Micron stock (MU) investors have already been rewarded handsomely this year. The harder question now is whether AI has fundamentally changed the business enough to keep those gains from looking like another peak in a familiar memory cycle.
That’s a massive question. If demand remains tight for years rather than quarters, and Micron can lock more business into long-term agreements, the company could start to look less like a volatile commodity producer and more like a business with durable earnings power.
Veteran D.A. Davidson analyst Gil Luria believes the market is still underestimating that shift, as reported by TheFly.
“Micron is on a growth trajectory for the next 3-5 years, which is what the market has not yet acknowledged,” he said.
His thesis rests on a simple shift in AI economics, as he expects a major demand-supply imbalance while raising a bigger question for investors: whether Micron’s earnings should still be valued as temporary.
D.A. Davidson sees a 3-5 year Micron growth story
Luria argues that AI has effectively transformed memory from a highly cyclical commodity into a “mission-critical” component of AI infrastructure.
Consequently, Luria bumped his Micron price target to $3,000 from $2,100 while reiterating a Buy rating, pointing to roughly 187% upside as per the latest closing price.
He believes Micron is on a three- to five-year growth trajectory that investors haven’t fully incorporated into the stock’s valuation and expects memory demand to outstrip supply through both 2027 and 2028.
According to him, “early in their journey of understanding MU’s value,” arguing that this process should ultimately result in a materially higher earnings multiple.
Read more: Morgan Stanley revisits Micron stock price target amid stellar growth
To break things down, Micron’s thesis is basically straightforward, where more capable AI systems need enormous quantities of fast memory.
Luria’s note argues that additional memory can enable larger context windows, faster processing, and better AI performance. That changes the economics of memory because GPUs and custom accelerators aren’t particularly useful if the systems around them cannot feed those processors data quickly enough.
Micron itself has been making essentially the same argument.
Reuters reported that its president and COO, Manish Bhatia, recently described memory as “the chief constraint in AI” as data centers become the world’s largest market for memory and storage.
Moreover, Micron expects memory supply-and-demand conditions to become even tighter in fiscal 2027 and 2028 than in 2026, as reported by TrendForce.
Also, the numbers behind Luria’s thesis are substantial.
Micron’s fiscal Q4 sales jumped to $54.23 billion from $11.32 billion a year earlier, while adjusted EPS reached $33.42. Full-year fiscal 2026 revenue hit $133.19 billion, with adjusted EPS of $75.52.
Management also guided to about $61.5 billion in first-quarter revenue, above Wall Street’s $57.02 billion consensus. More importantly, long-term supply commitments rose to $32 billion from $22 billion, while remaining performance obligations climbed to roughly $150 billion from $100 billion.
That backs up Luria’s view that longer-term contracts could make Micron’s historically cyclical business more predictable. Even “de-specing” may not be bearish if using less memory ultimately hurts AI performance and pushes customers to buy more later.
Micron’s biggest AI opportunity may also be its biggest risk
What stands out to me is how quickly AI is turning memory from a supporting component into a major infrastructure cost.
Bank of America estimates that memory now accounts for roughly 35% of AI infrastructure spending, giving Micron a much larger share of every dollar flowing into AI data centers.
That said, the opportunity comes with an obvious risk: If memory prices climb too far, customers could reduce specifications or slow purchases. Yet BofA has become more bullish, and, as I covered following Micron’s latest results, the firm raised its fiscal 2027 revenue estimate to $275.4 billion from $230.3 billion.
More Micron:
- Michael Burry increases his bet against popular chip giant
- Bank of America doubles down on Micron stock after AI bombshell
- Micron stock jumps as investors look beyond GPUs in AI chip trade
BofA also increased its fiscal 2028 estimate to $317 billion from $244.1 billion. Its EPS estimates jumped to $171.78 and $197.90, respectively.
I think the more revealing comparison is the pricing setup.
D.A. Davidson’s $3,000 target values Micron at roughly 19 times fiscal 2027 earnings, implying about $158 per share. BofA actually forecasts higher earnings at $171.78, yet carries a much lower $1,550 target.
That tells me the debate is increasingly about what Micron’s earnings are worth. With 26 strategic customer agreements and more than 75% of fiscal 2027 output already committed, Micron may be becoming less cyclical. The question is whether that change is durable enough to justify a permanently higher multiple.
Micron looks cheap on earnings, but the market still doubts the cycle
There’s little you can take away from or question about Micron from an AI-demand standpoint. The real question is whether its earnings visibility has improved enough to justify paying something closer to a broader-market multiple for profits investors have historically treated as temporary.
That tension shows up in the valuation data.
Seeking Alpha lists Micron at just 5.94 times forward non-GAAP earnings, compared to a 23.74 times sector median, which is unusually inexpensive.
Still, I would not call the stock cheap based on those multiples alone. Memory companies often look cheapest near cyclical peaks, and Micron’s forward enterprise value to sales multiple of 4.05 times is actually above the sector median of 3.63 times.
Luria clearly thinks this cycle is different.
If AI demand, long-term supply agreements, and tighter memory markets make earnings more durable, then the multiple itself may need to reset higher.
If Micron’s current trajectory really lasts another three to five years, I feel that the $3,000 target is not just a bet on another memory-price spike. It is a bet that Wall Street will eventually stop valuing Micron as it did in the previous cycle.
Related: Nvidia stock flashes massive signal as Wall Street leans in