Micron Technology was my highlight of the week. Going into Wednesday's earnings after the close, I had one simple question: can they deliver again? They did. Thursday's reaction was strong, and the stock hit a new all-time high. And still: not a buy for me

The numbers were outstanding, there's no arguing with that. Revenue nearly quadrupled from $9.3 to $41.5 billion. Net income jumped from $1.9 to $28.2 billion. Adjusted earnings per share climbed from $1.91 to $25.11. And for the final quarter of fiscal year 2025/26 (ending in August), Micron is guiding for further growth and new records across the board.

What impresses me even more than the numbers are the long-term prospects. CEO Sanjay Mehrotra announced 16 completed strategic customer agreements (SCAs) that will, in his words, "fundamentally transform" the business model. These contracts cover roughly 20% of DRAM volume and one third of NAND volume from 2026 through 2030. They are structured as take-or-pay agreements, backed by deposits and commitments of over $22 billion. Pricing is designed to keep average selling prices per bit and gross margins at historically high levels. Eventually, more than 50% of revenue is expected to be covered by SCAs. That, to me, is a genuine paradigm shift. Memory chips have always been a commodity business: cyclical, price-driven, winner-takes-cheapest. The SCAs break that logic, at least in part. Long-term price bands and binding purchase commitments create a kind of stability this market has never had before.

There is also a structural argument on the supply side. The memory market is tight, and it is staying tight. Micron expects demand to outpace supply well into 2027, with no clear timeline for when supply can catch up. Micron also forecasts that the total addressable market for High Bandwidth Memory (HBM) will exceed $100 billion in 2027. HBM is currently the most critical component in building out global AI infrastructure.

And yet I'm not buying, even with a price-to-cash-flow ratio below 10. Despite everything I just described, I'm cautious on the stock in the near term. We've seen several significant corrections this year already: a roughly 35% drawdown in March, and three separate pullbacks of more than 20% in May and June. What concerns me most right now: investment in AI infrastructure has reached a scale that is, not without reason, being compared to the conditions that precede a bubble. If the AI growth narrative starts to crack, things could get uncomfortable fast. Whether the postponement of the OpenAI IPO to 2027 is already a trigger is an open question. For me, it is a first signal of fading risk appetite in the AI sector. I wouldn't be surprised if the market takes a break from the AI trade. Micron would not be spared.

Is Micron Technology a buy? At current prices, I am not buying Micron Technology despite a record quarter and a price-to-cash-flow ratio below 10. The business model is genuinely transforming: 16 completed take-or-pay agreements covering 20% of DRAM and one third of NAND volume through 2030, backed by over $22 billion in commitments, reduce the cyclical commodity risk that has historically defined this sector. The main risk is timing: AI infrastructure investment has reached a scale that is drawing comparisons to pre-bubble conditions, and the postponement of the OpenAI IPO to 2027 looks to me like a first signal of fading risk appetite in the sector.

Disclaimer: This newsletter is for informational purposes only and does not constitute investment advice. I am not a financial advisor. Always do your own research before making any investment decision.

Disclosure: I may hold direct or indirect positions (including options) in any securities mentioned in this newsletter. My opinions are my own and always honest.