In this YouTube short, Jason M. Lemkin highlights Anthropic’s competitive edge over OpenAI, noting that Anthropic’s model training costs are roughly a quarter of OpenAI’s while it is catching up in about half the time. He points out the compounding advantage of lower costs combined with faster growth, projecting potential revenues of $30 billion in five years. Lemkin argues that this creates a “double‑code‑red” scenario for OpenAI investors, who may have secured a worse deal compared to Anthropic’s backers. The discussion includes an Uber‑vs‑Lyft analogy to illustrate how out‑accelerating a rival while being more cost‑efficient can be a daunting game‑theory position for the incumbent.
Chapters
The one that jumped out at me when I contrast it with the fact that Anthropic has caught OpenAI, right, in half the time, is that their training costs are a quarter of OpenAI. Their training costs for models are a quarter of the OpenAI. Now, maybe that's because they're focused. They don't have to do video, they don't have to do images, they don't have to do a lot of consumer stuff. But if you just think about it for a moment, the compounding effects of catching OpenAI in half the time, right, at, at roughly the same revenue or more, $30 billion in 5 years and having training costs that for now are a quarter of it, you know, that's a double code red. It's one thing if, if, if, if, if you have two classic startups where one has bled money and it's artificial, there's other things. But if you have a dramatic cost benefit and you're out-accelerating your competitors, um, and there's management team turmoil at your competitor, uh, it really feels like the investors in OpenAI got a much worse deal in the last round than the Anthropic ones did. Just crazy. Just having both. You usually don't have both together with your competitor. You're out-accelerating your competitor and your training costs are a fraction of your competitor. Good God, that just compounds.
That's actually a good point. 'Cause you take the Uber-Lyft struggle, right? Uber had the, oh my God, we're out-accelerating. Oh, but by God, we're spending every dollar we have to do it and we show no fear, right? In this case, you're out-accelerating the opposition while being more efficient on a bunch of interesting measures. No, you're right, Jason. That's a scary fact pattern if you're, If you're running the game theory and you're the other guy, it's like, hmm, that's not good. They're growing faster than us. The gross margin economics are roughly the same, slightly better. And their costs below the line, and to a rounding error, costs are compute and scientists to run the compute for training are better.
