacity to HBM, the marketplace is now experiencing shortages of some traditional memory products, too. SK Hynix is also a leader in the DRAM market, with a 35% share in the third quarter. Hence, some of its customers have started issuing pre-purchase orders for DRAM and NAND products for 2026.
The company also delivered exceptional performance in the third quarter, as revenues soared 39% year over year to 24.4 trillion Korean won (about $16.6 billion). Operating profit also rose 62% to an all-time high of 11.4 trillion Korean won (about $7.7 billion).
AI winners
SK Hynix expects the AI memory market to grow by almost 30% annually through 2030. The company is also a key HBM supplier to Nvidia; Micron supplies the GPU leader with smaller HBM volumes.
With a favorable demand environment for their offerings and strong technological advantages, both Micron and SK Hynix are well positioned to keep capturing a significant share of this growing and profitable market.
Manali Pradhan, CFA has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Advanced Micro Devices, Amazon, Microsoft, Nvidia, and Palantir Technologies. The Motley Fool recommends the following options: long January 2026 $395 calls on Microsoft and short January 2026 $405 calls on Microsoft. The Motley Fool has a disclosure policy.
Why Micron and SK Hynix Could Quietly Become the Real AI Winners was originally published by The Motley Fool
Is This the Most Underrated AI Infrastructure Play of the Decade?
Rick Orford,
The Motley Fool
November 23, 2025
Key Points
Micron Technology is pivotal in AI with its HBM and Nvidia's validation tools.
Supply constraints and HBM shortages give Micron pricing power and margin potential.
Analysts rate MU a strong buy as HBM4 and SOCAMM2 drive the company's growth forward.
November 23, 2025
Key Points
Micron Technology is pivotal in AI with its HBM and Nvidia's validation tools.
Supply constraints and HBM shortages give Micron pricing power and margin potential.
Analysts rate MU a strong buy as HBM4 and SOCAMM2 drive the company's growth forward.
Micron Technology (NASDAQ: MU) isn't getting a lot of attention right now despite being one of the world's three dominant memory chip manufacturers. It produces the memory and storage components inside nearly every computing device we use: DRAM and NAND Flash memory.
Moreover, while almost every company in this space is South Korean, which includes Samsung Electronics and SK Hynix, Micron is the only U.S.-based memory chip manufacturer, which could give it a boost in the current geo-political environment.
Let's see why investors would do well to take a closer look at this stock now.
Image source: Getty Images.In AI, what's driving Micron's momentum is HBM, or high bandwidth memory. This is a relatively new technology that stacks multiple memory chips vertically, rather than spreading them out flat. And, to Micron's credit, this technology has become essential in the AI infrastructure buildout, especially for Nvidia.
Micron's partnership with Nvidia
Micron confirmed that HBM3E chips will power Nvidia's latest Blackwell architecture. Nvidia CEO Jensen Huang specifically mentioned that it's using Micron's G7 memory, which delivers 1.8 terabytes per second. At the same event, Nvidia introduced its GeForce RTX 50 series GPUs and a desktop-sized AI supercomputer called Project DIGITS, which utilizes Micron's DDR5X memory with 128GB capacity. Huang is arguably the most influential figure in the AI space right now, so this is a huge validation for Micron.
Moreover, Micron began sending samples of its 192GB SOCAMM2 memory modules to customers back in October. These memory chips are designed specifically for AI data centers and were developed through a five-year partnership with Nvidia. Reportedly, these modules use one-third the power of standard memory while delivering 2.5 times higher bandwidth, which makes them a must-have for AI infrastructure.
How has Micron stock performed?
Micron stock currently trades at about $225 per share. Over the past year, it has gained over 130%, and over five years, it's up nearly 270%. Those returns have significantly outpaced the broader market, which also reflects investor enthusiasm for Micron's position in high-growth markets amid demand for AI-optimized memory.
HBM market share expansion in a supply-constrained market
According to reports, Micron held less than 10% of the global HBM market back in 2022. Today, the company aims to grow its share to somewhere between 20% and 25% through 2026. This is entirely possible as the global HBM market itself is expected to grow 26% to $9.2 billion in 2026, and more than double by 2028. Micron could capture a significant portion of that growth, given its position in the AI infrastructure buildout.
Moreover, there are industrywide shortages of HBMs. SK Hynix's and Micron's 2025 output is already largely sold out, and management expects these tight supply conditions to continue through 2025 and into 2026, supporting healthy profit margins. Micron also plans to start shipping HBM4 in 2026, its next-generation HBM.
The company is in a supply-constrained environment in which demand exceeds supply, so Micron inevitably gets pricing power. It can charge more since customers need the product and alternatives are limited, and that pricing power directly helps profitability.
Verdict
Is Micron Technology a buy at these levels? Consider that a consensus among 37 Wall Street analysts rates Micron stock a "strong buy," and that rating has been gradually improving over the last three months. In light of Micron's role in the AI ecosystem and its overall financial performance, I would agree.
For those who are looking to invest in the AI space without paying a premium, Micron is a different name that doesn't dominate the headlines. Not only that, the market might not yet fully recognize Micron's critical role in the AI infrastructure stack. Sure, it probably won't capture the same margins as chip designers and manufacturers, but still, it's an indispensable supplier to the AI ecosystem, and investors typically pay a premium to get that exposure.
