Pod Insights Odd Lots The AI Model That Tanked the Stock Market About this episode On Monday, the stock market tanked, seemingly in reaction to the emergence of DeepSeek, an open source AI model developed in China. Nvidia, the semiconductor giant that has been the largest winner of the AI boom, erased $589 billion in market cap, for the biggest one-day wipeout in US stock-market history. Other chipmakers and big tech giants also swooned. So how did DeepSeek do it? Is it a big threat to the American AI giants like OpenAI and Anthropic? What does this say about export restrictions on US chips? On this special emergency session of the podcast, we spoke with Zvi Mowshowitz, an AI expert who authors the excellent Substack, Don’t Worry About the Vase. He answered all our questions and more to help understand what it means. Read more: AI-Fueled Stock Rally Dealt $1 Trillion Blow by Chinese UpstartWorld’s Richest People Lose $108 Billion After DeepSeek Selloff Only Bloomberg.com subscribers can get the Odd Lots newsletter in their inbox — now delivered every weekday — plus unlimited access to the site and app. Subscribe at bloomberg.com/subscriptions/oddlots See omnystudio.com/listener for privacy information.
Transcript Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
0:00 Speaker 00 Finance is increasingly being built on digital rails. But as the adoption of virtual assets, tokenization and stable coins grows, can regulation not only keep pace but help accelerate it? Find out more later in the podcast.
0:15 Speaker 01 Games on a break? Don't doom scroll. Try rips on Underdog. Buy a digital pack, rip it open and reveal a real physical card. Football, basketball, even Pokemon. Packs start at just $5. Ship it straight to your door or sell it back instantly at full market value. Max value up to $125,000. Download Underdog and get in on the chase. Deposit $5, get five free packs. Real rips, real cards, real cash. Only on Underdog. Unleash your dog. 18 plus, terms apply. Top grails are extremely rare.
0:48 Speaker 02 Bloomberg Audio Studios. Podcasts, radio, news.
1:03 Joe Weisenthal? Hello and welcome to another episode of the Odd Lots Podcast. I'm Joe Weisenthal.
1:08 Tracy Alloway? And I'm Tracy Alloway.
1:10 Joe Weisenthal? Tracy, the deep seek sell off.
1:12 Tracy Alloway? That's right. It's pretty deep. Has anyone made that joke yet?
1:15 Joe Weisenthal? We're in deep seek. Yeah. I don't think anyone has made that joke.
1:19 Tracy Alloway? I will say, like, you know, it's bad in markets when all the headlines are about standard deviations.
1:25 Joe Weisenthal? Yeah, right.
1:25 Tracy Alloway? And then, and then you know, it's really bad when you see people start to say, it's not a crash, it's a healthy correction.
1:33 Joe Weisenthal? Yes.
1:33 Tracy Alloway? That's the real cope.
1:34 Joe Weisenthal? But just for like real scene setting, you know, we've done some very timely interviews about tech concentration in the market lately and how so much of the market is this big concentrated bet on AI, et cetera. Anyway, on Monday, I think people will be listening to this on Tuesday, markets got clobbered. Nvidia, one of the big winners as of the time I'm talking about this, 3:30 PM on Monday, down 17%. So we're talking major losses really across the tech complex. Basically, it seems to be catalyzed by the introduction of this high performance, open source, Chinese AI model.
2:10 Joe Weisenthal? Called DeepSeek. I was born from what we know out of a hedge fund. Apparently it was very cheap to train, very cheap to build. You know, the tech constraints at this point didn't seem to be much of a problem. They may be a problem going forward. But yes, here is something the entire market betting on a lot of companies making AI and are now concerned about, of course, a cheap Chinese competitor.
2:31 Tracy Alloway? I just realized, Joe, this is actually your fault, isn't it?
2:34 Joe Weisenthal? Yeah.
2:34 Tracy Alloway? Last week you wrote that you were a DeepSeek AI bro and look what you've done. You've wiped $560 billion off of Nvidia's market cap.
