I can't talk about too much, but we do quite a bit with softbank[1], and I live in Japan working with HQ. The bets they are making are actually pretty predict

I can't talk about too much, but we do quite a bit with softbank[1], and I live in Japan working with HQ. The bets they are making are actually pretty predictable, they are all more "sure" wins. Many of these companies will not exit for a long time. They are making some (in my opinion) questionable bets on more early stage companies though. If the bets are < 100 million, some of these are more traditional R&D bets that might not give proper returns. Of note here is that "softbank" is actually a massive investment fund though. There are multiple sources of money, not just 1 "vision" fund. Even the way the decision making is done is pretty fragmented. Edit: wrong press release link [1]https://www.forbes.com/sites/aarontilley/2017/09/18/skymind-deep-learning-tool-robots/ https://www.forbes.com/sites/aarontilley/2017/09/18/skymind-... [2]: https://skymind.ai/press/softbank https://skymind.ai/press/softbank