For these reasons, Micron presents a compelling investment opportunity for those seeking AI exposure and growth without paying a premium that's commanded by better-known names.
Rick Orford has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Nvidia. The Motley Fool has a disclosure policy.
Why the Selloff in Micron Technology Stock Never Made Sense
Rich Duprey
Mon, November 24, 2025
Micron Technology Inc.
Quick Read
Micron Technology (MU) controls 23% of the global DRAM market and 12% of NAND flash.
UBS raised global HBM demand forecasts to 28 billion gigabits by 2026.
Micron trades at a forward P/E of 10 versus the semiconductor industry average of 25x.
Micron Technology (NASDAQ:MU) shares soared to an all-time high of $260 earlier this month, riding the wave of AI-driven optimism in the semiconductor sector. But like most chip and tech stocks, the memory chip maker has trended lower since, caught in broader market jitters over AI hype fears.
Last week, the selloff intensified, with Micron plunging 10% in a single session -- triggered partly by Nvidia's (NASDAQ:NVDA) earnings report, which, despite beating estimates, highlighted rising costs and slower data center growth. Investors panicked, dumping memory plays like Micron amid fears of overcapacity. Yet this drop never made sense. Here's why.
An Iron Grip on Memory Markets
Micron stands out thanks to its commanding position in DRAM and NAND flash memory. DRAM, essential for temporary data storage in servers, PCs, and smartphones, accounts for over half of Micron's revenue. The company controls about 23% of the global DRAM market, behind only Samsung and SK Hynix.
NAND flash, used for long-term storage in solid-state drives (SSDs) and mobile devices, adds another 40% to sales, with Micron holding a 12% share. These segments benefit from cyclical upturns, but AI is turning cycles into a structural boom.
What sets Micron apart is its pivot to high-bandwidth memory (HBM), the ultra-fast DRAM variant critical for AI accelerators. HBM enables the massive data throughput needed for training large language models.
Micron began shipping HBM3E samples in 2024 and is ramping production this year, positioning it as a key supplier alongside SK Hynix. In the AI era, where Nvidia's GPUs and Advanced Micro Devices (NASDAQ:AMD) Instinct chips demand terabytes of high-speed memory per rack, Micron's HBM expertise locks in long-term growth.
Analysts project the HBM market exploding from $4 billion in 2023 to nearly $100 billion by 2030, with Micron capturing a growing slice through its advanced 1-gamma node tech, which boosts density and efficiency.
This dominance isn't just defensive -- it's a growth engine. Micron's fiscal fourth quarter results showed 46% year-over-year revenue growth to $11.3 billion, driven by AI server demand and strong data center sales, which now account for 56% of total revenue. Even as consumer markets like PCs and smartphones recover slowly, AI hyperscalers like Microsoft and Amazon are stockpiling Micron's memory for cloud infrastructure.
Supply constraints in advanced nodes, including limited HBM production capacity, shield margins from commoditization risks. Micron's capex investments -- $13.8 billion in fiscal 2025 -- target expanding HBM output to 20% of DRAM sales by 2026, ensuring it rides the AI wave without overextending.
Wall Street's HBM Boost
Adding fuel this morning, MU stock climbed nearly 6% on no company-specific news, but a fresh UBS report lit the spark. The investment firm upped its global HBM demand forecast, now expecting 17.3 billion gigabits of end-consumption this year (a 1% increase from prior estimates) and 28 billion gigabits in 2026 (up another 1.6%).
This revision stems from robust procurement by Nvidia and AMD, including higher unit forecasts for Nvidia's next-gen Rubin GPUs (3 million in 2026) and AMD's MI450 series, tied to deals like OpenAI's supercomputer builds. Taiwan Semiconductor Manufacturing's capacity expansions and tight node utilization through late 2026 further tighten supply.
While UBS spotlighted SK Hynix and Samsung as direct buys -- citing SK Hynix's 70% share of Nvidia's HBM4 supply and 13% blended average selling price (ASP) growth -- the read-through for Micron is clear and positive.
As the third major HBM player, Micron benefits from the same demand surge, especially with its U.S.-based fabs qualifying for CHIPS Act subsidies. This could lift Micron's HBM annual revenue run rate of $8 billion far higher by 2027, according to consensus estimates, pushing overall earnings higher.
Gross margins were already at 45% in Q4, and stand to expand as HBM commands 5x to 10x premiums over standard DRAM. For MU going forward, this means accelerated free cash flow ($3.7 billion in 2025) funding dividends, buybacks, and more AI R&D. The selloff ignored these fundamentals, creating a disconnect from reality.
Key Takeaway
The recent plunge in Micron Technology stock was a textbook overreaction, handing savvy investors a rare discount on a proven AI winner. Even after today's 6% bounce, Micron remains a compelling buy.
Trading at a forward P/E of just 10 compared to the semiconductor industry's 25x makes Micron's dirt cheap for its prospects. Wall Street forecasts over 40% compound annual EPS growth through 2030, yielding a PEG ratio under 1, a hallmark of undervalued growth.
In a sector buzzing with pricier picks such as Nvidia or AMD, MU offers the best risk-reward for AI exposure.