2:44 Joe Weisenthal? Might be, might be.
2:45 Tracy Alloway? That's you.
2:45 Joe Weisenthal? Anyway, one of the interesting questions, though, is that this was sort of announced in a white paper in December. Why did it take till January 27th for really to freak people out? Big questions. Anyway, let's jump right into it. We really do have the perfect guest. Someone who was here for our election eve special, a guy who knows all about numbers and AI and quant stuff. And he writes a Substack that has become for me a daily absolute must read where he writes an extraordinary amount. I don't even know how he writes so much on a given day. We're going to be speaking with Zvi Mashavitz. He is the author of the Don't Worry About the Vase blog or Substack. Zvi, you're also a DeepSeek AI bro. You've switched to using that?
3:28 Speaker 05 So I use a wide variety of different AIs. So I will use Claude from Anthropic. I will use O1 from ChatGPT from OpenAI. I'll use Gemini sometimes and I'll use perplexity for web searches, but I'll, yeah, I'll use R1, the new DeepSeek model for certain types of queries where I want to see how it thinks and like see the logic laid out. And then I can judge like, did that make sense? Do I agree with that?
3:54 Tracy Alloway? So one of the things that seems to be freaking people out as well as the market is that purportedly this was trained on like a very low cost. Something like $5.5 million for DeepSeek V3. Although I've seen people erroneously say that the $5.5 million was for all of its R1 models. And that's not what it says in the technical paper. It was just for V3. But anyway, oh, I should mention, it also seems like a big chunk of it was built on Llama. So they're sort of piggybacking off of others' investment. But anyway, $5.5 million to train.
4:33 Tracy Alloway? Is that A, realistic? And then B, do we have any sense of how they were able to do that?
4:39 Speaker 05 So we have a very good sense of exactly what they did because they're unusually open. And they gave us technical papers that tell us what they did. They still hid some parts of the process, especially with getting from V3, which was transferred of the 5.5 million to R1, which is the reasoning model for additional millions of dollars, where they tried to make it a little bit harder for us to duplicate it by not sharing their reinforcement learning techniques. But we shouldn't get over anchored or carried away with the 5.5 million dollar number. It's not that it's not real, it's very real. But in order to get that ability to spend 5.5 million dollars and get the model to pop out, they had to acquire the data. They had to hire the engineers, they had to build their own cluster. They had to over optimize to the bone their cluster because they're having problems with chip access thanks to our export controls. And they're training on H800s.
5:26 Speaker 05 And the way that they did this was they did all these sorts of mini little optimizations, including like just exactly integrating the hardware, the software, everything they were doing in order to train as cheaply as possible on 15 trillion tokens and get the same level of performance or, you know, close to the same level of performance as other companies have gotten with much, much more compute. But that doesn't mean that you can get your own model for 5.5 million dollars, even though they told you a lot of the information. In total, they're spending hundreds of millions of dollars to get this result.
5:57 Joe Weisenthal? Wait, explain that further. Why does it still take hundreds of millions? And does this mean if it takes hundreds of millions of dollars that the gap between what they're able to do versus the, say, American labs is perhaps not as wide as maybe people think?
6:10 Speaker 05 Well, what DeepSeek is doing is they have less access to chips. They can't just buy Nvidia chips the same way that, you know, OpenAI or Microsoft or Anthropic can buy Nvidia chips. So instead, they had to make good use, very, very efficient killer use of the chips that they did have. So they focused on all of these optimizations and all of these ways that they could save on compute. But in order to get there, they had to spend a lot of money to figure out how to do that and to build the infrastructure to do that. And, you know, once they knew what to do, it cost them $5.5 million to do it. And they've shared a lot of that information. And this has dramatically reduced the cost of somebody who wants to follow in their footsteps and train a new model, because they've shown the way of many of their optimizations that people
6:55 Speaker 05 didn't realize they could do or didn't realize how to do them, that can now very easily be copied. But it does not mean that you are $5.5 million away from your own V3.
