LLMs

“So actually the next frontier for LLMs is in enterprises, and I believe the best LLMs for this next, next wave essentially will be enterprise LLMs.”

“It's really much easier for someone who's maybe not be as sophisticated with SQL to suddenly start asking and answering questions of their data in ways that would have taken them three, six, 12 months to learn beforehand.”

“Yeah, I think even just zooming back out at a technical level, I think actually public data is running out for all, all of what LLMs can take advantage of.”

“And so if I take like a blog post that I've written, and then I ask one of these models to produce like three or four suggested blog post titles, I actually get three to four times Better click-through rates on the titles, because I'm aggregating all that, tha…”

“The learning is, as of six months ago, that has changed, and it is working absolutely. It's making the job seamless. Like, the LLMs now, inside Google Ads, are making targeting better, ok? They're better at showing the right creative. They're better at showing…”

“For example, like all of the YC companies that are doing like AI agents that does the entire SDR for you, like the output of that is like, eh, it's not great. It's okay, but it's not great, ok? What we're seeing working, however, is like very specialized LLMs …”

“And unless AI SDRs are able to do that, I think it might be a temporary dead end for now until we get an order or two of magnitude of improvements in LLMs, which could happen very quickly.”

“The LLMs are not good at planning, right? They're good at predicting the next step. And as part of that, you can't rely on LLMs to do the end-to-end Processing yet.”

“In AI, you cannot have predictability. LLMs are not predictable even on that.”

“They claim million tokens, ten million tokens. When you start going up to that size of input. The quality deteriorates on the output”

“They're going to LLMs first. So that, that has been a significant shift. What they do is they first, they go to the LLMs for their research process, their discovery process. And what we're seeing is that they're, they really do trust it. And so when they cite …”

“My go-to-market team uses now about 90% of the people uses LLMs every week.”

“We have seen cases where our data and systems team built things with LLMs, but it was lacking the business context, and therefore the models didn't work very well.”

“I think we've already crossed a threshold. Yeah. For which the context links Is larger than any context that human can have. Right. And I think that's huge because, I mean, just think back two years ago we had 16 K context link. Right now we have a one million…”

“I think one of the things that's happened with LLMs, we're now very firmly in this world of, hey, it's all about building the software that goes around the core LLM, the harness as we speak, as we talk about it.”

“So instead, what we're using is we're using like all the LLMs to help come up with the plans, but the plans themselves Of how you're going to do a recovery in a cyber scenario are deterministic and then are explainable to the customer.”

“I do still believe that, like, AI agents are not that good today actually at doing something that is very relationship driven. But they're getting much better at it, right? So, like, the problem is when you look at LLMs, they look at the distribution site, and…”