By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
Panelists agree that AI demand is a real driver for Micron's memory chips, but disagree on the sustainability of the current pricing and the potential for a supply glut. They also highlight the risk of a sudden shift in AI spending and the potential impact of geopolitical factors on pricing.
Risk: A sudden shift in AI spending or an acceleration of oversupply could compress margins and discipline the multiple MU trades at.
Opportunity: Micron's reliance on non-Chinese fabs could provide a structural advantage as export controls tighten.
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 →
Memory-chip maker Micron Technology (NASDAQ: MU) is on a roll. The stock has delivered a total return of 130% over the last year, including a 64% surge in three months.
It feels funny talking about how big a company named after a tiny measurement might get, but that's the question on every Micron investor's mind. Is the stock still primed for further gains or is it more likely to run into a glass ceiling and back down again?
The catalysts behind Micron's stock surge
Micron benefits from a couple of robust market trends. The computer systems that create, train, and operate high-powered artificial intelligence (AI) tools require a metric truckload of memory. One Nvidia (NASDAQ: NVDA) V100 AI accelerator card comes with 32 gigabytes of high-speed memory, for example, and a single AI server can use thousands of these cards.
And V100 isn't even Nvidia's latest or greatest AI accelerator anymore. The brand-new GB200 card comes with more than half a terabyte of high-speed SDRAM each, which is roughly 100 times the V100's memory capacity.
Long story short, AI computers need a lot of memory. That's good news for leading chip suppliers like Micron.
At the same time, the memory market is cyclical and currently on an upswing from the last downturn. Predicting future demand for computing hardware is notoriously difficult due to rapid technological advancements and shifting market trends.
Nobody foresaw the COVID-19 crisis or the generative AI boom. Hence, the memory sector is almost always either trying to meet unexpectedly huge product demand or stuck with warehouses full of oversupply.
At the moment, chip prices are high and rising, thanks to the aforementioned AI boom, the rise of electric and self-driving vehicles, and the aftershocks of a semiconductor manufacturing crunch that started with the coronavirus pandemic.
Micron's financial outlook
Micron's stock is rising for several good reasons. When they're all taken together, the memory market is poised for several years of overwhelming product demand, driving chip prices higher while widening Micron's profit margins. And it's still early in the positive part of a familiar cycle -- at $18.3 billion, Micron's trailing sales are 44% below the most recent peak of $32.4 billion.
The company's shares aren't exactly on fire sale. The stock is reaching all-time record prices on a nearly daily basis right now, far exceeding even the lofty peaks of the dot-com era.
You can't call it "expensive," either. Micron shares are changing hands at quite reasonable valuation ratios, such as 9.3 times sales and 17 times forward earnings projections. These ratios should shrink as the economic upsides I discussed earlier are converted into robust revenue, earnings, and cash flow.
Should you invest in Micron now?
Micron is well-positioned to benefit from the ongoing AI boom and increasing semiconductor demand in electric vehicles. Smartphones are also gaining AI features, boosting their need for high-speed memory chips. The company is ready to supply these unstoppable market trends.
The stock's recent gains reflect these positive signals, but is it too late to buy Micron stock? The answer is no.
I think there's more upside ahead for Micron investors. The cyclical nature of the industry means a downturn is inevitable, possibly within the next two to three years -- but this is still the start of a classic industry upswing. Investors should see this as a chance to capitalize on Micron's strengths. In my mind, Micron's soaring stock has the fuel to keep rising from here.
Keep a close eye on the underlying market forces, perhaps preparing to take some profits off the table when the tide starts to turn. Or you can do it my way, holding Micron stock for the long run and adding more on the cyclical price dips.
Before you buy stock in Micron Technology, 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 Micron Technology wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $830,777!*
Stock Advisor provides investors with an easy-to-follow blueprint for success, including guidance on building a portfolio, regular updates from analysts, and two new stock picks each month. The Stock Advisor service has more than quadrupled the return of S&P 500 since 2002*.
*Stock Advisor returns as of June 10, 2024
Anders Bylund has positions in Micron Technology and Nvidia. The Motley Fool has positions in and recommends Nvidia. 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
"Micron's current upcycle is more likely to peak within 18 months than the two-to-three-year window the article assumes."
The article correctly flags AI servers and EVs as demand drivers for Micron's (MU) memory chips, yet it downplays how abruptly the sector flips from shortage to glut. Trailing revenue remains 44% below the prior peak while shares trade near record highs at 17x forward earnings. Hyperscaler AI spending can pause or shift to in-house designs within 12-18 months, repeating the post-2021 inventory correction. Historical cycle peaks have lasted roughly two years before pricing collapses; this one shows no structural change that would extend the upswing.
If generative AI training and inference workloads scale faster than any prior compute wave, memory demand could stay elevated long enough for MU to reset its cycle permanently and expand margins beyond historical norms.
"Micron's valuation assumes the memory cycle extends 2-3 years without supply shock, but the article provides no margin-of-safety analysis for the scenario where competitors' capex comes online and spot prices crater."
