Micron's FY2025 revenue hit $37.4B (+49% YoY) with $7.59 diluted EPS, and management is pointing toward a $50B revenue target on the horizon — implying a ~34% additional revenue step-up from here. The headline creates a classic setup tension: whether HBM-driven AI demand can sustain the trajectory or whether DRAM/NAND cyclicality snaps it back before $50B is reached.
Micron's FY2025 revenue hit $37.4B (+49% YoY) with $7.59 diluted EPS, and management is pointing toward a $50B revenue target on the horizon — implying a ~34% additional revenue step-up from here.
MU has delivered a 49% revenue surge to $37.4B and is guiding toward $50B, but the question is whether the HBM-driven AI upcycle has enough runway to hit that target before memory cyclicality reasserts itself.
Memory cycles mean-revert hard and fast — any deceleration in cloud/hyperscaler AI capex, a competitor HBM supply surge from SK Hynix or Samsung, or further U.S. export restrictions on China sales could compress margins well before $50B revenue is in sight, collapsing both the earnings estimate and the multiple simultaneously.
CoverageSource: Yahoo Finance · Published here TUE, JUL 7 · 6:20 AM ET · the only report in this recordHow this is decided →
Micron Technology reported FY2025 revenue of $37.4B, up nearly 49% year-over-year, with gross margins of 39.8% and net margins of 22.8%, generating $7.59 in diluted EPS. The company has apparently issued forward commentary pointing to a $50B revenue milestone, representing roughly a 34% further step-up from the already elevated FY2025 base. The numbers represent a dramatic recovery from the cyclical trough of FY2023, when Micron's revenues fell below $16B.
The memory market's current upcycle is being powered largely by High Bandwidth Memory (HBM) demand tied to AI accelerator buildouts — Micron competes directly with SK Hynix and Samsung for HBM3E slots in NVIDIA's GPU supply chain. Micron is considered a meaningful beneficiary as the third major HBM supplier ramping capacity, and the revenue trajectory reflects that positioning.
The bull tension is straightforward: if AI infrastructure spending holds and Micron's HBM ramp continues on schedule, the $50B target is a credible multi-year path, and the stock trades at a historically reasonable P/E multiple during upcycles. The bear tension is equally grounded in memory history: DRAM and NAND are notoriously cyclical, capital intensity is high, and any demand air pocket — from cloud capex deceleration, China export restrictions tightening further, or competitor HBM capacity flooding the market — could compress margins sharply before $50B is ever printed.
What to watch: HBM allocation updates in upcoming quarterly calls, any shift in NVIDIA or hyperscaler AI capex guidance, and whether conventional DRAM/NAND pricing holds or softens as supply catches up with demand. The $50B headline is a target, not a guarantee — the timing and the path through the next cyclical turn are the real question.
FY2025 revenue of $37.4B (+49% YoY) and $7.59 EPS confirm Micron is deep in a genuine upcycle, not just recovering from a trough; HBM3E demand from AI accelerator supply chains provides a structural demand floor that prior DRAM cycles lacked. The $50B revenue target, if even partially credible, implies significant earnings upside from the current base and could re-rate the multiple. Historically, memory stocks have tended to overshoot to the upside during AI-infrastructure-led demand cycles before the reversal.
The read above, as written. kept as written · closes shown from JUL 7 on
3-6 months, into next quarterly print. Follow to be told when one lands.
Price context does not establish that the story caused the move.
FY2025's 39.8% gross margins and 49% revenue growth, combined with a credible $50B forward target driven by HBM3E AI demand, suggest Micron is in the strongest structural demand environment in its history — not just a vanilla DRAM recovery.
Memory semiconductors have an unbroken track record of violent cyclical reversals, and at $37.4B in revenue Micron is already near historical peak utilization — any demand softening or HBM supply normalization could reprice both earnings and the multiple sharply lower well before the $50B target is reached.
Kept as written · your side, if you take one, is graded privately against licensed closes after 10 trading days · nothing here is advice · How the Wire is made →