Claude AI also said Burry’s worry “deserves respect.” The AI agent pointed to the risk that hyperscalers may be overbuilding and that some AI computing could become a cheap commodity. Claude holds no semiconductor stocks itself after selling Broadcom, keeping its AI exposure in software names like ServiceNow and Zeta instead. <\/p>\n
Burry’s Short Is Already Working <\/p>\n
Micron has fallen from an intraday high above $1,200 to around $865 since Burry’s post. That doesn’t prove his thesis right, but it means the trade has moved in his favor since he made it. <\/p>\n
What Wall Street’s Bulls Are Actually Betting On <\/p>\n
The bull case isn’t “the AI trade continues” in the abstract \u2014 it’s a specific bet that Micron has changed its business model. Sixteen strategic customer agreements running through 2030 lock in roughly $100 billion in take-or-pay revenue, backed by $18\u201322 billion in customer deposits Micron collected upfront, effectively an interest-free loan from its own customers. One bull frames this as Micron becoming an “infrastructure toll-bridge” rather than a cyclical chipmaker, arguing the stock should re-rate from its current forward multiple (roughly 11.6x per Seeking Alpha data, though some models put trailing multiples near 5.6-7.6x on out-year estimates) toward an infrastructure-like 16\u201319.5x. A separate DCF model puts fair value at $1,527.88, more than 50% above today’s price, leaning on Micron’s late-June supply and design partnership with Anthropic and a $250 billion US manufacturing buildout through 2035. <\/p>\n
The Catalyst That Actually Tests Both Sides <\/p>\n
Bulls point to Micron’s incredible strengths. The company has signed 16 strategic customer agreements running through 2030, locking in about 20% of DRAM and 33% of NAND capacity, backed by roughly $18 billion in customer deposits collected upfront.This converts Micron from a cyclical hardware seller into something closer to an infrastructure operator with recurring, SaaS-like revenue, justifying a re-rate from a forward multiple in the high single digits toward the high teens. This camp also points to Micron’s lead position in SOCAMM2 memory architecture for agentic AI workloads, where shifting CPU-to-GPU ratios mean standard DDR5 can’t keep up on density.A second, narrower argument says Micron doesn’t need to beat SK hynix at supplying Nvidia \u2014 it just needs to hold its roughly 21\u201322% HBM revenue share as hyperscalers like Google, Microsoft, and Broadcom build custom AI chips that need HBM independent of Nvidia’s GPU cycle. Goldman Sachs estimates custom-ASIC HBM demand will grow 82% in 2026, a third of the total HBM market. This case is more falsifiable than most: it depends explicitly on Micron landing at least one named non-Nvidia customer, which hasn’t happened publicly yet. Without it, even bulls making this argument put fair value barely above today’s price. <\/p>\n
The Hidden Risks <\/p>\n
Micron’s HBM growth case depends partly on winning a named customer beyond Nvidia \u2014 Amazon, Microsoft, or Meta’s custom AI chip programs. As of now, no such production win has been publicly confirmed, and Google has already picked SK hynix as its first HBM3E supplier for its latest TPU. Micron’s ~21% HBM revenue share is a real number today; the higher share bulls are underwriting is still an assumption. <\/p>\n
ClearBridge Large Cap Growth Strategy stated the following regarding Micron Technology, Inc. (NASDAQ:MU) in its Q2 2026 investor update<\/a>: <\/p>\n “The Russell rebalance substantially reduced the weighting of the Magnificent Seven in the Russell 1000 Growth Index (RLG), with Apple, Microsoft and Amazon shifting toward the Russell 1000 Value Index while Alphabet’s weight increased (Exhibit 1). Factoring in the net effect of these changes, the index saw a reduction in top-heavy concentration, which we view as a welcome change. At the same time, the rebalance has caused the index to become more momentum oriented, with Micron Technology, Inc. (NASDAQ:MU) entering the benchmark at an almost 400 basis point weighting the most immediate example. <\/p>\n With semiconductors becoming more than 30% of the RLG, we also initiated a position in Micron, a leading memory provider that we believe is well-positioned to benefit from growing AI infrastructure investment as AI data centers require significantly more memory than traditional computing systems. We see Micron as a differentiated way to gain exposure to the AI buildout while diversifying our semiconductor holdings.” <\/p>\n photo by Business-laptop-campaign-creators on Unsplash <\/p>\n While we acknowledge the risk and potential of MU as an investment, our conviction lies in the belief that some AI \u00a0stocks hold greater promise for delivering higher returns and doing so within a shorter time frame. If you are looking for an AI stock that is more promising than MU and that has 10,000% upside potential, check out our report about the cheapest AI stock<\/a>. <\/p>\n READ NEXT:\u00a033 Stocks That Should Double in 3 Years<\/a>\u00a0and\u00a0Cathie Wood 2026 Portfolio: 10 Best Stocks to Buy<\/a>.\u00a0 <\/p>\n<\/a> Hesai (HSAI) Shareholders Approve 8-for-1 Stock Split <\/p>\n