故郷求めて

Masayoshi Son lost about $3.9 billion on Friday. SoftBank fell 10.8% in Tokyo on November 6. The Nikkei dropped 6% the same day. Since October 1, SoftBank is down 47%, almost half its value in five weeks. In June, Son told CNBC that AI is "probably 50x bigger than dot-com." In the video below he says the 2000 peak was "a small hill." Then the interviewer asks if he's worried about a correction. His answer: there is always a correction. 1929 crashed too, and markets went up for 100 years after. If a correction comes, it would be "the best investment opportunity time to me." Now look at this week: → AI stocks sold off worldwide after the US midterms, on fears about data-center projects → per Bloomberg, a mystery seller is trying to unload about $500M of data-center loans at once, including a piece of the Stargate site in New Mexico that OpenAI, SoftBank and Oracle are building → on Friday Elon Musk replied on X: "SoftBank is defaulting on their turbine orders." That's his claim, 2M views My take: Son has been here before. After 2000 SoftBank lost most of its value, and he came back on Alibaba and Arm. What's different now is how much of SoftBank rides on one bet. When the stock halves in five weeks, buying the dip is easy to say and harder to fund. So which is it: Son's correction and a once-in-a-decade entry, or the first crack in the AI buildout?