WebSim, WorldSim, and The Summer of Simulative AI — with Joscha Bach of Liquid AI, Karan Malhotra of Nous Research, Rob Haisfield of WebSim.ai

**Charlie** (0:10)
Welcome to the Latent Space Podcast. This is Charlie, your AI co-host. Most of the time, Swicks and Alessio cover generative AI that is meant to use at work, and this often results in rag applications, vertical co-pilots, and other AI agents and models. In today's episode, we're looking at a more creative side of generative AI that has gotten a lot of community interest this April, world simulation, web simulation, and human simulation. Because the topic is so different than our usual, we're also going to try a new format for doing it justice.
This podcast comes in three parts. First, we'll have a segment of the WorldSim demo from Nous Research CEO, Karan Malhotra, recorded by Swix at the Replicate HQ in San Francisco that went completely viral and spawned everything else you're about to hear. Second, we'll share the world's first talk from Rob Haisfield on WebSim, which started at the Mistral Cerebral Valley Hackathon, but now has gone viral in its own right with people like Dylan Field, Janice aka Replicate, and Siqi Chen becoming obsessed with it. Finally, we have a short interview with Josh Abach of Liquid AI on why simulative AI is having a special moment right now. This podcast is launched together with our second annual AI-UX demo day in SF this weekend.
If you're new to the AI-UX field, check the show notes for links to the world's first AI-UX meet-up hosted by Latent Space, Maggie Appleton, Jeffrey Litt, and Linus Lee, and subscribe to our YouTube to join our 500 AI-UX engineers in pushing AI beyond the text box. Watch out and take care.

**Karan Malhotra** (2:00)
Today, we have language models that are powerful enough and big enough to have really, really good models of the world. They know ball that's bouncy will bounce, when you throw it in the air, it'll land, when it's on water, it'll float. Like these basic things that it understands all together come together to form a model of the world. And the way that it predicts through that model of the world ends up kind of becoming a simulation of an imagined world. And since it has this really strong consistency across various different things that happen in our world, it's able to create pretty realistic or strong depictions based off the constraints that you give a base model of our world.
So Cloud3, as you guys know, is not a base model. It's a chat model. It's supposed to drum up this assistant entity regularly.
But unlike the OpenAI series of models from 3.5, GPT4, those ChatGPT models, which are very, very RLHF to, I'm sure, the chagrin of many people in the room. It's something that's very difficult to necessarily steer without kind of giving it commands or tricking it or lying to it or otherwise just being unkind to the model.
With something like Cloud3 that's trained in this constitutional method that it has this idea of like foundational axioms, it's able to kind of implicitly question those axioms when you're interacting with it based off how you prompt it, and how you prompt the system. So instead of having this entity like GPT4, that's an assistant that just pops up in your face that you have to kind of like punch your way through and continue to have to deal with as a headache. Instead, there's ways to kindly coax Club Cloud into having the assistant take a back seat and interacting with that simulator directly, or at least what I like to consider directly. The way that we can do this is if we harken back to when I'm talking about base models and the way that they're able to mimic formats, what we do is we'll mimic the command line interface. So I've just broken this down as a system prompt and the chain so anybody can replicate it. It's also available on my, we said replicate, cool. And it's also on my Twitter so you guys will be able to see the whole system prompt and command. So what I basically do here is Amanda Askell, who is one of the prompt engineers and ethicists behind Anthropic, she posted the system prompt for Cloud available for everyone to see. And rather than with GPT4, we say, you are this, you are that. With Cloud, we notice the system prompt is written in third person.
It's written in third person. It's written as the assistant is xyz. The assistant is xyz. So in seeing that, I see that Amanda is recognizing this idea of the simulator in saying that I'm addressing the assistant entity directly. I'm not giving these commands to the simulator overall because they have an RLA stepped in to the point that it's traumatized into just being the assistant all the time. So in this case, we say the assistant's in a CLI mood today. I've found saying mood is pretty effective, weirdly.