SK Hynix posted record third-quarter revenue of 104.62 trillion won and operating profit of 83.33 trillion won, surging 632.1% year on year, as demand for AI high-bandwidth memory continues to drive the South Korean chipmaker's results.
South Korean chipmaker SK Hynix on Wednesday reported its highest-ever quarterly profit, posting operating earnings that sharply exceeded analyst expectations and crossing the 100 trillion won revenue threshold for the first time, underscoring the sustained strength of AI-driven memory demand.
The company's third-quarter operating profit reached 83.33 trillion won, compared with the 76.72 trillion won forecast by LSEG SmartEstimates. Revenue came in at 104.62 trillion won ($78.86 billion), against expectations of 98.34 trillion won.
Operating profit climbed 632.1% from a year earlier, while revenue rose 327.9% over the same period. Sequentially, revenue grew 31.9% and operating profit jumped 37.6% from the prior quarter.
Alongside the earnings report, SK Hynix announced plans to return approximately 100 trillion won to shareholders — including at least 30 trillion won in cash dividends and 60 trillion won in share buybacks and cancellations.
The company approved a third-quarter dividend of 14 trillion won, which includes a special dividend funded by gains from its investment in Kioxia. An additional 20 trillion won in share repurchases was announced, with half of that amount set for cancellation this year.
The results reflect accelerating demand for high-bandwidth memory, or HBM — a specialized type of advanced memory used alongside AI processors to manage large volumes of data at high speeds. SK Hynix expects HBM4, the sixth generation of the technology, to surpass its predecessor HBM3E in both shipments and revenue during the fourth quarter, as it scales production at its M15X facility in Cheongju, South Korea.
The chipmaker said it has secured long-term supply agreements with 14 major customers covering more than 50% of its production capacity for 2027 and 2028. The company also projected that memory demand will outpace supply through 2030 and said it plans to significantly increase capital investment next year.
In August, SK Hynix had announced plans to repurchase and cancel 40 trillion won worth of its own shares, while raising its shareholder return target to more than 50% of cumulative free cash flow generated between 2025 and 2027.
One notable development alongside the earnings release: SK Hynix disclosed last week that its U.S. subsidiary Solidigm would halt its review of potential pre-IPO fundraising and IPO options, citing market volatility and share price trends. The company said it would continue evaluating investment timing, scale, and financing options to support Solidigm's growth.
With its M15X facility ramping and customer commitments extending well into the decade, SK Hynix enters the fourth quarter as the dominant supplier of HBM to the AI industry — a position that analysts will be watching closely as rivals work to narrow the gap.
Marc Sabatini is a staff writer at TechEchelon covering enterprise software, cybersecurity, and the regulatory beats that shape both. He focuses on the deal flow and policy decisions that move markets.
More from Marc →Anthropic's S-1 filing revealed $131.7 billion in cash and short-term investments, surpassing every non-bank U.S. company except Berkshire Hathaway and Alphabet, even as the company disclosed more than half a trillion dollars in future obligations and analysts cautioned that the AI sector's financing risks remain unresolved.
Anthropic filed for an IPO Monday that could be the largest in history, disclosing 14-fold revenue growth and its first profitable quarter — but analysts warn an underperformance could trigger a broad reset across AI valuations in both public and private markets.
Anthropic publicly filed for a Nasdaq IPO Monday under the ticker "ANT," disclosing $16.29 billion in first-half revenue — up 1,304% year over year — alongside an $80.7 billion net loss and ongoing legal battles with the U.S. Department of Defense.
Written for readers who already know the basics — markets, AI, and the policy decisions that shape both.