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 12 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
Andreessen: Google could have built a GPT-4-level ChatGPT by 2019
“Google developed the transformer in 2017. And then they basically let it sit on the shelf, right? Cause it was a research project. They didn't productize it. They were very worried about, you know, from people I've talked to, they were very worried about the, …”
Andreessen: Google Could Have Built a GPT-4 Level Chatbot by 2019
“I talked to somebody senior who was there at the time and I asked them, you know, when could you have had chat GPT with GPT four level output if you had just gotten, you know, gone, gone flat out starting in 2017. And they said by 2019.”
Douglas: Generalist models will obsolete specialized fine-tuned models
“I really do think that similar to how we saw with large pre-trained models before with small fine-tuned models made it like, had gains over the sort of GPT-II era, but then were obsoleted by GPT-IV being generally good at everything. I think, to be honest, you…”
Coogan: DeepSeek's training data relies heavily on OpenAI's GPT-4 API
“I think the training data, a lot of it is synthetic data that's been pulled from the GPT-IV API.”
Coogan: DeepSeek-V2 matched near-GPT-4 performance with one-twentieth the FLOPs
“DeepSeq V-II training required one-twentieth as much energy, essentially, one-twentieth the flops of GPT-IV, while not being far off in performance.”
Coogan: AI training scaling laws show diminishing returns since GPT-4
“When you 10 X the energy and tokens and data and all the work on the training, you do get a smarter model, but the increases have been diminishing. So as you spend 10 times more, It, you don't get the same increase from GPT three to GPT four.”
Altman: OpenAI knows how to build a GPT-4 equivalent for video
“It was really not until GPT-IV where these text models started providing real value for people, and we know how to go make the GPT-IV equivalent of video models, and we will do that, and then a lot of these things that are currently annoying, like doors, or, y…”
Srinivas: Launching Perplexity around GPT-4 would have been too late
“You want to build at the right moment where, like, kind of, like, how we built perplexity around the GPT 3.5 time, not GPT four time. It would have been too late then.”
Srinivas: Most ChatGPT users do not know o1 or GPT-4 differences
“In fact, most people using tracks between the world don't even know there's a model called O one or O three and don't even know what the difference is for GPT four.”
Coogan: OpenAI o1 is built on the GPT-4 foundation model
“The foundation model, yeah, it's still running GPT-IV, but obviously there's a lot of there's a lot of training to have that happens on top and a lot of UI stuff.”
Will Brown: Morgan Stanley deployed integrations on day GPT-4 launched
“Like, the day GPT-IV launched, Morgan Stanley had integrations, because we had been working on it, and these were, like, we had press releases for these, like, we are ready to go”
Coogan: Ramp significantly accelerated receipt scanning using Google OCR and GPT-4
“Early on, I remember they were they were one of the huge beneficiaries of GPT-IV for the receipt scanning because they could just do exactly what we did with the red book where they would OCR it with Google, take all the text in the picture, and then throw tha…”