AI rewrites the venture capital playbook

Good morning from the Financial Times. Today is Thursday, September 10th, and this is your FT News Briefing. The US Treasury Secretary is disappointing bond traders, but it looks like he helped give the yen a boost. Plus, tech investors are shooting for the stars. And so, the risk is multiplied, and the capital involved, the capital at risk is multiplied. Let's go big, okay. I'm Sonya Hudson in for Mark Filipino, and here's the news you need to start your day. US government bond yields rose to their highest level in almost three years yesterday, yields move inversely to prices. The sell-off came after Treasury Secretary Scott Besent said he's expanding his plan to buy back long-term government debt to $6 billion. And that disappointed investors. Now, it's a step up from the Treasury's initial $4 billion figure last month, but it's less than the roughly $8 billion that Wall Street expected. Higher bond yields push up borrowing costs and mortgage rates for US consumers, as well as for companies that want to raise cash. Besent said he hopes the expanded buybacks will stop the deepening bond sell-off, but he said that the route does not "reflect the underlying market fundamentals." Scott Besent has also been intervening in the currency market. The yen has been doing well this week, partially due to a recent unusual move by the US Treasury. Besent worked with the Japanese government to buy up the currency earlier this summer because of its importance to US financial markets. The FT's Tokyo Bureau Chief Leo Lewis says initially there were doubts. There was this big intervention that you just mentioned earlier in the summer, but over the weeks that followed, the yen started falling back, and there was a sense that, you know, are they going to step back into the market? And then rather to everyone's surprise, the yen started moving seemingly on its own. One reason the yen may have strengthened is because traders are expecting the country's central bank to raise interest rates next week. The Bank of Japan has been in the process of normalizing, which is to say that the Bank of Japan had for a very long time been this big outlier. It had kept rates very, very low, at zero or below for a long period of time. Normalization in its terms is bringing interest rates into positive territory. What's a state now is that it is under pressure to significantly accelerate that process of normalization. Markets are closely watching this Bank of Japan decision next week, and so is Besent. Here he is talking about it earlier this week. So when we intervene with the Japanese yen, I have pretty good insight into what the Japanese, what the Bank of Japan is going to do, what Japanese policy makers are going to do. The Bank of Japan is under pretty overt pressure from the US side to get ahead of the curve and to accelerate or become more aggressive in its schedule of interest rate increases. There's been some speculation it might do back-to-back interest rate increases and do another one in October, which again would be very, very rare. There's a lot going on in the minds of people who are worried that they could very easily or very quickly get on the wrong side of a big yen move. Things are looking good for the yen right now. Does that mean that Besent's intervention worked? Because there was such a substantial show of willpower by the Japanese and they had the kind of permission more than in position and endorsement from the US side. The market is left with the sense that they could do it again at any time, and if they have managed through the intervention to slightly reset the way that markets look at the yen and the risks around the dollar yen exchange rate, then that will have been an effective intervention. Leo Lewis is the FT's Tokyo bureau chief. A leading candidate for the IMF's chief economist job lost out on the post after he was found to have criticized the Trump administration's tariff policy, that's according to people briefed on the process. London's School of Economics Professor Ricardo Rice had been slated to lead the fund's research and analysis arm, but the IMF nixed his appointment at the 11th hour. IMF insiders told the FT that the incident made many people worry the role would go to a Trump ally, and that that would undermine the fund's credibility. The US is the IMF's largest shareholder, and can exert a strong informal influence over the organization. The job ended up going to Silvana Tanreiro. She's a former policymaker at the Bank of England and another highly regarded economist at the London School of Economics. An IMF spokesperson declined to comment on the hiring process for its chief economist role. They told the FT that the selection process for the role had been "rigorous and competitive." Rice also declined to comment. Moonshot capitalism is back. The AI boom is fueling a resurgence in ambitious bets, some of which are literally out of this world. Early SpaceX backers have earned huge returns and valuations for traditional software companies are falling. That's forcing tech investors to embrace riskier and more capital intensive deal making. Here to talk more about this is Tim Bradshaw. He's the FT's global tech correspondent. How has the venture capital investment landscape changed recently? Tell us a bit about that. In the years following the 2000.com busts, a lot of venture capital investors went looking for very predictable, safe tech investments. They found them in software as a service company. That seems to be running out of steam now that we are into the AI boom. Many investors are worried that AI model companies like Anthropic and OpenAI will end up eating all of those traditional enterprise software businesses lunch. Hardware is very much back in fashion. A lot of the new more ambitious ideas that are getting funded are in the physical economy, whether that's in energy or space. We've seen more than $150 billion invested in these so-called deep tech deals, excluding AI since the start of 2024, which is more than the $130 billion in the entire decade to the end of 2019. Is there a particular type of company or a particular sector that people have told you will really take off in the future? I think a lot of them are just in some way inputs or outputs of the AI. So anything that can radically lower the cost of powering AI chips or put them in novel places, out to sea, up in space, inside nuclear power stations are all getting funded. There is renewed interest from several startups in brain, computer, interface technology. One of them is literally called the merge, which is you merge a brain with an AI. Then there's also a lot of other ideas that are really enabled by AI technology, nuclear fusion itself. There are startups that are working on that now that say it's easier to model these fiendishly complicated systems because you can do it in software using AI before you have to physically make the thing and test it in the real world. So AI is both the source of demand for novel ideas for lowering energy costs and also an enabler to take on tougher ideas that previously would have just been much more expensive in upfront capital costs. There's obviously a lot of excitement brewing around these types of companies, but moonshot investments are by definition riskier. Can you tell us about some of the potential pitfalls here for investors? So venture capitalists in the tech industry have always been used to making fairly unlikely bets. But I think the idea with this kind of moonshot trend is that the biggest hits are going to get bigger. And so you've got to be more ambitious in what you're backing. They require a lot more capital to get off the ground because they're working in the physical world. And they're also harder technically to pull off. And so first you have to solve the science problem, then assuming you do solve the science problem, you have to sell it. And the commercialization risk of this technology is as hard as it would be for any other venture-backed company. Just because you have a great scientific or engineering founder doesn't mean that they'll be great at commercialization. And so the risk is multiplied. And the capital involved, the capital at risk is multiplied. It's go big or go home. So it's taking high risk, high reward to a whole other level. What are some of the risks with this type of investment to the wider market and the wider economy? I don't think there's a huge risk of fallout for regular people on the street from this. We're talking about tens of billions of dollars a year going into this deep tech outside of AI. And so I think the risk really is that if the AI boom slows down, then all of these other bets will be off. That will stop the overflow of capital into these other kind of slightly crazier and more ambitious ideas. Well, as always, we will keep our eye on the AI industry. Tim Bradshaw is the FT's global tech correspondent. Thanks, Tim. Thank you. Before we go, one of London's grandest hotels is sending a message to the ultra wealthy, even royalty have to pay their bills. The doorchester has secured a court order to sell a car belonging to Shake Nasser bin Abdullah Althani, a member of Qatar's ruling family. The hotel sued Althani after he allegedly ran up an unpaid bill of more than 450,000 pounds. The Shake's lawyers told the FT that he has only recently become aware of these proceedings, and that they dispute the bill. You can read more on all these stories for free when you click the links in our show notes. This is Better Daily FT News Briefing. Check back tomorrow for the latest business news.