Editorial Team
6 min read
Micron has signed 16 multiyear strategic customer agreements that commit buyers to specified memory volumes and, in many cases, pricing ranges. The contracts support its expansion plans and reduce risk for a portion of the business, but they do not yet make most of Micron’s revenue contractual or settle whether demand will outlast the supply shortage.
Micron Technology, a U.S. manufacturer of DRAM, NAND and NOR memory and storage products, is asking customers to take on some of the uncertainty that has long defined the memory market. Its 16 Strategic Customer Agreements, or SCAs, require buyers to purchase specified volumes over several years; the company says the contracts give it greater confidence to invest in capacity after a fifth consecutive quarter of record revenue.
The terms make the agreements more consequential than an ordinary supply forecast. They also set clear limits on the claim that Micron has remade its business: the signed agreements cover only part of its DRAM and NAND output, while the company’s expectation that half or more of revenue will ultimately sit under SCAs depends on agreements still to be completed.
Micron Technology’s Folsom, California office entrance. Photo: Micron Technology. Source: TheStreet.
Micron said in its prepared earnings remarks that most SCAs run from calendar 2026 through the end of 2030; automotive agreements generally run three years. They are take-or-pay arrangements, meaning customers make binding commitments to buy specific volumes.
The company says the 16 agreements represent roughly 20% of its DRAM volume and a third of its NAND volume over that period. Four involve very large customers and three medium-sized customers; the remaining agreements are with smaller automotive customers. Micron did not identify the buyers. A contemporaneous call report said management described many of the major customers as data-center customers, but the unnamed counterparties and the amount each has committed remain undisclosed.
That matters because memory producers normally face both demand swings and price swings. The contracts allocate scarce supply in advance, but they do not show how much negotiating leverage any individual large customer retains, or how concentrated Micron’s commitments are.
Company-reported long-term-contract coverage is 20 percent of DRAM volume and about 30 percent of NAND volume. Source: Seoul Economic Daily.
For SCAs with fixed prices or price bands, Micron is now reporting remaining performance obligations, an accounting measure of contracted future performance. In the prepared remarks, the company said the figure was more than $5 billion at the end of fiscal Q3; for agreements signed so far, including some signed after that quarter, it was approximately $100 billion.
The company says the $100 billion is calculated from minimum committed volumes and minimum prices, and is not its forecast of total contract revenue. The distinction is important: it documents a contractual minimum rather than five years of expected sales, and it does not measure all demand for AI-related memory.
Pricing is not entirely fixed. Micron says its largest agreements generally put a ceiling on existing-product prices at the current calendar-Q2 market price and a floor beneath it. A modest share of SCA-related revenue has either fixed prices or no price bands. Premiums for newer products will be negotiated later. If all planned SCAs are completed, the company expects agreements with fixed prices or ceilings at or near the current calendar-Q2 price to account for about 40% of revenue—a future mix target, not the coverage of the 16 signed agreements.
Micron projects $22 billion in customer deposits and related financial commitments under the agreements it has signed, of which about $18 billion would be cash deposits. The company says the cash will appear in financing-related cash flows rather than free cash flow and will be returned to customers toward the latter half of the agreements’ terms.
That makes the deposits a financing resource for a capital-intensive expansion, not booked revenue. Micron forecasts about $27 billion of fiscal-2026 capital spending, net of anticipated government incentives, after projecting about $10 billion of spending in fiscal Q4 alone. It expects fiscal-2027 quarterly capital expenditure to exceed that Q4 level, with more than half of the year-over-year increase coming from construction spending.
The spending plan gives the contracts practical weight. Micron says its first Idaho fab is on track for first wafers in mid-2027 and a second in late 2028, and that it broke ground on the first of its New York fab cluster in January. It also cited manufacturing and packaging expansion in Taiwan, Singapore, Japan and Virginia. Those schedules and the company’s view that labor, permitting, energy infrastructure, cleanroom capacity and technical transitions constrain supply are forecasts and assessments, not evidence that a shortage will necessarily persist.
Company-reported fiscal-Q3 revenue was $31.3 billion for DRAM and $9.9 billion for NAND. Source: Micron Technology earnings call.
Micron reported fiscal-Q3 revenue of $41.456 billion for the quarter ended May 28, 2026, versus $9.301 billion a year earlier, in its quarterly filing. Its prepared remarks attributed the quarter’s record DRAM and NAND revenue chiefly to higher prices and favorable product mix during tight industry conditions.
Management forecasts fiscal-Q4 revenue of $50 billion, plus or minus $1 billion, and roughly 86% gross margin. Those are company forecasts made in a period when it says industry demand exceeds supply; they are not achieved results. The signed SCAs can reduce downside uncertainty for the contracted share, but their volume coverage also shows how much of Micron’s business remains outside them.
Rivals are pursuing long-term supply arrangements too. A report says Samsung Electronics and SK hynix have used such agreements in earlier memory upcycles, though it describes the newer contracts as longer and more binding. That comparison narrows the novelty claim: the company is presenting greater durability and scale, not the existence of supply contracts, as the change.
The next evidence is not another optimistic demand forecast. It is disclosure showing how much revenue—not just DRAM and NAND volume—has moved under signed contracts, how much of that revenue has price floors, fixed prices or ceilings, and how concentrated the commitments are among customers.
