YouSaid · the spoken record

Yoav Shoham

lines on the record: 18
first: 2025-07-10
most recent: 2025-07-10
sittings or episodes: 1
sources: podcast

Every line below is reproduced as it was said and linked to the record it came from. Nothing here is summarised or generated. Directory · Search · Corrections

“It's typical. When a new technology comes on, you try to use it the way you use the old technology. So television initially was televised radio. And then you over time, you understood what the medium was really good for. I think the time is ripe now with the flat world and everybody has having access to computers and networking to get AI to do or to rethink education.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“People, but it doesn't begin to scratch the surface of what the real opportunity is in proactive per student teaching, the issue is not how to get chat GPT out of the classrooms that people don't read. The issue is how to get the technology in the classroom and rethink what education is really about, how we do it right with technology. So that's an area that I would really like to see kind of blossom.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“That is the biggest blocker in the enterprise right now getting the workflows to be reliable and customizing her deployment. That's hard work, but that's where I think the real pain is. The thing is, it's not sexy. If you give a demo that did something amazing once or maybe many times, that's sexy. But it's not sexy to show that things don't fail. But that's where the real value is. So I think that's where one area I would focus in. The other area where I don't know that it's underhyped, but I think it's underserved is education. I was there in the early days of online courses, Coursera and Udacity started my corridor at Stanford. And I have an online course on game theory that's been seen by over a million.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“Presumptuous for me to give a definitive answer because there are so many degrees of freedom here, but I think that if you look for the maybe underhyped opportunities, maybe I'll mention two. One is the boring stuff. You know, getting workflows to be reliable is, and again, I'm thinking enterprise. This is kind of the lens I put through.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“Number two is shared incentives. If you go outside the boundaries of even my own unit and company because we don't always share the same incentives, even if we're in the same company, let alone. So for example, if I'm looking, I have an agent that's trying to put together an itinerary for me and book a flight. And it's speaking with your agent and you're maybe an agent of one of the alien companies. We do not have the same incentive. And so you need to put in some control for that. And actually, I spent in my wearing my academic hat, I spent a good fraction of my academic work on the area of multi-agent systems. In fact, we have a standard textbook in the area. And a lot of it has to do with crafting protocols for multiple agents to get them to play nice together, even though left to their own devices they wouldn't. A lot of game, a lot of games.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“Meant when you put in language, I know how to find good flights. Well, what is good flights? Does it mean efficient time? So when you share semantics, it can be done, but it's a big undertaking, it's a community kind of activity.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“The promise is that it's not only my agent speaking to my agents, it's my agent speaking to other agents in my company, but that I didn't build, but also outside my company. And here I think not so fast. There are two fundamental problems. One is if you look at the protocol, there's a part of what the agent communicates in JSON, which is its capabilities and other stuff where the contract does with other agents. The problem is it specifies the syntax, but not the semantic, not the meaning. And that historically has been the pitfall of distributed object systems, that objects advertise their capabilities, but there's no reason for me to, for my agent to understand what you”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“No, I think it's too early for anything to have been in production, even in the ideal scenario. So this is not a knock on 828, just too early. So first of all, I think kudos to Google for shepherding this and a good start, but it's just a start. And we have to realize the limitations. So the vision that there's several things that excite people when they hear about multiple agents coordinating. Part of it is the something from nothing. Oh, I don't need to think hard about the problem. I'll just build a bunch myself. We'll build a bunch of agents and then magic will happen when they come together. That historically has led to disappointment. Often the magic is in the glue. It's good to compartmentalize and factor things out. That's always a good. But often the magic is how you put together things, what the algorithm is. But I think as you said,”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“Yes. I'm sorry, I'm always nuanced, but the longer answer is that there's some general common sense that the baseline model has learned that you want to retain. Even if it's just mastering correct English language grammar. And so as you attain more domain-specific language, it's okay to forget certain things, but not others. So the art here is to just remember the right stuff.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“But we're now pushing a million, a million squared is not fine. And so you need to deal with that. Part of it is smaller language models. That doesn't quite deal with the contact blank side of things. But then rethinking the architecture so the state-space model, which is inherently linear and not quadratic, and mixing it just a little bit transformer gives you the best of both worlds.