Nvidia shares fall after report flags slower South Korea HBM expansion

Nvidia’s shares traded lower in the afternoon session after market chatter pointed to a slowdown in South Korea’s high-bandwidth memory buildout, a key input for many AI accelerators. The selloff was reported by Yahoo Finance, citing a move of about 3.7% in Nvidia’s afternoon trading.

The immediate catalyst described in the market report was a claim that SK Hynix is slowing its expansion of high-bandwidth memory, or HBM. HBM is a specialized type of memory designed to deliver higher bandwidth than conventional DRAM, allowing AI processors to move data faster, which is critical for training and inference workloads.

Investors appear to be treating HBM capacity as a potential constraint or timing risk for companies that rely on large amounts of memory to support AI systems. When memory suppliers announcement slower buildout, markets often reprice expectations around the pace at which AI hardware can be shipped at scale, even if the chip designer’s own plans have not changed.

For Nvidia, the linkage is direct. Nvidia’s data center platforms and many of its graphics and compute products depend on high-performance memory configurations. Even without changes to Nvidia’s product road map, any perceived risk to memory availability or supply ramp dynamics can pressure sentiment toward the entire AI hardware chain.

The episode also highlights how Nvidia’s stock can react to developments upstream of its own manufacturing and sales. The HBM supply chain includes memory producers and advanced packaging steps that can become bottlenecks when capacity is tight, demand is front-loaded, or projects are rephased.

Nvidia did not provide a statement in the market report about the specific SK Hynix timing or whether it has already adjusted inventory or sourcing plans in response. The report therefore centers on the market’s interpretation of the memory supplier’s actions rather than on any disclosed change from Nvidia itself.

Looking ahead, investors will likely watch for additional clarification on SK Hynix’s HBM investment pace and for any corresponding commentary from the broader memory and AI hardware supply chain. Nvidia’s next disclosures, including any updates in quarterly reporting about supply, demand visibility, and component availability, could also influence whether the market views the HBM slowdown as a transient timing issue or a more meaningful constraint.