BofA raises Micron stock price target to $950 on AI demand outlook

BofA Securities raised Micron’s price target to $950 from $500 while keeping a Buy rating, citing structurally tighter memory supply and AI-driven demand that should support pricing through CY28. The firm values Micron’s sustainable AI/HBM business at about $240 per share and the cyclical DRAM/NAND business at about $710 per share. Micron also began sampling 256GB DDR5 modules for AI servers, reinforcing the bullish AI memory thesis.

Analysis

The market is starting to price Micron less like a cyclical memory name and more like a constrained AI infrastructure asset. The key second-order shift is that capex alone no longer guarantees supply response: packaging capacity, HBM integration, and power delivery are becoming the real bottlenecks, which should keep pricing power stronger for longer than prior memory upcycles. That changes the relative winner set inside semis: component vendors with pricing leverage and high mix exposure to AI memory should continue to take share of incremental industry profit pools from equipment and slower-turning analog/handset-linked names.

The most interesting asymmetry is that the stock may still be underestimating how long earnings can stay elevated if DRAM/NAND never revert to classic boom-bust behavior. If the supply-demand ratio truly stays tight through the next 12-24 months, the market will likely reward MU with a multiple closer to AI compute peers rather than legacy memory peers, but that rerating can be noisy because the base business is still heavily consensus-sensitive. The implied message is that every quarter of stable margins reduces the probability of a sharp de-rating, while any sign of accelerated capacity adds more to the stock’s downside than upside from here.

The broader weakness in adjacent semiconductor names looks less like a sector-wide demand shock and more like a relative rotation toward the few names with direct AI monetization and scarce supply exposure. That creates potential underperformance for companies with weaker end-demand visibility or more mature product cycles, especially if investors start funding MU longs by trimming slower-growth chip exposure. The contrarian takeaway is that the move in MU may be less about a one-time target raise and more about a regime change in memory scarcity; if that is right, the consensus is still too anchored to past cycle duration.

Near term, options activity suggests the market expects another large move, but not necessarily in a straight line. The next catalyst window is the next couple of earnings prints and any commentary on HBM/DDR5 qualification pace; a single quarter of softer pricing would likely be enough to trigger a fast unwind because positioning is now crowded on the bullish side. The risk case is not demand collapse, but supply creeping up faster than expected if packaging constraints ease or if peers redirect capex aggressively into AI memory.

Key Decisions for Investors

• Stay long MU versus a basket of legacy chip names over the next 1-3 months; the cleaner expression is MU long / QCOM-INTC-ON-SWKS short basket, targeting continued multiple divergence if AI memory tightness persists.
• Buy MU on pullbacks rather than strength; use any 3-5% intraday selloff after the next sector-wide risk-off day to add, with a 3-6 month horizon and a tighter stop if management commentary turns to supply normalization.
• Express the view through MU call spreads into the next earnings window; upside is still meaningful if the market continues rerating the stock as an AI infrastructure beneficiary, while premium outlay is constrained if volatility stays elevated.
• Fade the weaker levered peers, especially QCOM and SWKS, on rallies over the next 4-8 weeks; they remain more exposed to handset/consumer cyclicality and less to the AI memory scarcity trade.
• Do not chase the move in equipment names on this headline alone; if MU’s margins stay resilient, vendor margin upside is likely capped relative to memory producers, making a memory-over-equipment relative trade more attractive.