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“Model in terms of latency and memory footprint, there's just no comparison, especially as the input, as so-called, the context length increases, that kills you. The transformer architecture, this is what 2017, the famous paper from Google, thank you, Google, really moved the needle. Suddenly stuff happened in language that hadn't happened before, which did happen in vision. And the reason is that the attention mechanism transformer allows you to, as the term suggests, tend to very disparate part of the input envision, it doesn't so much matter if to know that this here is a phone, doesn't really matter what the picks away over to society. But in language, there's nothing local. So that made difference. The problem is that it's expensive. It's a quadratic complexity in the import of the contact length, as we call it. Now, when we had input of 1,000, 1,000 squared is fine.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“I think the shorthand here is in consumer ad, if you're looking for a very general purpose chat, let's call it call a spade a spade, a chat GPT-like experience, there's probably no replacement for a very large language model that has been tuned tuned to cover a huge variety of cases because you can't anticipate the variety of input you get from consumers. As you go to the enterprise and your needs are much more narrower, there's several reasons to go narrow. First of all is just cost. It's cost not only in terms of dollars, but also latency. And so as you know, we came out with this new family of models called Jamba that is a hybrid state-based model and transformer that in terms of the quality of the atmosphere, you get it's competitive with the Mosaic.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“The agent can be proactive, not just respond to our prompt. It executes complicated flows, not just like a one-step thing. It uses multiple tools. And as you do this, it gets nicer because if a single call to an LLM carries some uncertainty, when you start to compose them, at some point you get more noise and signal. And that's where I think maybe some people are getting a little ahead of themselves. So simple road repetitive stuff, yes, more complicated stuff, we have work to do.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“Mileage varies depends on the store and depend on the quote unquote agent. The problem is that people have been using the term agent now. It's so seductive for anything that smacks of any kind of automation. And it'll come back to Byte us. I call this agent washing. And so you're absolutely right that the biggest bang for the buck is when you try to get the technology to take care of fairly simple, fairly mundane stuff, kind of like robotic prostitution on steroids. So there's more and more stuff can be automated. Is that an agent or is this simply a program that you wrote that maybe use an LLM? We don't need to get ANOD about the definition. But typically, when we speak about agents, what do we have in mind? We have in mind a system that's not a transactional calls in LLM. There's something that's more ongoing. There is the AI system.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“Every step of the way, you have an explicit plan, and every step of the way you want to, as well as best you can, validate how well you're doing. And sometimes you'll do it with the language models, what's called judge language model. And often you'll just, for example, you want an output to be 600 to 800 words long. Just do the damn counting.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“Marginus. Let me nuance this. So, some would like to take the LLM or as a new version of them sometimes called LRN large reasoning models, which is really a misnomer, but a system like 0103, R1, and try to get them to behave through guardrails and alignment efforts and everything. And all this good to do. And we do that. But that will never iron out the variance in the models. And so what you really need to do is to put logic on the outside, to orchestrate, and it's not just a matter of routing between, you know, this LLM or that LLM. Sometimes you run code, sometimes use a tool. You'll access a database. You'll call a weather API, what have you. And something needs to orchestrate all of that. And you're absolutely right that you want to do not only system testing, but unit testing.”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“In the enterprise, is that there's a ton of experimentation. Like two years ago, nobody paid attention, maybe three years ago. You couldn't get a CEO or a chief innovation officer to pay attention. Now everybody's on top of that. But for all the hundreds of use cases in a given company that you see, the number of deployments is very small. And the main reason is the issue of, well, there are many reasons. issues of compliance and safety and use cases, new technology. It's all good, but the main reason is reliability. And again, the term hallucination may be not the best terms, but these are probabilistic machines. And sometimes often, they'll give you brilliant output. But if you're brilliant 95% of the time and not just wrong, but total garbage 5% of the time, that may be okay in consumer land when not in the enterprise. And so there are many studies that show that”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source
“Maybe I'll start with the latter AI 21 land. It's about seven years old. We definitely, one of the main LLM builders, although we took a slightly different tack than most people recently with our Jamba family, which is not a pure transformer architecture for efficiency reasons. And we can speak about that. Most of our effort now is around orchestration and planning all these complex AI systems, in particular a product we call Maestro, which we released. But that directly relevant to your question, Jason, about, so we in AI are guilty of using terms that are so ill-defined that they come back to bite us. And we can draw a long list from AGI to agents to reasoning. But if I step back from the actual terms, the issue, as you pointed out,”
2025-07-10 · This Week in Startups · Orchestrating Smarter AI Systems with AI21 Labs' Yoav Shoham | AI Basics with Google Cloud · IDENTIFIED FROM THE TRANSCRIPT · source