Micron Technology shares fell sharply after investors began to reassess whether the economics of the artificial intelligence buildout can hold, even as the company continues to benefit from a persistent memory shortage tied to data-center demand. Micron stock has dropped by 32% from its June record high, reflecting growing concern that rising chip and infrastructure costs could eventually curb enterprise spending on AI software and services.
Key takeaways
- Price move: Micron shares have retreated 32% from their June record high.
- Catalyst: Investors are focused on whether soaring hardware costs will pressure AI customers to slow usage or shift to cheaper models.
- Financial backdrop: Micron’s recent results remain strong, with AI-linked demand continuing to drive revenue and earnings growth.
- Implication: The stock may look inexpensive on traditional valuation metrics, but demand-sustainability risk is now a central question for the market.
What drove Micron lower
According to Bloomberg, about 118 gigawatts of data-center capacity could be installed across the U.S. by 2030 to support the AI boom. Nvidia CEO Jensen Huang has also said that building one gigawatt of capacity requires $50 billion in capital investment. If Bloomberg’s projection holds, Huang’s framework implies an aggregate buildout cost of $5.9 trillion by 2030 for major technology firms.
The central issue for investors is that large-scale spending must translate into returns. Some companies can recoup costs by leasing computing capacity or charging for the use of AI models and software applications. But as hardware costs rise, the economics of AI delivery become more challenging, potentially forcing AI vendors and cloud providers to reprice offerings or modify usage policies.
In recent reporting, Microsoft and Anthropic implemented price increases for certain AI software products, with early signals that the higher cost structure is affecting customer behavior. Uber Technologies, for example, burned through its 2026 AI budget in four months using Anthropic’s Claude Code, according to the article. The company’s chief operating officer also indicated it was becoming hard to justify the spending pace.
Other large users have reportedly taken steps to manage AI consumption. Amazon and Walmart have also reportedly capped AI usage for employees to limit budget blowouts. Separately, a survey cited from UBS Group found that 60% of businesses are routing tasks to cheaper and more efficient AI models in an effort to reduce costs.
Any reduction in AI software spending could translate into slower growth in computing demand, which ultimately feeds through to semiconductor and memory suppliers. The stock decline appears to reflect Wall Street pricing in that potential second-order risk.
Micron’s fundamentals have not cracked yet
Despite the selloff, Micron’s latest financial performance has remained robust. The company reported record total revenue of $41.4 billion in fiscal 2026 third-quarter results (ended May 28), representing a 346% increase from the year-ago period. The company attributed the growth to AI-related memory demand across data centers, computers, smartphones, and even vehicles.
Micron also delivered earnings of $24.67 per share for the quarter, up 1,368% versus the prior year period. The backdrop for the strength is the ongoing global memory shortage, which has helped support pricing power and improved margins.
Guidance for the current quarter, which ends at the end of August, points to further momentum, with the company indicating revenue of $50 billion and earnings of $30.73 per share. However, the market’s concern is that demand growth could cool before those targets are fully reflected in final results.
Valuation looks supportive, but uncertainty remains
On conventional valuation measures, the stock may not reflect the bearish shift in sentiment. Based on trailing 12-month earnings of $44.23 per share and a recent closing price cited in the article of $823 on July 31, Micron traded at a price-to-earnings ratio of 18.6. That compares with a Nasdaq-100 price-to-earnings ratio of 32.6 in the same reference, suggesting the memory supplier has been priced at a discount to large-cap technology peers.
The article also notes Wall Street expectations for fiscal 2027 earnings of $153.74 per share, which would imply a forward price-to-earnings ratio of 5.3. Such figures typically indicate that the market is pricing in less growth than the company’s recent momentum suggests.
Still, investors appear to be weighing a risk that the same forces boosting demand—AI infrastructure buildouts—could eventually give way to tighter spending constraints. The article highlights two potential pressure points: a possible slowdown in AI hardware demand if customers become more cost-sensitive, and supply-side normalization if manufacturers expand capacity and pricing power weakens.
What to watch next
Investors will likely focus on whether Micron’s guidance and near-term bookings can withstand any pullback in AI-related usage patterns. Ahead of the next set of quarterly updates, markets may also turn to signals about broader AI infrastructure spending—particularly from large cloud providers and software vendors—as well as any incoming data that could influence expectations for technology-capex and demand elasticity.
