SK Hynix Q2 profit surges 557% on AI chip demand, misses estimates

• Operating profit reached 60.5 trillion won ($41.62 billion), up from 9.2 trillion won year-over-year, but fell short of the 64 trillion won analyst consensus
• Quarterly revenue surged 257% to 79.3 trillion won, fueled by robust demand for advanced memory chips from AI server applications
• The earnings miss was attributed to SK Hynix's higher exposure to high-end AI memory chips, which benefited less from the stronger price rally in conventional memory products compared to competitors

SK Hynix Q2 Profit Surges on AI Chip Demand But Falls Short of Expectations

SK Hynix, the South Korean chipmaker and Nvidia supplier, reported second-quarter operating profit of 60.5 trillion won ($41.62 billion) for the April-June period, representing a 557% increase from 9.2 trillion won year-over-year and reaching a record high. However, results missed analyst estimates of 64 trillion won according to LSEG SmartEstimate.

Key Financial Metrics:

• Operating profit: 60.5 trillion won (up 557% YoY)
• Revenue: 79.3 trillion won (up 257% YoY)
• Period: Q2 2026 (April-June)

Growth Drivers:

The exceptional profit surge was fueled by robust demand for advanced memory chips as major technology companies increased AI data center infrastructure spending. The company attributed results to sustained demand growth and price increases in high-performance AI server products, surpassing its previous quarterly record.

Market Implications:

Despite record profits, SK Hynix underperformed expectations due to its strategic positioning. The company's higher exposure to premium AI data center memory chips paradoxically limited gains, as it benefited less from a stronger price rally in conventional memory chips compared to competitors with more diversified product portfolios.

This result highlights the nuanced dynamics in the semiconductor market, where AI-focused chipmakers face both opportunities and constraints. While AI infrastructure investments continue driving demand for specialized memory solutions, the miss suggests that conventional memory chip pricing may be outpacing the high-end segment in the current market cycle.

The results demonstrate continued strength in AI-related semiconductor demand while raising questions about product mix optimization among memory chip manufacturers.