Intel forms 23-company Korean AI alliance as it pitches HBM-free inference GPU

Intel's strategic pivot is not mere market expansion but a structural disruption of the HBM monopoly. By substituting high-bandwidth memory with low-power DRAM for inference, Intel attacks the energy efficiency bottleneck, compelling SK Hynix to shift from a commodity supplier to an ecosystem partner. This fractures NVIDIA's closed 'GPU+HBM' loop, shifting competitive metrics from raw FLOPS to system-level power efficiency. The 23-company Korean alliance serves as a geopolitical hedge, diversifying supply chain risks away from single-vendor dependency. Over the next 18 months, the AI market will bifurcate into high-bandwidth training and low-power inference tiers. If Intel leverages its foundry capabilities to close the yield gap, it will redefine global pricing power; otherwise, it risks marginalization. The core insight: defining the energy-efficiency standard for inference determines who captures the next capital cycle.