DeekSeek Resists Bringing on Investors

[00:00:00] Ready to launch your business? Get started with the commerce platform made for entrepreneurs. Shopify is specially designed to help you start, run, and grow your business with easy, customizable themes that let you build your brand. Marketing tools that get your products out there. Integrated shipping solutions that actually save you time. From startups to scale-ups, online, in person, and on the go. Shopify is made for entrepreneurs like you. Sign up for your $1 a month trial at shopify.com slash setup.

[00:00:33] Welcome to Tech News Briefing. It's Monday, March 17th. I'm Shara Tipkin for The Wall Street Journal. Federal funding cuts are putting the bioscience industry at risk. What do these cutbacks mean for the future of drug development in the U.S.? Then, DeepSeek shook up the world of artificial intelligence when it introduced its chatbot. But that doesn't mean it's ready to take on investors.

[00:00:58] Our reporter Rebecca Fung explains why the Chinese startup isn't in a hurry to get outside investment. But first, the bioscience industry is facing federal funding cuts. As part of the Trump administration's efforts to save money, it's freezing grants that universities count on to fund their research. Courts have blocked the cuts, but it's still causing uncertainty in the bioscience

[00:01:26] industry. And that uncertainty could threaten drug development and the U.S.'s leadership in the fields. Stephen Rosenbush, bureau chief of the journal's Enterprise Technology Group, joins us now with more. Stephen, how does funding usually work? I spoke to researchers at the University of Washington in the biochemistry department, which includes the Institute for Protein Design. One of their chief investigators, David Baker,

[00:01:54] won a Nobel Prize for his work last year designing proteins not found in nature. This group relies on the government for around 90 percent of its funding. Typically, these grants are paid to support research projects that can go on for four or five years. Then they may conclude or they may be renewed for another four or five years. So these programs are functioning now, but there's a lot of uncertainty. How are universities adapting?

[00:02:23] They are cutting back. The uncertainty in funding has led to a significant reduction in the number of graduate students that the University of Washington is admitting in biochemistry and elsewhere. One of the deans there told me that throughout the school, which now actually has a very broad-based hiring freeze in place, that even prior to that broad-based university-wide hiring

[00:02:50] freeze, that graduate admissions were down somewhere in the range of 25 to 50 percent. The hiring of postdocs is either slowed or may slow, and the number of faculty positions is being reduced as well. What has the Trump administration said about the funding? I did reach out to the National Institutes of Health and to the White House. I haven't heard anything back. Are these cutbacks all part of the DOGE effort or what's driving the funding cuts?

[00:03:19] Yes, it's part of a much broader-based effort to pause, cut back, reassess federal funding, generally shrink the size of the federal government, cut back on its budget, and maybe in a larger sense, reduce its overall role in the economy. Is there any sort of sense how big these federal funding cuts are? Do we have any sort of figures?

[00:03:44] It's hard to know exactly what's been paused or what's still on pause, but the overall sense that I have is that the cuts themselves, the funding themselves, isn't an enormous number, especially relative to the size of the U.S. budget, let alone the U.S. economy. The concern that David Baker in particular has is that there really isn't that much actual cost savings to be found in terms of

[00:04:13] hard dollars, but the cost of those savings could be very, very significant to the U.S. over time when it comes to the ability to develop drugs that depend on this bioscience research and other products. What impact could funding cuts have over the long term? Over time, there's a risk that if these cuts or that uncertainty continues, that the capacity of the U.S.

[00:04:41] to produce these drugs will be at risk and that, in David's words, we could end up buying more of these products and drugs from China, which has invested very, very heavily in this scientific work over the last 15 years and which also critically has a really stable funding base. That was our Enterprise Technology Bureau Chief, Stephen Rosenbusch.

[00:05:05] Coming up, DeepSeek is hugely popular with users and investors want in, but the Chinese startup isn't so sure. We'll find out why after the break. DeepSeek has been attracting tons of attention and users with its AI chatbot that's smart but was cheap

[00:05:33] to build. Now, investors want a piece of the Chinese company, but DeepSeek doesn't want the hassle. WSJ reporter Rebecca Fung joins us from Hong Kong to talk about the latest developments. Rebecca, we've seen OpenAI, Anthropic, and other AI companies raise literally billions of dollars to fund their operations. Why is DeepSeek so resistant?

