Micron Hits $1,000 Again as Wall Street Bets AI Has Broken the Memory Boom-and-Bust Cycle

For years, investors treated memory chips as the ultimate commodity business. Demand would surge, manufacturers would add capacity, prices would collapse, and profits would evaporate. Then the cycle would start again. Artificial intelligence is forcing investors to reconsider that playbook. 

Micron Technology (MU) has moved back above $1,000 a share, extending a rally that has taken the stock from $123 a year ago to four-digit territory. The move is not simply about another quarter of strong memory pricing. A growing body of evidence suggests AI could make the industry's economics more durable than they have been in previous cycles.

The Memory Cycle Is Starting to Look Different

Micron's fiscal third-quarter revenue reached $41.46 billion, up from $9.30 billion a year earlier. Non-GAAP gross margin reached 84.9%, compared with 39% in the year-ago quarter, while operating cash flow jumped to $25.39 billion from $4.61 billion.

DRAM revenue alone increased 343% year-over-year (YoY) to $31.3 billion, with prices rising in the low-60% range. NAND revenue climbed 361% to $9.9 billion, with pricing increasing in the mid-80% range. Micron says industry demand for DRAM and NAND continues to exceed supply and expects tight conditions to persist beyond 2027 because of AI-driven demand and structural supply constraints.

AI is not merely increasing memory demand. It is increasing the amount of memory required per computing system while also making capacity more difficult and expensive to add.

New Street Research recently upgraded Micron to “Buy” with a $1,250 price target, arguing that AI-driven memory demand has made the company's business structurally less cyclical than in previous industry cycles. New Street sees a potential $2 trillion to $3 trillion Micron valuation by 2030.

Bank of America is reaching a similar conclusion from a different angle. Its bull case assumes Micron can generate $236.16 of EPS by fiscal 2030, compared with $136.24 under consensus assumptions. That forecast requires roughly 80% gross margins and 34% annual EPS growth through fiscal 2030.

I wouldn't make $236 EPS my base case. But the fact that multiple analysts are questioning the old memory-cycle model is becoming the bigger story.

The consensus among the 41 analysts tracked by Barchart is that MU stock is a “Strong Buy,” and the average price target comes to $1,460.11.

Micron Is Becoming More Than a Memory Supplier

The memory chipmaker has also started acting less like a traditional chip manufacturer and more like an AI infrastructure company. Last week, it launched the $250 million Micron Ventures Paradigm Fund, its third and largest venture fund, to invest across the AI technology stack, including model architectures, compute infrastructure, enterprise applications, and physical AI. The new fund brings Micron Ventures' total committed capital to $550 million.

That puts Micron in the same broad strategic camp as Nvidia (NVDA) and Alphabet (GOOG) (GOOGL), which have used investments and partnerships to expand their exposure to the AI ecosystem beyond their core businesses.

Micron's move is also practical. If the next generation of AI requires new memory architectures, accelerators, networking, or software, Micron has a financial incentive to understand those technologies before they reach the mainstream.

The company is already signing multiyear strategic customer agreements that it says should improve the durability and predictability of its financial results.

Key Takeaway

In short, MU stock breaking above $1,000 looks less like a victory lap for another memory cycle and more like a test of whether the old memory-cycle playbook still applies.

Granted, Micron's current 85% gross margin is unlikely to remain untouched forever. Competition will keep pricing under pressure, and eventually new capacity will arrive. But investors don't need 80% margins forever to make the bullish thesis work. They need AI demand to remain strong enough that new capacity is absorbed faster than the industry can build it.

So far, the evidence points that way. Micron's record results, multiyear customer agreements, expanding AI memory markets, New Street's $1,250 target, and even Micron's decision to invest $250 million in AI startups all point toward the same conclusion: This memory boom may have more legs than the market once believed.