Samsung Electronics is attempting to reclaim lost ground in the artificial intelligence memory race by moving aggressively into next-generation high-bandwidth memory, unveiling sample shipments of its new 12-layer HBM4E chips months ahead of broader market adoption.

The announcement marks a critical moment for Samsung, which has spent the past two years watching rivals SK Hynix and Micron Technology capture much of the explosive demand tied to AI servers and accelerators, particularly within the supply chain of Nvidia.

Samsung said the new HBM4E chips are more than 20% faster than its previous HBM4 generation and are built using its sixth-generation 10-nanometer-class DRAM process alongside a 4-nanometer logic base die manufactured through its foundry business.

The move has put the industry on notice because the AI semiconductor race is increasingly being shaped not just by compute chips themselves, but by access to advanced memory capable of feeding enormous volumes of data into AI accelerators fast enough to avoid bottlenecks.

HBM has become one of the most critical technologies in modern AI infrastructure. Advanced AI systems from Nvidia, AMD, and Google rely heavily on stacked memory architectures that deliver dramatically higher bandwidth and energy efficiency than conventional DRAM. That has transformed HBM from a niche product into one of the semiconductor industry’s most profitable and strategically important segments.

Samsung’s latest move comes only three months after it began shipping HBM4 samples, an unusually fast progression that signals urgency inside the company to close the gap with SK Hynix, which currently dominates the market.

According to Counterpoint Research, SK Hynix held 57% of the global HBM market in the fourth quarter of 2025, compared with Samsung’s 22% and Micron’s 21%.

Analysts say timing matters enormously in the HBM business because early suppliers often lock in multiyear customer relationships tied to AI infrastructure buildouts worth tens of billions of dollars.

“In the HBM market, early movers tend to secure the bulk of orders,” said Jeff Kim of KB Securities-Jefferies, pointing to Samsung’s delayed entry into earlier HBM3 and HBM3E cycles.

The stakes are especially high because hyperscalers and AI model developers are now racing to secure long-term memory supply amid fears of shortages as AI compute demand accelerates globally.

Samsung’s customer list already includes major AI players such as Advanced Micro Devices, Nvidia, and Google. But qualification by Nvidia remains the industry’s most important benchmark because Nvidia systems dominate the AI training market.

Samsung shares rose as much as 6.5% following the announcement, substantially outperforming South Korea’s broader market, as investors bet the company may finally be regaining momentum in AI semiconductors.

Part of that optimism also stems from Samsung’s expanding role beyond memory.

Earlier this year, AI company Anthropic identified Samsung as a strategic infrastructure partner in a funding round that valued the startup at $965 billion. Notably, Samsung was singled out not only for memory capabilities but also for logic-chip manufacturing, highlighting growing expectations that the company could become a larger beneficiary of the AI infrastructure boom through its foundry business.

That matters because Samsung is one of the very few companies globally capable of manufacturing advanced chips at scale outside of Taiwan Semiconductor Manufacturing Company.

With TSMC’s advanced-node capacity expected to remain heavily booked for years amid surging AI demand, some analysts believe Samsung could increasingly emerge as an alternative manufacturing partner for AI chip designers seeking additional supply flexibility. The company already strengthened that narrative last year after announcing a $16.5 billion chip supply agreement with Tesla.

Samsung’s broader challenge, however, goes beyond simply launching newer products. The company must convince customers that it can consistently deliver top-tier yields, power efficiency, and manufacturing reliability in an industry where execution failures can delay entire AI server deployments.

That is particularly important in HBM, where packaging complexity has become nearly as important as transistor performance. Advanced HBM stacks require precise thermal management, ultra-dense interconnects, and high production yields to meet hyperscaler requirements.

Samsung’s aggressive HBM4E rollout is seen as an indication that the company understands the path the next phase of the AI boom is taking. Analysts believe the next phase may be determined not only by who designs the most powerful AI chips, but by who controls the surrounding memory and manufacturing ecosystem needed to keep those chips running at scale.

In effect, the battle over AI dominance is increasingly becoming a supply-chain war.

Dell Technologies delivered its strongest earnings report since returning to public markets in 2018, marking loudly how the artificial intelligence spending wave is reshaping the fortunes of once-mature hardware companies and redrawing the competitive market of enterprise technology.

