Can the AI boom sustain SK Hynix's record earnings?

Net profit reached 39.8 trillion won in the July-September quarter. Operating profit also exceeded the analyst estimate of 76.7 trillion won tracked by LSEG, highlighting the strength of the company's earnings momentum. Expanding AI data centres are driving demand for high-bandwidth memory (HBM) and advanced server DRAM. SK Hynix has benefited from stronger demand for server DDR5 memory and the ramp-up of its next-generation HBM products.

Higher prices for key memory products have also supported revenue growth and profitability. SK Hynix expects HBM4 to overtake HBM3E in both shipment volume and revenue in the fourth quarter, reflecting the growing importance of next-generation memory for AI applications. The company also plans to ramp up its M15X facility to maximum capacity by the end of the year. The expansion is aimed at supporting rising demand for advanced memory products.

SK Hynix expects memory demand to outpace supply as investment in AI infrastructure continues to expand. The company has signed long-term agreements with 14 major customers covering more than 50% of its total production capacity for 2027 and 2028. It also plans to increase investment to expand production capacity next year. However, the outlook for a prolonged shortage remains dependent on future demand, supply additions and the pace of AI investment.

SK Hynix plans shareholder returns of approximately 100 trillion won in 2026, including at least 30 trillion won in cash dividends and more than 60 trillion won in share repurchases and cancellations. The company also approved a third-quarter cash dividend of 20,000 won per share, including a special dividend. The planned payouts reflect its strong cash-generation outlook and efforts to return more capital to shareholders.

SK Hynix's record earnings highlight its strong position in the AI memory supply chain, supported by demand for advanced HBM products, long-term customer agreements and capacity expansion. Investors will now watch whether AI infrastructure spending and memory prices can sustain the current pace of earnings growth. While supply constraints could support pricing, any slowdown in AI investment or faster-than-expected capacity additions could challenge the industry's growth outlook.