7:04 Tracy Alloway? So the other thing that is freaking people out is the fact that this is open source, right? We all remember the days when OpenAI was was more open, and now it's moved to closed source. Why do you think they did that? And like, how big a deal is that?
7:21 Speaker 05 So this is one of those things where they have a story, and you can believe their story or not believe their story. But their story is that they are essentially ideologically in favor of the idea that everyone should have access to the same AI, that AI should be shared with the world, especially that China should help pump out its own ecosystem, and that they should help grow all of the AI for the betterment of humanity. And they're going to get artificial general intelligence, and they are going to open source that as well. And this is their, the main point of DeepSeek. This is why DeepSeek exists. They're disclaiming even having a business model, really. And, you know, they're they're an outgrowth of a hedge fund, and the hedge fund makes money. And maybe they can just do this if they choose to do that. Or maybe they
8:03 Speaker 05 will end up with a different business model. But it was obviously very concerning from a lot of angles if you open source increasingly capable models, because, you know, artificial general intelligence means something that's, you know, as smart and capable as you and I, as a human. And perhaps more so. And if you just hand that over in open form to anybody in the world who wants to do anything with it, then we don't know how dangerous that is, but it's existentially risky at some limit to unleash things that are smarter, more capable, more competitive than us, that are then going to be free and loose to, you know, engage in whatever any human directs them to do.
8:44 Tracy Alloway? I have a really dumb question, but I hear people say artificial general intelligence all the time, AGI. What does that actually mean?
8:53 Speaker 05 There is a lot of dispute over exactly what that means. The words are not used consistently, but it stands for artificial general intelligence. Generally, it is understood to mean you can do any task that can be done on a computer, that can be done cognitively only, as well as a human.
9:12 Joe Weisenthal? I mean, it does. Most of these things do things much better than me. I don't know how to code. And so, but I get that there are still some things, maybe they wouldn't be as good as approving some of the, are you a human tests. Everyone's talking about Jevon's paradox. And so, we see Nvidia and Broadcom shares, these chip companies, they're getting crumbled today. And one of the theories like, oh, no, with all these optimizations and so forth, researchers will just use those and they'll still have max demand for compute. And so, it won't actually change the ultimate end for compute. How are you thinking about this question?
9:44 Speaker 05 So I'm definitely a Jevon's paradox, bro, right now, from the perspective of this debate.
9:49 Joe Weisenthal? So that you don't think it'll have a negative impact and just the amount of compute demanded?
9:53 Speaker 05 The tweet I sent this morning was Nvidia down 11% pre-market on news that its chips are highly useful. And I believe that what we've shown is that, yes, you can get a lot more, in some sense, out of each Nvidia chip than you expected. You can get more AI. And if there was a limited amount of stuff to do with AI, and once you did that stuff, you were done, then that would be a different story. But that's very much not the case. As we get further along towards AGI, as these AIs get more capable, we're going to want to use them for more and more things, more and more often. And most importantly, the entire revolution of R1 and also OpenAI's O1 is inference time compute. What that means is every time you ask a question,
10:37 Speaker 05 it's going to use more compute, more cycles of GPUs to think for longer, to basically use more tokens or words to figure out what the best possible answer is. And this scales, not necessarily without limit, but it scales very, very far. So OpenAI's new O3 is capable of thinking for, you know, many minutes. It's capable of potentially spending, you know, hundreds or even in theory, thousands of dollars or more on individual query. And if you knock that down by an order of magnitude, that almost certainly gets you to use it more for a given result, not use it less, because that is in fact starting to get prohibitive. And over time, you know, if you have the ability to spend remarkably little money
11:18 Speaker 05 and then get things like virtual employees and abilities to answer any question under the sun. Yeah, there's basically unlimited demand to do that or to scale up the quality of the answers as the price drops. So I basically expect that as fast as Nvidia can manufacture chips and we can put them into data centers and give them electrical power, people will be happy to buy those chips.
