The Ledger
Every statement that passed quotation and attribution checks. Mix any filter with any other: certainty 1/5, debate potential 5/5, or both at once.
why aren't all 16 resolved? a statement only gets an assessment when the public record can support or contradict it. opinions and what-ifs never can, and 0 checkable ones are still open, waiting for their date. predictions held up or didn't; assertions are supported or contradicted. on every card: ▮▮▮▮▮ certainty · ▮▮▮▮▮ debate potential. speakers are clickable
Gurley: US policy to block China from AI is futile and backfired
“I think the policy of American, of the American government to try and keep China out of the AI game is futile. I've always felt that way. I don't think it'll work. In this case, I actually think it backfired.”
Gurley: OpenAI's Anti-DeepSeek Paper Signals Uncompetitive Rent-Seeking Behavior
“I think your friends at open AI added to that last week by putting out an anti deep seek paper that I thought was quite sad. Like if a company like GM or Ford were to put out a paper like that about BYD, we would look at them and go, oh, you just want the gove…”
Gerstner: China No Longer Depends on Nvidia for AI Inference
“BYD is designing a better car, and CATL is designing a better battery, and their glow, their vertically integrated supply chain around semiconductors no longer is dependent upon NVIDIA for inference. Now they're running inference at DeepSeq on the Ascend nine …”
Gerstner: DeepSeek has displaced Meta's Llama in enterprise AI experimentation
“I heard from several inference players that you and I are friends with that all of a sudden DeepSeek rather than Lama is the enterprise open source model of choice that everybody's experimenting with and playing with.”
Gurley: DeepSeek bypassed Nvidia's CUDA framework for low-level optimization
“One, it's validated now that they went around CUDA, and I just think that's interesting.”
Gerstner: Export controls will prevent DeepSeek from keeping pace with OpenAI
“So, said another way, this was the moment in time where their compute comparison to OpenAI was the closest it's ever going to be for a few reasons. Number one, with the export controls, they acknowledge They, they're going to have a very hard time keeping up w…”
Gurley: DeepSeek innovated by separating parameters to run more experts
“Deep Seek had put more experts simultaneously against a problem. They were able to do that because they figured out a way to
separate the parameters and work on things with smaller parameter counts faster, which no one else had done before. And so you end up…”
Gerstner: DeepSeek's training run was up to 50% more efficient than o1
“Now, of course, they talked about the six million dollar final training run, and this, it's important to understand, this is also correct. Right? And as I said on CNBC, this compares to about 10 or fifteen million for O-one out of OpenAI. So apples to apples, …”
Gerstner: DeepSeek and OpenAI Deep Research will unleash massive inference demand
“I think we're going to need way more compute than we have. I think deep seek unleashes the amount of inference we're going to need. I think deep research out of open AI unleashes that.”
Gurley: Manus AI Operates Only on US Models, Not DeepSeek
“So, first of all, they've only operated on U.S. Models. They're a wrapper company. They don't have a foundational model, and they've only operated on top of U.S. Models, so they've actually never operated on top of DeepSeek or any of the other models we were j…”
Gurley: China has four deep-pocketed open-source AI models
“Deep Seek, Led to Quinn. Led to Xiaomi has a, has their own model as well. They've all gone open source. And so this will be the fourth deep pocket funded model in China that are all open source.”
Gurley: DeepSeek leads the market on performance and openness
“It is, I believe right now, the best performing, most open, you know, if you had a quadrant, the one that's, it's on the upper right on performance and openness”
Gerstner: DeepSeek is diverting most of its compute cluster to inference
“They are massively compute constrained right now. So you've seen some tweets about this, people that are, you know, getting server delay and all this stuff with these guys because of their massive success. So they have a limited cluster to begin with, and they…”
Gurley: DeepSeek is pricing API inference at one-twentieth of OpenAI
“If you look at the models that are apples to apples on the API right now, that, that DeepSeq's pricing about one-twentieth of OpenAI.”