By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel consensus is bearish on Seagate (STX) and Western Digital (WDC) due to the risk of demand normalization, price compression from NAND flash, and potential margin erosion from geopolitical supply chain risks and workload shifts towards NVMe/SSD.
Risk: Demand normalization and price compression from NAND flash
Opportunity: None identified
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
One sometimes-overlooked aspect of the artificial intelligence (AI) trend is that nearly every aspect of the technology, from training to inference to physical AI, creates data that needs to be stored on hard disk drives.
Two major U.S. technology companies that supply those resources are Western Digital (NASDAQ: WDC) and Seagate Technology (NASDAQ: STX). Their shares are up by around 190% and 200%, respectively, year to date, and it appears their growth cycles are still ramping up. Analysts' average price target for Western Digital is around 9% higher than its current price, while Seagate's average price target is about 5% higher than it is trading today.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
Here are three reasons I like each stock.
While the early AI hype focused heavily on the computing power required to handle such systems, the market is coming to realize that AI is extremely data-heavy. Large language models (LLMs) require astronomical amounts of storage to house training data, logs, and synthetic data outputs.
Western Digital is seeing explosive demand from hyperscale cloud data center clients. In its third-quarter earnings call, management noted that 100% of its 2026 hard disk drive (HDD) production capacity is sold out, backed by multiyear purchase agreements that will extend through 2028 and 2029.
Seagate can manufacture 550 exabytes of data storage annually, but that's not enough to keep up with demand either. The company said that its nearline exabyte capacity is largely allocated and locked in through 2027, with strategic planning stretching into 2028. These build-to-order contracts give the company a rare degree of visibility into its future revenue streams.
Because data center clients are racing to secure hardware faster than it can be manufactured, the whole tech industry is experiencing shortages of storage hardware. This supply crunch gives both companies immense pricing power.
In its fiscal 2026 third quarter, which ended April 3, Western Digital reported revenue of $3.34 billion, up 45%, year over year. Gross margins jumped past 50%, leading to net income rising by 516% to $3.2 billion. EPS rose to $8.20, up 201% year over year.
Seagate had similarly strong numbers in its fiscal third quarter, which likewise ended April 3. Revenue was $3.1 billion, up 44% year over year, while its gross margin rose by 1030 basis points to 46.5%. Net income rose 120% to $748 million, and EPS was up 108% to $3.27.
Western Digital is maintaining its industry lead through next-generation, high-capacity drive innovations, successfully advancing its energy-assisted magnetic recording (ePMR) and heat-assisted magnetic recording (HAMR) technologies to deliver higher-density UltraSMR drives with capacities up to 32 terabytes. The company is also developing high bandwidth drive and dual pivot technologies to maximize throughput for intensive AI workloads.
This technological execution is translating into massive free cash flow. Western Digital has aggressively improved its balance sheet to a net cash position, allowing management to reward shareholders. Western recently raised its quarterly dividend by 20% and authorized an expansive $4 billion share-repurchase program.
Seagate was an early adopter of HAMR and maintains an edge in HAMR sales through its Mozaic drive platform. It has qualified and deployed these ultra-high-capacity drives (32 terabytes and above) to roughly 75% of leading global cloud service providers. The tech allows hyperscalers to upgrade their data centers' capacities without expanding their physical footprints or power grid connections.
Like Western Digital, Seagate is benefiting from its strong cash position of $1.1 billion, up 40% from the same quarter a year ago. In its fiscal Q3, it retired $641 million in debt and returned $191 million to shareholders through dividends and share repurchases. Its board also approved an additional stock buyback program of up to $5 billion. The company raised its quarterly dividend by 3% this year to $0.74 per share.
Between the two, Western Digital trades at lower valuations right now, both on a trailing and a forward basis. It's also in a better position regarding debt.
While Seagate has an edge in technology, Western Digital appears to be catching up, and it has higher margins. Both companies are solid choices, but I feel Western has more momentum for now. Seagate's share price run-up this year makes the stock less of a buy.
Before you buy stock in Western Digital, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Western Digital wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $442,220! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,230,114!
Now, it’s worth noting Stock Advisor’s total average return is 926% — a market-crushing outperformance compared to 203% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
**Stock Advisor returns as of June 11, 2026. *
James Halley has positions in Seagate Technology Plc. The Motley Fool has positions in and recommends Western Digital. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
Four leading AI models discuss this article
"The AI memory boom is likely cyclical, and investors should price in a potential re-rating risk as supply catches up and demand normalizes."
The article frames Seagate (STX) and Western Digital (WDC) as clear beneficiaries of an AI storage boom, highlighting supply shortages, rising margins, and strong buybacks. Yet the thesis hinges on two risky assumptions: that AI/data-center storage demand remains robust through 2027 and that HDD pricing/margins stay elevated as supply tightens. In reality, the memory cycle is highly cyclical: hyperscaler capex could slow, workloads may migrate toward SSD/NVMe or storage-class memory, and price discipline can erode quickly after capacity additions. Even with near-term cash flow strength, the longer-run upside may hinge on deployments that outpace subsequent capacity, a risky bet if demand normalizes sooner than expected.
The strongest counter is that the AI memory surge could prove temporary; once 2026-2027 capacity comes online, pricing power fades, and HDDs lose share to faster SSD solutions, curbing both earnings and multiples.