11:40 Tracy Alloway? At the risk of angering the Jevons paradox bros, just to push on the Nvidia point a little bit more. So my understanding of DeepSeek is that one of the reasons it's special is because it doesn't rely on like specialized components, custom operators. And so it can work on a variety of GPUs. Is there a scenario where, you know, AI becomes so free and plentiful, which could in theory be good for Nvidia, but at the same time, because it's easy to run on a bunch of other GPUs, people start using, you know, more like ASIC chips, like customized chips for a specific purpose?
12:20 Speaker 05 I mean, in the long run, we will almost certainly see specialized inference chips, whether they're from Nvidia or they're from someone else. And we will almost certainly see various different advancements. The today's chips are going to be obsolete in a few years. That's how AI works, right? There's all these rapid advancements, but.
12:37 Speaker 05 You know, I think Nvidia is in a very, very good position to take advantage of all this. I certainly don't think that like you'll just use your laptop to run the best AGIs and therefore we don't have to worry about buying GPUs is a poor position. It's certainly possible that rivals will come up with superior chips. That's always possible. Nvidia does not have a monopoly, but Nvidia certainly seems to be in a dominant position right now, me.
13:15 Speaker 00 What's next for virtual assets? I'm Jacob Greaves, host of Economy Scale, a series exploring growth in key global sectors. And in this episode, we're diving into the digital forces shaping finance. Sean McHugh from Dubai-based Virtual Assets Regulatory Authority has this to say.
13:30 Speaker 06 We're seeing a lot of institutional players come in, digital asset dedicated hedge funds, high frequency traders and that sort of thing, and folks who service institutional clients.
13:39 Speaker 00 As more capital enters the market, adaptive regulation is likely to become increasingly key. Reese Merrick of Ripple says.
13:47 Speaker 07 A lot of regulators globally would like to keep the industry at arm's length. That's the opposite here. I think also what you'll find is that regulation and frameworks can come out, but some of them are maybe a slight iteration on some 80-year-old policy and not actually in line with what the technology can do and offer. And again, that's very different in Dubai.
14:06 Speaker 00 Listen to the full episode wherever you get your podcasts. This series is brought to you by Bloomberg Media Studios and the Dubai Department of Economy and Tourism.
14:46 Joe Weisenthal? It seems to me, I mean, I know there's others, but it seems to me in the US, there's like three main AI producers of models that people know about. There's OpenAI, there's Claude, and then there's Meta with Llama. And it's worth knowing that Meta is green today, that the stock is actually up as of the time I'm talking about this, 1.1%. Just go through each one real quickly, how the sort of DeepSeek shock affects them and their viability and where they stand today.
15:15 Speaker 05 I think the most amazing thing about your question is that you forgot about Google.
15:19 Joe Weisenthal? Oh, yeah, right. Yeah, that's very telling, isn't it?
15:21 Speaker 05 But everyone else has forgotten about Google as well.
15:24 Joe Weisenthal? Gemini, yeah.
15:24 Speaker 05 It wasn't that surprising.
15:25 Joe Weisenthal? Yeah, that's a very-
15:26 Speaker 05 Gemini flash thinking, their version of O1 and R1 got updated a few days ago. And there are many reports that it's actually very good now and potentially competitive. And effectively, it's free to use for a lot of people on AI Studio. But nobody I know has taken the time to check and find out how good it is because we've all been too obsessed with being DeepSeek pros.
15:47 Joe Weisenthal? Yeah.