The company’s shares surged about 30% in after-hours trading on Thursday after it posted quarterly results that comfortably exceeded Wall Street expectations, powered by explosive demand for AI infrastructure, servers, and enterprise computing systems tied to generative AI deployment.

Dell reported first-quarter revenue of $43.84 billion, sharply above analyst estimates of $35.43 billion. Earnings per share came in at $4.86, also far ahead of forecasts of $2.94. The stock had already climbed nearly 4% during regular trading, closing at $317.05 before the earnings release triggered a further rally.

The results highlight how Dell, long associated with the personal computer boom of the early 2000s and its famous “Dude, you’re getting a Dell” advertising campaign, has transformed itself into a critical supplier of AI infrastructure for corporations, cloud providers, and governments racing to expand computing capacity.

The company’s resurgence is centered on its rapidly expanding AI server business. Dell disclosed that AI-related revenue surged to more than $16 billion year over year and projected that AI-derived revenue could reach $60 billion in 2026, a figure that would place the company among the biggest beneficiaries of the global AI infrastructure buildout.

The forecast also underlines the extraordinary scale of spending now flowing into data centers, accelerated computing, and networking equipment as enterprises attempt to deploy increasingly complex AI models. Demand for AI servers has become one of the fastest-growing segments in the global technology market, benefiting hardware makers closely aligned with chip giants such as NVIDIA.

Dell also guided full-year company-wide revenue to between $165 billion and $169 billion, signaling confidence that corporate AI spending remains in its early stages rather than nearing a peak.

Legacy infrastructure firms in the tech industry that spent years battling slowing growth are now finding renewed momentum from AI-driven demand. Investors have increasingly rotated back into companies viewed as essential suppliers of the physical backbone underpinning generative AI systems.

That dynamic has fueled sharp rallies across several former internet-era leaders. Cisco Systems shares have climbed 56%, while Intel has surged more than 200% amid growing optimism that AI-related compute demand could revive parts of the semiconductor and networking industry that had struggled through years of cyclical weakness.

Dell’s performance also came off as part of a major change in enterprise purchasing priorities. Corporate customers are now spending aggressively on high-performance servers, storage systems, and networking gear capable of supporting AI training and inference workloads, even as broader IT spending in some traditional areas remains subdued.

The company has positioned itself as a one-stop provider of AI infrastructure, combining servers, storage, cooling systems, and cloud integration services. Analysts say that an integrated approach is helping Dell secure large enterprise and government contracts at a time when organizations are looking for turnkey AI deployment solutions rather than piecing together systems from multiple vendors.

Government demand is emerging as another powerful tailwind. Dell recently secured a nearly $10 billion contract with the U.S. Department of Defense, further strengthening its position in federal technology infrastructure and AI modernization efforts.

The results are also believed to wield broader impacts for financial markets. Investors who once viewed AI primarily as a software or semiconductor story are increasingly recognizing that the boom requires massive investment in physical infrastructure, from data centers and servers to energy systems and advanced networking.

That has widened the pool of AI beneficiaries well beyond chipmakers alone.

Dell’s resurgence became a broad market due to the company’s history. Founder Michael Dell took the company private in 2013 in a controversial deal aimed at restructuring the business away from the pressures of quarterly reporting. The company returned to public markets in late 2018 through a complex financial transaction involving VMware.

At the time, many investors still viewed Dell primarily as a PC maker operating in a slowing market. The latest earnings suggest the company has successfully repositioned itself as a central player in the AI infrastructure economy.

Dell’s comeback also meets the intensifying competition among hardware and cloud providers to capture the next phase of AI spending. Tech companies globally are racing to build AI-ready computing environments as demand accelerates for agentic AI systems, enterprise copilots, and large-scale inference workloads.

Investors now appear to be betting that Dell’s infrastructure footprint gives it an advantage in that race, particularly as organizations seek alternatives to building AI systems entirely through hyperscale cloud providers.

The scale of Dell’s projected AI revenue growth suggests the company expects the current AI spending cycle to persist for years, not quarters. That outlook aligns with forecasts from major technology firms and investment banks that estimate global AI infrastructure spending could reach several trillion dollars over the coming decade.