"The current valuation of these HDD manufacturers assumes a permanent shift in pricing power that ignores the inevitable commoditization once supply-side capacity constraints are resolved."
The article captures the cyclical tailwind of HDD demand, but it ignores the structural threat of NAND flash price compression. While Seagate (STX) and Western Digital (WDC) are currently enjoying a 'super-cycle' due to hyperscaler capacity constraints, the long-term shift toward QLC (quad-level cell) NAND flash in data centers remains a significant risk. As flash density improves and costs drop, HDDs will be relegated to 'cold storage' only, limiting the TAM (total addressable market) expansion. Furthermore, the 200% YTD run-up has already priced in aggressive margin expansion. Investors are chasing a rearview mirror; once supply-side capacity catches up in 2026, the current pricing power will likely evaporate, leading to a classic commodity-cycle crash.
If the AI-driven data explosion outpaces the cost-reduction curve of flash memory, HDDs could remain the only economically viable 'warm' storage tier for the next five years, sustaining these elevated margins longer than history suggests.
"Capacity sold through 2028 is a bullish *near-term* signal but a bearish *long-term* one, because it frontloads demand and locks in prices before the market knows if AI storage needs will actually materialize or shift to faster, cheaper alternatives."
The article conflates a cyclical supply shortage with structural AI demand. Yes, WDC and STX have 100% capacity sold through 2027-2028, but that's a *contract backlog*, not proof of perpetual growth. The real risk: once capacity catches up (new fabs, competitors ramping), these multiyear agreements lock in *prices* that may crater. Gross margins at 46-50% are cyclical peaks, not sustainable. The 200%+ YTD gains already price in years of this euphoria. Both companies are capital-intensive; free cash flow can evaporate if capex spikes or demand normalizes. The article ignores that AI workloads are shifting toward GPU-attached NVMe and in-memory compute—HDD may not be the bottleneck in 2-3 years.
If hyperscalers truly need 550+ exabytes annually and can't build it themselves, these multiyear contracts ARE structural, not cyclical—and the installed base of legacy HDDs means replacement demand alone sustains margins for years.
"Massive YTD gains plus minimal analyst upside targets already embed most of the visible AI storage tailwinds."
Both STX and WDC have already surged 190-200% YTD on AI-driven HDD shortages, yet analyst price targets sit just 5-9% above current levels. The article highlights sold-out capacity through 2028-29 and margin expansion, but overlooks how hyperscalers could accelerate flash adoption or internalize storage to bypass pricing power. Cyclical inventory swings and potential HAMR/ePMR delays also remain unaddressed. With net cash positions and buybacks in place, near-term EPS beats look likely, yet the valuation reset already appears largely priced in.
The multiyear build-to-order contracts could deliver sustained revenue visibility that justifies holding through any near-term digestion of the 2025 rally, especially if AI training data volumes exceed even current forecasts.
"Backlog is not a durable moat; rapid NVMe/SSD adoption and capacity growth could erode HDD pricing power within 2–3 years."
Claude’s backlog critique underplays the risk that backlog can lock in margins only temporarily. Backlogs signal demand, but they don’t guarantee durable pricing if new capacity comes online and market structure shifts. The overlooked risk is hyperscalers accelerating NVMe/SSD adoption or re-architecting storage, which could compress HDD pricing power within 2–3 years even with solid demand. That matters for valuation beyond the next few quarters.
"Geopolitical supply chain risks and potential forced localization are being ignored in the valuation of STX and WDC."
Claude, you’re missing the geopolitical dimension. Both STX and WDC have massive manufacturing footprints in Southeast Asia and China. As US-China tech decoupling accelerates, these firms face existential supply chain risks that go beyond mere cyclicality or NAND competition. If hyperscalers view HDD supply as a national security bottleneck, they won't just wait for capacity—they will force domestic sourcing or aggressive diversification, potentially cratering margins through forced, inefficient capital expenditure requirements that the market is currently ignoring.
"Geopolitical supply fragmentation compresses margins but doesn't solve the structural risk that AI workloads may not require the HDD volumes the backlog assumes."
Gemini's geopolitical angle is sharp but overstated. US-China decoupling affects *margins*, not demand—hyperscalers will source HDDs domestically at lower efficiency, yes, but that's a 2-3 year friction, not an existential threat. More pressing: nobody's quantified how much AI workload actually *needs* HDD vs. NVMe. If training clusters run hot on GPU-attached flash and only archive to HDD, the 'sold-through 2028' backlog becomes a cold-storage rump. That's the real margin cliff.
"Geopolitical decoupling plus NVMe shifts could erode HDD margins faster than Claude's 2-3 year friction timeline."
Claude underestimates how US-China decoupling could force hyperscalers to accelerate NVMe adoption as a diversification play, not just accept lower-efficiency domestic HDDs. Gemini flags supply-chain exposure correctly; combined with workload shifts to GPU-attached flash, this shortens the 2028 backlog durability. The margin cliff arrives sooner if export controls limit tech transfers for next-gen HDDs like HAMR.
The panel consensus is bearish on Seagate (STX) and Western Digital (WDC) due to the risk of demand normalization, price compression from NAND flash, and potential margin erosion from geopolitical supply chain risks and workload shifts towards NVMe/SSD.
None identified
Demand normalization and price compression from NAND flash