SK Hynix, one of the world's biggest memory chip makers, reported its best quarter ever on July 29, 2026. Revenue was 79.3187 trillion won (about $54 billion USD). Operating profit — what's left after paying the costs of making the chips — was 60.5426 trillion won. Net profit for the quarter reached 93.9226 trillion won. The boom was driven by demand for AI memory chips (SK Hynix).

The growth compared to a year ago is enormous. Operating profit jumped 557% from the same period a year earlier, about $41.62 billion USD (Reuters). Revenue more than tripled (CNBC). The operating margin — the percentage of revenue left after production costs — was about 76% (Investing.com).

Yet SK Hynix shares fell about 10% after the announcement. The reason: the record profit still fell short of what analysts had predicted (Reuters).

The company also has a strong balance sheet. SK Hynix had 88 trillion won in net cash — cash minus debt — at the end of June 2026, giving it plenty of room to keep investing in new chip capacity (Reuters).

Here's why that stock drop is worth thinking about. Think of it like a student who scores 95 on a test but whose parents expected 98. The result is excellent, but expectations were even higher. Analysts had assumed SK Hynix would sell its AI memory chips at even higher prices or in even greater quantities than it did. A 76% margin — keeping 76 cents of every dollar of revenue after production costs — is extraordinary for a company that builds multi-billion-dollar factories. For context, margins that high are usually seen at software companies, which don't need expensive manufacturing plants.

A 557% profit increase that still disappoints the market tells you something about expectations. When you compare today's profit to a period when profits were very low, the percentage looks huge but doesn't tell you much. What matters more is whether the company's actual selling prices and shipment volumes matched what it had hinted they would be. The gap between revenue (79.3 trillion won) and operating profit (60.5 trillion won) shows that the company's AI memory chips are generating profit margins that would have seemed impossible two years ago.

There's also a puzzle in the numbers. Net profit (93.9 trillion won) was higher than operating profit (60.5 trillion won). That difference likely comes from non-operating sources — things like gains from investments in other companies or currency exchange effects, though the reported facts don't break that down. With 88 trillion won in net cash, the interest earned on that cash alone could add meaningfully to net profit when interest rates are high.

The 10% share price drop on a record quarter fits a pattern: semiconductor stocks trade on what investors think will happen next, not on what just happened. If the forecast miss reflected even slightly lower chip prices or a cautious outlook for the rest of the year, the market may be betting that profit margins have peaked. That's different from saying the business is in trouble. A cyclical peak followed by a gradual return to normal is very different from a collapse in demand, and the reported facts don't tell us which one this is.

That 88 trillion won in net cash matters beyond just the balance sheet. In a chip market where some competitors can't afford to invest as much, SK Hynix can keep spending on new capacity without borrowing heavily. That could help it gain market share, but it could also mean that if all this new supply outpaces demand, the current boom could end sooner — though that's speculation beyond what the company reported.

For anyone following the AI chip story, the things to watch are chip selling prices, how much the company is shipping versus how much it can produce, and whether big cloud computing customers are stockpiling inventory. The forecast miss, whatever its exact cause, suggests that analysts may have been too optimistic and need to adjust their expectations.