BofA Nearly Doubles Micron Price Target to $950 Amid AI Chip Rally

Bank of America has substantially raised its price target on Micron Technology ($MU), nearly doubling the forecast from $500 to $950, in a sweeping endorsement of the memory chip sector's AI-driven growth trajectory. The aggressive upgrade, delivered by analyst Vivek Arya, extends beyond Micron alone—BofA simultaneously raised price targets on Marvell Technology ($MRVL), Advanced Micro Devices ($AMD), Coherent ($COHR), and Nvidia ($NVDA), signaling broad conviction that artificial intelligence infrastructure demand will drive semiconductor outperformance for years to come.

The timing reflects a fundamental shift in how Wall Street views semiconductor supply dynamics, particularly in memory chips that have historically faced cyclical pricing pressures. Arya's rationale centers on what he identifies as structurally lower memory supply elasticity—a critical insight suggesting that traditional capacity-adding patterns no longer apply due to converging constraints in capital availability, advanced packaging capabilities, power delivery infrastructure, and geopolitical fragmentation.

Key Details: The Supply Crunch That Changes Everything

Arya's upgraded thesis rests on a straightforward but powerful observation: the semiconductor industry cannot simply build its way out of AI-driven demand like it has in previous cycles. Multiple structural bottlenecks are creating a persistent supply gap:

• Capital constraints: Building advanced memory fabrication plants requires tens of billions of dollars in capex, concentrating capacity among a handful of wealthy manufacturers
• Packaging limitations: Advanced packaging technologies—essential for high-bandwidth memory and complex chip stacking—remain capacity-constrained and difficult to scale
• Power infrastructure: Data centers pursuing AI require unprecedented power densities, straining grid connections and forcing geographic concentration
• Geopolitical fragmentation: U.S.-China semiconductor tensions are pushing companies toward redundant supply chains, reducing efficiency and global capacity utilization

These constraints mean Micron, AMD, Marvell, Coherent, and Nvidia face a rare environment where demand growth will likely outpace supply additions for an extended period—the inverse of the typical semiconductor cycle. This structural imbalance justifies both higher absolute valuations and reduced earnings volatility typical of memory stocks.

BofA's analysts project the AI data center systems addressable market will reach $1.7 trillion by 2030, a staggering figure that dwarfs current annual semiconductor industry revenues and suggests the growth opportunity remains in early innings. To put this in perspective, the entire semiconductor industry generated roughly $570 billion in revenue in 2023, meaning the AI data center market alone could represent three times the current total addressable market for chips.

Market Context: Hyperscaler Capital Deployment Fuels the Supercycle

The upgraded price targets arrive as major technology companies—Amazon, Google, Microsoft, and Meta—commit hundreds of billions of dollars to AI infrastructure spending over the next several years. These hyperscalers face existential competitive pressure: whoever deploys the largest, most efficient AI training and inference infrastructure will capture disproportionate value in enterprise and consumer AI applications.

This capital deployment cycle differs fundamentally from previous cloud buildouts. AI workloads demand:

• Advanced GPUs and custom AI accelerators (driving Nvidia, AMD, and design wins for Marvell interconnect chips)
• Specialized memory architectures including high-bandwidth memory (benefiting Micron and competitors Samsung, SK Hynix)
• Optical interconnects for data center communication (supporting Coherent's products)
• Power delivery and thermal management components

The competitive landscape has shifted from a six-year smartphone supercycle to what appears to be a multi-decade AI infrastructure buildout. Unlike consumer electronics demand, which faces saturation and replacement cycles, enterprise AI investment appears constrained primarily by capital availability rather than technical limitations or market saturation.

Regulatory and geopolitical tailwinds reinforce this trend. U.S. export controls on advanced chips to China are accelerating multinational companies' decisions to nearshore and diversify manufacturing, requiring them to order larger buffer inventory levels. Biden administration CHIPS Act subsidies are incentivizing new U.S. capacity, though this remains years from production, leaving near-term supply tight.

Investor Implications: Valuation Reset Amid Secular Tailwinds

For investors, BofA's price target cascade signals something unusual in the semiconductor sector: multiple years of above-trend earnings growth supported by structural supply constraints rather than temporary demand surges.

Micron particularly benefits from this thesis. The memory chip market has historically traded at cyclical valuations—often 8-12x forward earnings during upswings—because oversupply can materialize quickly. A structurally constrained supply environment could justify 20x+ forward multiples over time, particularly if the company maintains pricing power while production grows modestly. At $950, Micron's implied valuation assumes significant multiple expansion beyond historical norms.

Nvidia, already the semiconductor industry's largest company by market capitalization, benefits from being the category-defining leader in AI accelerators. AMD's recent design wins with hyperscalers position it as a credible secondary supplier, capturing pricing power while Nvidia manages allocation constraints. Marvell's high-speed connectivity products are essential infrastructure supporting data center AI expansion. Coherent's optical interconnect technology becomes increasingly valuable as data centers scale to house hundreds of thousands of AI chips in tight physical proximity.

The broader market implication: semiconductor stocks could outperform for years if AI infrastructure spending materializes as expected. Historically, semiconductor supercycles (the PC era, smartphone explosion) drove 5-10x shareholder returns for leaders. Current AI infrastructure dynamics suggest similar magnitude opportunities—conditional on companies executing flawlessly and geopolitical tensions not escalating further.

However, risks remain material. If hyperscaler AI spending disappoints, if competitive dynamics intensify faster than expected, or if new manufacturing capacity unexpectedly comes online, the Bull case could quickly unwind. Investors should view BofA's upgrades as reflecting current consensus sentiment rather than a guarantee of future returns.

Bankof America's sweeping price target increases reflect Wall Street's growing confidence that artificial intelligence is not a cyclical opportunity but a secular transformation requiring decades of capital reinvestment. For memory chip, processor, and optical interconnect manufacturers, this thesis implies a rare combination of strong growth and margin protection—conditions that justify the analyst's aggressive valuation resets.