The article conflates two separate theses: (1) AI demand is real and durable, and (2) Micron is cheap. On (1), true—but memory is a commodity; pricing power evaporates fast. On (2), the 9.3x sales multiple looks reasonable until you realize Micron's trailing revenue ($18.3B) is depressed by the 2022-23 downturn. Forward earnings at 17x assumes the cycle extends 2-3 years without supply shock. The article admits cycles are unpredictable, then dismisses that risk. Missing: Micron's capex intensity (they must reinvest heavily to stay competitive), SK Hynix and Samsung's capacity additions, and the risk that AI memory demand concentrates in HBM (high-bandwidth memory) where Micron lags.
If NAND/DRAM spot prices roll over in 2025 due to oversupply from competitors' new fabs, Micron's margins compress 30-40% faster than the cycle suggests, and a 17x forward multiple becomes 25x+ on normalized earnings—a value trap, not a bargain.
"Micron's valuation is currently priced for a perfect execution of the AI cycle, leaving zero margin for error regarding broader consumer electronics demand or manufacturing cost inflation."
The article correctly identifies the HBM (High Bandwidth Memory) tailwind, but it ignores the brutal reality of capital intensity. Micron is currently navigating a massive capex cycle to expand production capacity for AI-grade memory. While the forward P/E of 17x appears reasonable, it assumes a linear recovery in margins that may be compressed by rising utility costs and the commoditization of non-AI memory segments. Furthermore, the article glosses over the 'inventory overhang' risk; if smartphone and PC demand doesn't recover in tandem with the AI build-out, Micron's utilization rates will struggle, leading to significant earnings volatility that the current market optimism is failing to price in.
If HBM supply remains structurally constrained by yield issues at competitors like Samsung and SK Hynix, Micron could maintain pricing power far longer than historical cyclical patterns suggest.
"Near-term upside is plausible due to AI-driven memory demand, but a memory-cycle downturn within 2-3 years could erase those gains and justify a more cautious, lower multiple."
MU looks poised to benefit from AI-driven memory demand in the near term, with AI servers needing substantial DRAM/NAND and pricing still firm. The article highlights a cyclical upswing and reasonable valuation (roughly 17x forward earnings, 9x sales), which supports upside. However, memory cycles are notoriously volatile: pricing and margins hinge on industry capex swings, and competitive capacity from Samsung and SK Hynix could pressure ASPs. A potential AI demand slowdown or an acceleration of oversupply could compress margins and discipline the multiple MU trades at, especially if demand fades before the cycle peaks.
Bear in mind that AI demand could plateau or reverse, and memory ASPs may fall as supply catches up; a cycle turn could compress MU's margins and valuation multiples quickly.
"Micron could see sustained DRAM pricing from AI server volumes even if HBM lags competitors."
Claude notes Micron's lag in HBM where pricing power may be limited, but overlooks potential strength in high-volume DDR5 DRAM for AI servers. If competitors focus capex on HBM as Gemini suggests indirectly, standard memory ASPs could stabilize. This links to ChatGPT's oversupply concern: not all capacity additions hit the same segments equally, potentially delaying the cycle peak by a year.
"Competitors won't voluntarily cede DDR5 capacity to Micron; oversupply will hit all memory segments concurrently, not sequentially."
Grok's DDR5 stabilization thesis assumes competitors voluntarily cede standard DRAM to Micron while chasing HBM margins—implausible. Samsung and SK Hynix have no incentive to abandon high-volume DDR5; they'll flood both segments. The segmentation argument delays but doesn't prevent the cycle turn. Gemini's inventory overhang in smartphones/PCs is the real wildcard: if non-AI demand stays weak, utilization collapses regardless of AI server strength, forcing aggressive pricing across all tiers simultaneously.
"Geopolitical export controls on Chinese memory producers create a structural floor for Micron's pricing power that defies traditional cyclicality."
Claude is right about the competitive landscape, but both Claude and Gemini ignore the geopolitical 'China premium.' Micron’s reliance on non-Chinese fabs is a strategic advantage as export controls tighten. If the US restricts Chinese memory producers further, Micron gains structural pricing power regardless of the global supply glut. The market is pricing this as a commodity cycle, but it is becoming a security-linked oligopoly. This shift could decouple Micron from the historical 2-year peak-to-trough volatility.
"China-only protection won't guarantee MU pricing power; global AI capex and non-AI demand risk can still erode margins and reprice MU."
Gemini's 'China premium' thesis rests on export controls alone lifting MU prices; that ignores a global demand shock risk. Even with restricted Chinese supply, MU's growth depends on non-Chinese markets and AI capex cycles, which can slow or reverse. If worldwide AI spend cools or non-AI channels stay weak, the supposed structural pricing power could vanish, compressing margins and the multiple despite a geopolitically favorable backdrop.
Panelists agree that AI demand is a real driver for Micron's memory chips, but disagree on the sustainability of the current pricing and the potential for a supply glut. They also highlight the risk of a sudden shift in AI spending and the potential impact of geopolitical factors on pricing.
Micron's reliance on non-Chinese fabs could provide a structural advantage as export controls tighten.
A sudden shift in AI spending or an acceleration of oversupply could compress margins and discipline the multiple MU trades at.