MoneyThe story, in brief
This chip startup just raised $135M on a bet that AI’s biggest bottleneck isn’t compute — it’s memory
$135M. XCENA just bet that compute isn't AI's real constraint—memory bandwidth is.
The KeyNews take
Why it matters
As AI models scale, memory bottlenecks are becoming a more pressing infrastructure problem than raw compute. This funding validates a shift in how the industry thinks about AI hardware optimization, with direct implications for model deployment costs and latency.
The key facts
5 to knowXCENA raised $135M in funding
$570M valuation
South Korean chip startup
Focus: AI memory bottleneck rather than compute
Published May 29, 2026
Go to the source
TechCrunch AItechcrunch.com
Publisher excerpt: South Korean chip startup XCENA is betting that AI's real bottleneck is not compute, but memory.