15:47 Speaker 05 And Google's had its rhetorical lunch eaten over and over and over again. December, like OpenAI would come out with advance after advance after advance. Then Google would have advance after advance after advance. And Google's would be seemingly actually, if anything, more impressive. And yet everyone would always just talk about OpenAI. So this is not even new. Something is going on there. So in terms of OpenAI, OpenAI should be very nervous in some sense, of course, because they have the reasoning models. And now their reasoning model has been copied much more effectively than previously. And the competition is a hell of a lot cheaper than what OpenAI is charging. So it's a direct threat to their business model for obvious reasons. And it looks like their lead in reasoning models is smaller and faster to undo than you would expect. Because if DeepSeek can do it, of course Anthropic and Google can do it and everyone else can do it as well.
16:32 Speaker 05 Anthropic, which produces Claude, has not yet produced their own reasoning model. They clearly are operating under a shortage of compute in some sense. So it's entirely possible that they have chosen not to launch a reasoning model even though they could or are not focused on training one as quickly as possible until they have addressed this problem. They're continuously taking investment. We should expect them to solve their problems over time, but- They seem like they should be directly concerned because they're less of a directly competitive product in some sense. But also, they tend to market to effectively much more aware people, so their people will also know about DeepSeek and they will have a choice to make. If I was Meta, I would be far more worried, especially if I was on their Gen AI team
17:15 Speaker 05 and wanted to keep my job, because Meta's lunch has been eaten massively here, right? Meta with Llama had the best open models, and all of the best open models were effectively fine tunes of Llama. And now, DeepSeek comes out, and this is absolutely not in any way a fine tune of Llama. This is their own product. And V3 was already blowing everything that Meta had out of the water. R1, there are reports that it's better than their new version that they're training now. It's better than Lava 4, which I would expect to be true. And so, there's no point in releasing an inferior open model if everyone on the open model community just being like, why don't I just use DeepSeek?
17:55 Joe Weisenthal? Tracy, it's interesting that, as Vee said, the people who should be nervous are the employees of Meta, not Meta itself, because Meta is up. And so, you gotta wonder, it's like, well, maybe they don't, I don't know, maybe they don't need to invest as much in their own open source AI if there's a better one out there, and now the stock is up. Anyway.
18:14 Speaker 05 The market has been very strange, from my perspective, on how it reacts to different things that Meta does. For a while, Meta would announce, we're spending more in AI, we're investing in all these data centers, we're training all of these models, and the market would go, what are you doing? This is another metaverse or something, and we're gonna hammer your stock and we're gonna drag you down. And then with the most recent $65 billion announced spend, then Meta was up. Presumably, they're gonna use it mostly for inference, effectively, in a lot of scenarios, because they have these massive inference costs to wanna put AI over Facebook and Instagram. So, you know, if anything, like, you know, I think the market might be speculating that this means that they will know how to train better Lamas that are cheaper to operate, and their costs will go down, and then they'll be in a better position.
18:58 Speaker 05 And that theory isn't crazy.
19:01 Tracy Alloway? Since we all just collectively remembered Google, I have a question that's sort of been on the back, in the back of my mind. And I think Joe has brought this up before as well. But like when Google debuted, it took years and years and years for people to sort of catch up to, to the search function. And actually, no one ever really caught up.
19:22 Joe Weisenthal? Never really, yeah.
19:23 Tracy Alloway? So Google has like dominated for years. Why is it when it comes to these chatbots, there aren't like higher, wider moats around these businesses?
19:34 Speaker 05 So one reason is that everyone's training on roughly the same data, meaning the entire internet and all of human knowledge. So it's very hard to get that much of a permanent data edge there unless you're creating synthetic data off of your own models, which is what OpenAI is plausibly doing now. Another reason is because everybody is scaling as fast as possible and adding zeros to everything on a periodic basis. In calendar time, it doesn't take that long before your rival is going to have access to more compute than you had, and they're copying your techniques more aggressively. There's just a lot less secret sauce. There's only so many algorithms. Fundamentally, everyone is relying on the scaling law. It's called the bitter lesson is the idea that, you know, you just scale more, you just use more compute, you just use more data, you just use more parameters.
