Major hyperscalers negotiating long-term supply agreements with DRAM suppliers to stabilize memory pricing amid aggressive AI infrastructure capex.

SK Hynix has announced it has struck approximately ten long-term supply deals with key customers, many of them AI players, aimed at stabilizing volatile memory prices. The company revealed the arrangements alongside Q2 results showing quarterly revenue of ₩79.3 trillion ($54.5 billion), a 257 percent year-over-year increase, and operating profit of ₩60.5 trillion ($41.6 billion), a 557 percent year-over-year increase. The company even achieved the unusual distinction of delivering more net income than revenue, benefiting from asset sales.

This growth was driven by an average 30 percent increase in DRAM prices and a 50 percent rise in NAND memory prices, alongside higher shipment volumes. For the third quarter, SK Hynix predicts slower growth, with NAND shipments rising by a low single-digit percentage and approximately 10 percent more DRAM shipping from its factories. The company anticipates higher average selling prices overall thanks to increased HBM4 volumes.

Executives pointed to the company's substantial capital expenditure on manufacturing capacity but dismissed concerns about oversupply, citing the AI industry's voracious appetite for memory extending well into the foreseeable future. Company president Song Hyeon-jong stated that SK Hynix's relationship with customers "is evolving beyond transactional relations into more strategic long-term partnerships," describing this as "evidence of the sustained demand coming from the AI ecosystem."

The company's major customers are securing deals lasting up to five years, structured to maintain price stability. "The objective is to reduce uncertainty arising from short-term market volatility while enhancing long-term business stability for both our customers and SK Hynix," Song said. "At the same time, it is equally important to secure effective purchase commitment, given the impact that demand volatility can have on the memory cycle. In addition to long-term volume commitments, the agreements include mechanisms such as deposits that can strengthen contract implementation and demand visibility."

Head of investor relations Park Seong-hwan argued that demand for memory and storage will remain robust for years to come, even if major AI companies accumulate excess datacenter capacity. He noted that hyperscalers leasing datacenters signals increased utilization rates for existing infrastructure, and dismissed more efficient AI models as a demand threat, contending instead that better models drive increased service adoption and therefore infrastructure requirements. "This view is also supported by the medium-to long-term demand outlook we have discussed with our key customers," Park said. SK Hynix also expects agentic AI to drive additional memory demand across servers, characterizing this as "a structural shift in demand where both AI memory and conventional memory are growing together."

Despite the strong financial results, investors appeared unconvinced, with SK Hynix's share price declining approximately 5 percent at the time of writing.