LLMs

“LLMs do not have that. They don't have any internal model of the world that allows them to predict.”

“Constraints are probably the single largest force that we have to improve the capabilities of these LLMs.”

“I don't know exactly what the breakthrough will be, but I'm really excited about the union of these LLMs plus reinforcement learning. And I'm excited about that because I think there's a lot more to come from reinforcement learning. And I know at Google D mine…”

“And I believe that we're not gonna quite get there, but we're gonna get pretty damn close to free guy when we build LLMs into video game technology, and that leads me to prediction four, is that video games are gonna be absolutely incredible with this stuff, a…”

“I personally don't necessarily think that the winner in LLMs will necessarily be The winner in visual creation. And so as a startup, the fact that you can pick whichever model or whichever offering an API you feel fits your needs best is great. Where in some c…”

“I think there will be an interface breakthrough that allows the search engine to gather either through things like email or history Or other inputs basically gather more of a contextual picture for you to be able to provide even better results. But I think it'…”

“To put these LLMs to work to solve real world problems, you got to actually take some of these models and then customize it, and the end result is not the biggest model. It is actually a much more customized, smaller model or a distal model to solve specific b…”

“So, I expect LLNs to evolve in the same way, where there are a lot of, ah, decisions, especially if you have a powerful computing device, ah, which, ah, many smartphones and others, ah, to be having that, You can actually run some of those, ah, simple LLM thin…”

“Now, what's funny, the reason it's complimentary is LLMs today are not very good at doing that LLM, you know, intent mapping. They make mistakes. You need to be able to check them, all those sorts of things.”

“I think they can interpolate between skills, and so if they've seen how to do A, and they've seen how to do B, they can get kind of the average of A and B but they don't just go completely beyond anything that they've seen.”

“So what we've done since then is we've invested quite a lot in this is quite recent that is rolling out in LLMs that actually tell interesting stories about the music. And we see very strong effects on this, on the retention of the application.”

“These deep learning based systems, they had flattened out in terms of if you added more user data or more parameters, they did not get better like the LLMs. There were no scaling laws. It's just like, it is what it is, and you could move at .2%. There's someth…”

“And I think what is going to happen with these LLMs is at least for some brands, you will start having literal relationships with them.”

“What tends to happen in these worlds is that the thing that is scarce gets even more valuable. So one bet would be that true human connection gets more valuable than ever. When a lot of what you talk to in the future may be LLMs.”

“A big issue there is, is that when humans or animals reason, We don't do it in token space. In other words, when we reason, we don't have to, you know, generate a text that expresses our solution and then generate another one, and then generate another one, an…”

“No, not, not in the current form. I mean, and whatever form of AI would be able to do that will not be LLMs. They might use LLMs as one component.”

“Well, I don't know if I would call it a wall, but it's certainly a diminishing return in the sense that, you know, we've kind of run out of natural text data to train those LLMs where they're already trained with, you know, on the order of you know, 10 to the …”

“One simple way of getting NNMs to kind of appear to reason is chain of thought, right? So you basically tell them to generate more tokens than they really need to in the hope that in the process of generating those tokens, they're going to devote more computa…”

“We are not going to get to human level AI by just scaling up LLMs. This is just not going to happen.”

“We also know that they can hallucinate facts that aren't true but they're really, in their purest form, they are incapable of inventing new things.”

“We are not going to get to human level AI by just scaling up LLMs. This is just not going to happen.”

“I think a big bottleneck these models have is their inability to learn on the job, to have continual learning. Their entire memory is extinguished at the end of a session. There's a bunch of reasons why I think this actually makes it really hard to get human-l…”

“And so sometimes people say, well, the reason Fortune 500 isn't using LLMs all over the place is because they're too stodgy. They're not they're not like, they're not thinking creatively about how AI can be implemented. And actually, I don't think that's the c…”

“Our competency is providing decision intelligence as early as possible to executives. That's not a competency of an LLM, but their advancement helps us do that better.”

“These type of things are going to definitely become real partners to people. People, when this technology has been bad or hardly workable, have gotten married to them. Before LLMs. So it's gonna happen again, and in greater numbers.”

“I think it's less that there's a lot of social media sort of oriented folks that have now moved into AI. It's more that I think a lot of the best product people were focused on that, you know, even four years ago, you know, pre-chat GPT you know, pre the emerg…”

“Almost every CISO I'm talking to now is, is concerned about Giving LLMs access to data for that reason. Not that the LLM is going to do something wrong, but it's going to let people search and find things way more powerfully and join information they didn't ev…”

“It does also seem like Meta is, is sort of fighting the last war on this where they're like, they've assembled a super team to build LLMs.”

“Email should be a very straightforward. It's just, it's mostly text. It's very structured. There's subject lines and like, it's something LLMs should be able to do a very good job of handling. There's a lot of context.”

“It just had different economics, and I genuinely think that's fundamentally changed. And we, we've talked about this a lot, like, and an AI company, we don't even know what the economics really are of them in terms of, like, what it costs to operate the more p…”

“So I do not buy that. In that I think LLMs will be, and I've experienced this myself, performative in a way that they think appeals to humans and hooks them emotionally.”

“They won't be banned. It's impossible. I mean, how can you ban them? Are you going to go and take your, is the government going to go grab your Mac mini out of your office where you've downloaded a version of deep seek and be like, all right, right to the poke…”

“Everyone a year from now is going to start asking much more detailed questions of Google Maps rather than saying, Restaurant Thai New York, you're going to start saying like, oh, I'm looking for the best pad Thai within a mile from me, and that's going to be p…”

“What I mean by this is today, if you want to train a top of the line, you know, our frontier model, that might cost you a couple billion dollars. But in a couple of years, that might be fifty million to, you know, a hundred million. And then a few more years o…”

“So that's why LLMs, I think, is going to go away as a The term, because they're not large language models only. They're visual. They're going to be used for robotics. They're going to be used for a lot of things.”

“The one thing that LLMs do the worst is humor. They cannot make jokes. And it's because I think, like, as you point out, they're just used to the average of averages and not throwing curveballs.”

“And of course, you know, a use case that probably 60% of the country is engaged in now is to use these LLMs for their own health diagnosis and research and people getting great value out of that.”

“We cannot set the standard that US AI models can't find bugs. That is a terrible, terrible, terrible standard. If you have a, if you are writing software with an LLM, it has to be able to understand what a bug looks like so it does not write those bugs.”

“If our American LLMs know nothing about security, they will create more security flaws. So we cannot create a standard where American LLMs are dumb about security. That would be a humongous own goal.”

“Now with LLMs you're able to reverse engineer that patch in a fraction of the time. So that part of AI and cyber is a legitimate step up in capability, which is the attacker's ability to quickly revert reverse engineer and weaponize a flaw in a patch, and then…”

“If it can solve arc AGI and it's not necessarily crushing on these economic factors and these just kind of general rote work things that we would like it to do it shows that instead of being general, it's very spiky intelligence and hence much less useful.”