During their recent episode, Taylor and Carlisle discussed China: No More Credit Card Payments. Here’s an excerpt from the episode:
Tobias: Well, we walked along the river.
Jake: The Bund?
Tobias: Is that the one? It was the Bund?
Jake: Yeah. Yeah.
Tobias: And our host said this was a cement factory three months ago. It was a mall with a whole lot of restaurants.
Jake: Yeah. Beautifully done.
Tobias: Art galleries. Yeah, it looked magnificent. Really impressive. You pay for everything. So, I think MasterCard and Visa are [chuckles] probably gone, I think, because you pay with– I use WeChat. You can pay Alipay. You can use Alipay. Every vendor has a cell phone. And if they don’t have a cell phone, they’ve got a sticker. You just scan the sticker with your app, and your app asks you if you want to make this payment. You click yes, you put in a little code and you’ve made a payment. It doesn’t run through MasterCard or Visa’s rails. It just charges your WeChat account. They say that they charge basis points for it. And in some instances, they charge nothing because they just want access to the vendor-
Jake: Yeah.
Tobias: -and the data. Yeah, so MasterCard and Visa are in trouble if that– Why isn’t somebody doing that here?
Jake: I don’t know.
Tobias: Facebook could do that here.
Jake: Because it just works right now and it’s low enough. I got the sense that DeepSeek was a pretty big moment for China, based on all the conversations, because obviously you talk about AI a lot, especially if you’re visiting VCs. But I think what it showed was that they weren’t as far behind as they worried that they were on AI. Really the CapEx that the US has been doing might who knows– If you’re able to get close to 80% of the result with 20% of the cost or less–
Tobias: Well, the numbers that they were quoting were 95% of the-
Jake: Yeah. Well, okay. [laughs]
Tobias: -quality with 5% of the cost.
Jake: Yeah.
Tobias: Which is bananas. I don’t know if it’s that either. But that’s the DeepSeek analogy that it’s like 95% of OpenAI’s 4.0 version, like their best version. And on the chart, it measures how smart these things are and how many questions it can answer. It’s right sitting, just underneath it for minimal cost.
Jake: Yeah.
Tobias: I don’t know if anybody could definitively say if that was real or if they had some training from– They had some advantage provided to them by the government, or they had spent more money than they had said that they had spent. But everybody seemed to think it was possible, because they were totally transparent with their model and their training data in the background.
Jake: Yeah. And the weightings.
Tobias: And the weightings. And so, it’s possible. One of the VCs who we talked to said– He interpreted this guy as being like an engineering type personality who just wanted to see this stuff out in the world, because he was making his money in a different– he’s making his money in the hedge fund. Yeah.
Jake: You know one thing I thought was interesting that a lot of the venture guys, I would have thought that there’d be something a little bit further out, like vision. Did you feel that same way?
Tobias: Yeah, 100%. The hard VC is like, they call it deep tech. It’s the stuff that’s like the cutting edge, the bleeding edge of tech. These guys try and put money into these really moonshotty type ideas. But then, when you come to hear what the moonshot type ideas are, they’re just what perhaps you’d expect them to be [Jake chuckles] just being a regular old– [crosstalk]
Jake: Yeah. Not even like a tech guy, right?
Tobias: They’re working on AI, they’re working on autonomous driving and they’re working on robotics.
Jake: Yeah.
Tobias: And so, they’re not working on cold fusion.
Jake: Yeah. [crosstalk] I guess some climate tech too, but I don’t even quite know what that battery.
Tobias: Yeah, we didn’t see– [crosstalk]
Jake: Yeah.
Tobias: We didn’t see what that meant.
Jake: Another thing that I noticed that I hadn’t really read anywhere else, is that it sounded like Chinese companies don’t want to use cloud-based AI solutions. Like, they want it on premises for data protection reasons. The US to me, I feel like they’re people are more comfortable, businesses are more comfortable with like “Okay, whatever. Microsoft’s got all our data.”
Tobias: Yeah, I don’t know. I just don’t know. My wife’s firm has its own internal AI, because they don’t want to share their data-
Jake: Yeah.
You can find out more about the VALUE: After Hours Podcast here – VALUE: After Hours Podcast. You can also listen to the podcast on your favorite podcast platforms here:
For all the latest news and podcasts, join our free newsletter here.
Don’t forget to check out our FREE Large Cap 1000 – Stock Screener, here at The Acquirer’s Multiple: