Intel (INTC.US) has appointed a new chief architect to spearhead its graphics processing unit (GPU) operations, CEO Patrick Gelsinger announced on Tuesday. He also indicated that the ongoing memory chip shortage affecting the technology sector is expected to last for at least another two years. GPUs, a type of chip pioneered by companies like Nvidia and AMD, serve as the computational backbone for large language models. Demand for graphics processors has surged as major corporations race to build out AI infrastructure and data centers. Speaking at the Cisco AI Summit, Gelsinger remarked to the audience that convincing the new executive to join Intel required considerable effort, though he did not disclose the individual's name.
Over recent years, the struggling American chipmaker has lagged behind major semiconductor competitors like TSMC (TSM.US), which have capitalized on the AI data center construction boom to achieve growth. While Intel's stock has rebounded over the past year, buoyed by significant investments from the U.S. government, SoftBank, and Nvidia, and growing investor optimism regarding its foundry business, the company still primarily manufactures chips for its own use. Earlier this month, production bottlenecks and supply chain issues overshadowed Intel's better-than-expected quarterly results. Investors are also seeking clearer guidance regarding potential anchor customers for its foundry operations.
Furthermore, Gelsinger addressed the widespread memory chip shortage impacting the global technology industry. He stated, "From what I understand, there are absolutely no signs of the situation easing." Gelsinger mentioned he had discussions with two key players in the memory sector who informed him that relief is not expected "before 2028." This assessment aligns with views from several major Wall Street institutions.
A recent UBS analyst report noted that against the backdrop of continued AI data center expansion, the global memory industry is experiencing significant structural divergence, with any meaningful supply relief unlikely to materialize before around 2028. Until then, the structural demand driven by AI data center construction will continue to bolster the memory chip supply chain's prosperity and pricing power. Analysts at Nomura project that the "memory supercycle," beginning in the second half of 2025, will persist at least until 2027, with substantial new supply capacity only emerging in early 2028.
The massive expansion of AI infrastructure is driving up demand for memory chips, thereby constricting the supply available for traditional computers and smartphones. This dynamic is triggering shortages, price increases, and potentially dampening consumer purchasing appetite for these products. Intel is the largest manufacturer of personal computer processors, which rely on memory chips for data storage and management. The world's leading memory chip producers include Samsung Electronics, SK Hynix, and Micron Technology (MU.US).
Gelsinger noted that Nvidia (NVDA.US), a leading supplier of AI processors, will further amplify memory chip demand with its recently launched Rubin platform and next-generation products. He stated plainly, "AI is going to 'consume' vast amounts of memory resources."
Comments