20:18 Speaker 05 And DeepSeek is saying, maybe you don't, you can do more optimizations, you can get around this problem and still get a superior model. But mostly, yeah, there's been a lot of just, I can catch up to you by copying what you did, also because I can see the outputs, right? I can query your model and I can use your model's outputs to actively train my model. And you see this in things like most models that get trained, you ask them, who trains you? And they will often say, oh, I'm from OpenAI.
20:49 Joe Weisenthal? The internet's gotten so weird. I just, the internet is so weird. Svej Masvidz, thank you so much for running over to the Odd Lots and helping us record this emergency pod on the DeepSeek sell-off. That was fantastic.
21:01 Speaker 05 All right, thank you.
21:14 Joe Weisenthal? Tracy, I love talking to Zvi. We gotta just sort of make him our AI, our AI guy.
21:20 Tracy Alloway? I mean, to be honest, we could probably have him back on again this week.
21:24 Joe Weisenthal? I know.
21:24 Tracy Alloway? Because there's gonna be stuff happening, right?
21:26 Joe Weisenthal? Maybe we will. And obviously, we could go a lot longer. This is a really exciting story. This is a really exciting story, and things are just getting really weird these days.
21:35 Tracy Alloway? It is kind of crazy how fast all of this is happening. And then the other thing I would say is just the bitter lesson, great name for a band.
21:45 Joe Weisenthal? Oh, totally, totally great. Maybe when we do our AI-themed prog rock band, Tracy, that could be our name.
21:52 Tracy Alloway? Yes, let's do that. Okay, shall we leave it there?
21:54 Joe Weisenthal? Let's leave it there.
21:55 Tracy Alloway? This has been another episode of the Odd Lots Podcast. I'm Tracy Alockway. You can follow me at Tracy Alockway.
22:01 Joe Weisenthal? And I'm Joe Weisenthal. You can follow me at theStalwart. Follow our guest, Zvi Moshavitz. He's at theZvi. Also, definitely check out his free Substack. It's a must-read for me. Don't worry about the Voz. Really great stuff every single day. Follow our producers, Carmen Rodriguez at CarmenArman, Dashell Bennett at Dashbot, and Kel Brooks at KelBrooks. For more Odd Lots content, go to Bloomberg.com/oddlots. We have transcripts, a blog, and a newsletter. And you can chat about all of these topics 24/7 in our Discord, discord.gg/oddlots. Maybe we'll get Zvi to do a Q and A in there with people.
22:35 Tracy Alloway? Oh, yeah. That'd be great. And if you enjoy Odd Lots, if you like it when we roll out these emergency episodes, then please leave us a positive review on your favorite platform. Thanks for listening.
23:19 Speaker 01 Underdog already has team picks, player props, money lines, spreads, you name it, they got it. But apparently, somebody over there said, "You know what's missing? Screaming like a maniac about a rookie card on your couch." And we love them for that. Because Underdog now has rips. You get a digital pack, starting at just five bucks. You check the live odds, see which grails you could pull, then rip it open to reveal a real, physical card. And the grails are no joke. They have cards up to $125,000 in value. And plenty to pull. From football and basketball to Pokemon. With new, limited edition packs dropping regularly. Even cooler, you can decide to keep the card and Underdog will ship it straight to your home. Or sell it back instantly at 100% of its fair market value.
24:03 Speaker 01 And the cash goes right in your account. No listing it online, no stranger lowballing you in the messages. Just rip, reveal, and cash out instantly. Or make some room on your shelf. Download Underdog and get in on the chase. Get five free rips packs when you deposit $5. Real rips, real cards, real cash. Only on Underdog. Unleash your dog. 18+, terms apply. Top grails are extremely rare.
Previous How Oaktree's Howard Marks Spots a Market Bubble Next Get Ready For Another Shock to Housing Affordability Page updated November 5, 2026 · transcript and summary are machine-generated
