This chip startup just raised $135M on a bet that AI’s biggest bottleneck isn’t compute — it’s memory

South Korean semiconductor startup XCENA has secured $135 million in funding. The company is focusing on addressing the memory bottleneck in artificial intelligence systems, rather than the computational power itself. XCENA believes that improving memory access and capacity is key to unlocking further advancements in AI. Their approach aims to overcome limitations that current AI models face when processing large datasets and complex operations. This investment signals strong confidence in their novel strategy for AI hardware development.

This development matters because overcoming memory limitations could significantly accelerate AI progress and reduce its energy consumption.