An October announcement that Nvidia would supply 260,000 GPUs to South Korea was seen as a diplomatic gain, but most supply is concentrated among large companies. About 20 percent is available for academia, research institutes and startups, with the government leasing capacity from cloud firms. The author argues AI demand is shifting toward low-latency inference, where domestic NPUs could be deployed, but cheap large-scale GPU supply could crowd them out. He calls for regional distributed inference hubs using idle public assets and local demand aggregation.

Generative AI startup Anthropic said it will void all over-the-counter trades of its shares conducted without explicit board approval and will not record them in company books. The policy also covers beneficial-interest deals, forwards, SPVs and tokenised securities structures. Some trades on Forge Global and Hiive and a list of unapproved intermediaries were cited. The move comes as implied off-market valuations have risen to around $1 trillion, lifting investor anxiety.