Micron also needs to execute its capacity plans while the agreements run. If new supply, changing product mix or softer AI infrastructure spending arrives before those additions are absorbed, the contracts’ price bands and still-uncontracted output will determine how much protection the new model actually provides.
Keel Infrastructure has completed its U.S. bitcoin-mining shutdown and is developing former power sites for high-performance computing. But its filing says none of the three priority U.S. sites had begun HPC operations or recorded related revenue as of Aug. 7, leaving contracts, funding and execution—not the shutdown itself—as the next test.
Bitdeer’s conditional lease for its Tydal, Norway data-centre conversion is material because the company plans to move 225 megawatts of the former mining site to colocation. But public accounts describe different contract values, terms and counterparties, leaving the customer, funding and revenue implications unresolved.
Nvidia is reportedly discussing an AI-infrastructure funding package of up to $500 billion with large private-capital firms. The talks underline the need to finance chips, data centres and power, but no commitments or risk-sharing terms have been disclosed.
Anthropic, Macquarie Asset Management and GIC have formed Theseus Infrastructure to develop, operate and lease dedicated data centers to Anthropic. The investors will own the platform and provide most project equity, but the companies have not disclosed sites, capacity, costs or a delivery timetable—and Anthropic already has a separate $50 billion infrastructure plan with Fluidstack.
Sen. Bernie Sanders has asked OpenAI, Anthropic and Meta to pause AI development, invoking their conditional safety commitments. The request is not a pause order, but it extends his earlier push to halt new AI data-center construction pending safeguards.
Intel has proposed a $15 billion underwritten common-stock offering, with a further $2.25 billion underwriter option. The gross base deal equals about half of Intel’s June cash and short-term investments, yet the preliminary prospectus leaves the price, share count, net proceeds and allocation of capital unspecified.
Archer has agreed to acquire Boeing subsidiaries Wisk Aero, Insitu and SkyGrid in a stock-and-warrant transaction that keeps Boeing economically and technologically involved. Insitu gives Archer an operating defense business; the deal leaves unresolved how Archer will fund and organize the autonomous Wisk program alongside its piloted Midnight aircraft.
The police watchdog is investigating an unnamed senior Derbyshire detective over alleged AI use in decision logs—records that may be disclosed in court—as ministers press ahead with a £75 million national police-AI programme.
Industries that buy memory chips say tighter supply and higher prices are threatening products from network equipment to medical devices. Their competing requests to Washington—add capacity, reserve supply for domestic industry, or keep Chinese suppliers out—leave Commerce with a policy conflict, not a settled response.
Moore Threads has approved a plan to issue H shares and seek a Hong Kong main-board listing after first-half revenue rose 147%. The proposal has no disclosed size, price, timetable or use of proceeds, leaving investors to weigh a fast-growing cloud-products business against continuing losses, cash use and GPU competition.
A media tally says at least 37 people have been arrested in U.S. data-center protests in 2026, though it is not an official database. Two local cases show a narrower, documented conflict: who can speak at hearings over proposed infrastructure, under what rules, and with what recourse when those rules are enforced.
Meta has released Apache-licensed weights for Muse Glimmer, a 30-billion-parameter agentic model intended for local deployment. The hardware and benchmark claims make the proposition concrete, but they are Meta's measurements, and the harder test is whether developers can safely operate the tools around it.
Mark Zuckerberg says Meta will pursue personal superintelligence for billions of people, with free versions, paid compute and private agents. The plan turns on capacity, release rules and governance that the company has not yet specified.
TSMC reported NT$467.58 billion in July revenue, up 44.7% from a year earlier and described by contemporaneous reports as a monthly record. The total supports its bullish AI-and-HPC outlook, but the disclosure does not break out the customers, products or AI contribution behind the increase.
South Australia plans to begin a royal commission on artificial intelligence on 1 October and report by 1 July 2027. Its immediate test is whether the government can define a focused remit on AI use, work, public services and infrastructure consequences before federal standards move ahead.
Meta has seeded a $1 billion fund for U.S. communities where it owns and operates data centers. The commitment places community benefits at the center of Mark Zuckerberg’s AI expansion case, but the company has not disclosed the timing, allocation rules or reporting that would let a host community judge its bargain.
California’s SB 903 would stop companion chatbots from being presented as psychotherapy and set consent, clinician-review and accountability rules for specified AI uses in mental-health care. The bill is not a blanket ban on AI support; its unresolved challenge is where clinical assistance becomes clinical judgment.
Microsoft’s latest in-house text-to-image model is second in Arena’s August 10 ranking, behind OpenAI’s GPT-Image-2. The result is an early competitive signal from a live head-to-head-vote leaderboard, while Microsoft’s promised reference, grounding and control features—and their wider rollout—remain to be detailed.
ChatGPT search can surface available times from OpenTable, Resy and Yelp and complete some reservations. The rollout makes ChatGPT a new place to start a booking, but coverage, confirmation and every post-booking change remain with the reservation providers.
Yang Qi, Game Science’s co-founder and Black Myth art director, says the team will keep AI-generated content out of Black Myth: Zhong Kui’s design and asset-production stages. The Weibo response is a limited production choice for an early project, not a disclosed company-wide AI ban or a release-date update.