[00:05:57] What we've known is that the founder has basically told his associates that he's really not in a hurry to get any investment, partially because he's slightly afraid that outsiders, like external investors, would basically interfere with DeepSeek's decisions. So far, it's been pretty much concentrated at the top few people. And this is from people familiar with the matter. And then the founder is also quite cautious about getting government-linked money because he still wants DeepSeek to

[00:06:26] have like global adoption and getting Chinese government-backed money might impact that ambition. What can you tell us about DeepSeek's founder? Actually, not that much is known about him. His name is Liang Wenfeng. He was born in 1985. He is like a math geek, basically. We had this interesting anecdote about DeepSeek's founder and how his love for the name DeepSeek is telling.

[00:06:54] DeepSeek's name is actually, in Chinese, is Huanfeng. It basically alludes to this ancient Han dynasty diagram with like a magic square. It's basically like this peculiarity where the rows, columns, and diagonals of the squares all add up to the same number. And then Liang, the founder, is super proud of it. That is the Chinese that discovered this concept long before the West. How has he funded the company so far?

[00:07:21] DeepSeek has an interesting starting story, basically. It started as a quant fund in 2015. The quant fund is called High Flyer. This may be slightly different from other AI companies, but starting as a quant fund, it earned a lot of money from 2015 to about 2020 because it was charging pretty hefty fees usually, like as other U.S. hedge funds. And at its peak, the quant fund had about 14 billion U.S. dollars.

[00:07:49] So you could calculate that's like quite a lot of annual income. And basically, that's how DeepSeek, which is this AI company born out of High Flyer, funded its chip purchasing and like hiring AI engineers. Yeah, they're basically using the quant fund's revenue to fund their AI ambition until now, essentially. So who is wanting to invest in DeepSeek or partner with the company? A lot of people want a share of DeepSeek, but there are two group of companies.

[00:08:18] There are the investors, basically the venture capital funds, the private equity funds, who all want to invest. And some of the state-linked funds all want to invest in DeepSeek, basically fund DeepSeek's operation. And then there are another group of companies, basically the massive technology giants like Tencent, Alibaba. So what we know is like in recent weeks, DeepSeek founder, Liang, did meet with executives from Tencent and Alibaba, basically to discuss potential cooperation.

[00:08:45] And right now, how they're, I guess you could use the word partnering with DeepSeek, is that these tech giants are testing DeepSeek to power features on their own apps. For example, Tencent is testing DeepSeek on its own messaging and payment app, WeChat. Since DeepSeek's algorithms, it is open source. So Tencent actually doesn't need to pay DeepSeek to do so.

[00:09:10] And Tencent users can basically opt for a chatbot that is powered by DeepSeek, but uses Tencent's own computer network. That's roughly how they're partnering with DeepSeek. Can you talk a little bit about how we know that there's this interest? Is this based on your reporting? Has the company or these potential partners talked about it? So Tencent and Alibaba very, very openly talked about it. Because of DeepSeek's rising popularity in mainland China,

[00:09:39] like anything that has anything to do with DeepSeek, basically, like their share price would jump and they would become popular. So Tencent and Alibaba very openly talked about that collaboration. The investors is just, we talked to contacts and sources, and we knew that the potential investors, some of the funds met up with DeepSeek. As you said, DeepSeek is really popular right now. People are paying a lot of attention to it. What sort of problems are facing the company?

[00:10:08] Right now, DeepSeek's chatbot itself is a bit overwhelmed. Because after January, it's shot up in fame. It's overwhelmed by millions of users because its computing power is proven at the moment not to be enough. So its chatbot has frequent service hiccups, and it's also facing kind of external pressure from authorities around the world that they're basically restricting the use of DeepSeek's chatbot over data security concerns.

[00:10:37] So where does the company go from here? What's up next? We actually really don't know. And that's up to DeepSeek to decide whether they want to monetize this chatbot or not. Because right now, they're not monetizing it. And that's what's been keeping investors at bay because investors are like, well, if you don't want to monetize it, why would I invest in this? So it's really up to DeepSeek. Whether they want to stay in almost like a nonprofit form and open source, or they want to monetize it. And that we don't know.

[00:11:07] That was our reporter, Rebecca Fung. And that's it for Tech News Briefing. Today's show was produced by Julie Chang with supervising producer Catherine Millsop. I'm Shara Tipkin for The Wall Street Journal. We'll be back this afternoon with TNB Tech Minute. Thanks for listening.