Verifiable Inference: Don't Trust, Verify | Crypto x AI Event
Episode publishedPodcast: The Delphi PodcastThe Delphi Podcast
Transcript preview
Preview only
Before we get started, we'd like to thank our sponsors for making this event possible. Special thanks to our platinum sponsor, Olas. Olas enables everyone to own a share of AI, specifically autonomous agent economies. We're also excited to highlight our silver sponsors, NEA, empowering decentralized applications and blockchain ecosystems. Venice AI, a private and uncensored alternative to popular AI apps. MIRROR, unified AI infrastructure secured by crypto. And Therick, the first on chain multi agent system. To learn more about all of our sponsors, check the description below and dive in. Enjo
Hi, everyone. Welcome to our panel on verifiable inference. I'm your host, Luke Saunders. I'm joined by the founders of 4 leading crypto AI projects, which are either directly building verifiable inference solutions or have a need to integrate inference in some trust minimized way. Through the session, we'll touch on what verifiable inference is, why it's important, dig into some real world applications, and some discussion on the different approaches these guys have taken. So, yeah, I'd love to kick it off with some some intros. If you could all do, like, a a minute or or 2 on who you are and
Thanks, Luke. So I'm Colin, one of the cofounders here at Inference Labs. We're building a network for verified inference, really focused on, you know, solving the hard problem of both, how do you optimize, you know, zk system for verification, but at the same time, how do you do maybe some more soft, verification using cryptoeconomics. We'll get more into that later on. We're also deployed on BitTensor. We have a subnet, Omron, subnet 2. It's a top 10 subnet, really focused around incentivizing, improvements in the z k space as well as in verified inference. So, I'll leave it there.
Awesome. Good stuff. Yeah. Ryan, you wanna go next? Hey.
Yeah. I'm Ryan. I'm one of the cofounders of SphereOne. We are essentially building AI agents to make blockchain transactions and processes easy to use and automate for payments. Nice. Jeremy?
Yep. So eyes on network, we're basically building the most scalable and also the fastest network dedicated to verifiable AI compute. So I think inference is part of the solution that we're providing. And also, we can go all the way from inference to execution so that anything happens on chain regarding AI compute, you can just make we we can just make it verifiable and also private. And right now, I think maybe the one one of the biggest differentiation, among other, great folks here is that Aizen is using t based solution mainly at the current stage, for, verifiable and also private compute o
Thanks. And the the last guest is Travis.
Hi. I'm Travis Good. So, we're building Ambient. Ambient is, a proof of work based l one that is designed to deliver verified inference on one huge model starting with, llama 3,140,000,000. We have a unique approach, to security that's incredibly efficient. Happy to talk about that later on, but that's the gist. Just
to set the context, could one of you just give a definition of what we mean by verifiable inference? Sure.
Keep exploring with a free account.
A free account preserves the short preview. Upgrade to Pro for the full transcript and episode chat.