The Delphi Podcast

Verifiable Inference: Don't Trust, Verify | Crypto x AI Event

Oct 15, 2024

Join Colin Gagich, co-founder of Inference Labs, Ryan McNutt from Sphere 1, and Travis Good of Ambient, as they dive into the crucial concept of verifiable inference. They discuss how decentralized solutions can combat AI's centralized control and explore real-world applications across sectors. The conversation highlights innovations like zero-knowledge proofs and trusted execution environments, emphasizing their role in enhancing security, transparency, and privacy in AI and blockchain integration. It's a critical look at the future of trustless technology.

01:06:54

Creator website

Episode guests

Colin Gagich

Ryan McNutt

Travis Good

AI Summary

AI Chapters

Episode notes

Podcast summary created with Snipd AI

Quick takeaways

Verifiable inference enables independent validation of AI outputs, essential for ensuring the integrity of results without revealing internal processes.

Decentralized AI frameworks can enhance verification methods, notably in biometric authentication, ensuring privacy and security in identity verification.

The discussion emphasizes the importance of diverse verification methodologies like trusted execution environments and ZK proofs to balance security and user trust.

Deep dives

The Role of Verifiable Inference

Verifiable inference is essential in confirming the integrity of AI outputs without needing detailed insight into the inference process. This approach allows users to independently validate results by only examining the input and output, minimizing reliance on trust in the model's operations. The discussion emphasizes that adopting a trustless framework is crucial, especially in scenarios involving significant data and financial transactions. Thus, verifiable inference serves as a cornerstone for creating reliable AI applications in decentralized ecosystems.

Intro

2min

Building Trust with Verifiable Inference

15min

Verifying AI with Trusted Execution Environments

7min

Understanding Zero-Knowledge Proofs in AI

23min

Understanding Model Support in Secure Computing Environments

2min

Challenges and Opportunities in Decentralized AI Inference

15min

The Intersection of AI and Blockchain: Opportunities and Challenges

2min

In this engaging debate for Crypto x AI Month, join Luke Saunders as he moderates a conversation on Verifiable Inference—a critical technology that ensures trustless AI by verifying the correctness of outputs without revealing internal workings. He is joined by leading founders from the crypto AI space, including Colin Gagich of Inference Labs, Ryan McNutt of SphereOne, Jeremy from Aizel Network, and Travis Good of Ambient, to explore:

► What verifiable inference is and why it's essential in the age of AI

► How decentralized models can offer solutions to centralized AI control

► Real-world use cases of verifiable inference across blockchain and AI applications

This panel digs into the technical approaches—from ZK proofs to trusted execution environments (TEEs)—and discusses how the future of AI and crypto requires trustless verification to ensure security, transparency, and privacy.

Watch more sessions from Crypto x AI Month here: https://delphidigital.io/crypto-ai --- Crypto x AI Month is the largest virtual event dedicated to the intersection of crypto and AI, featuring 40+ top builders, investors, and practitioners. Over the course of three weeks, this event brings together panels, debates, and discussions with the brightest minds in the space, presented by Delphi Digital.

Crypto x AI Month is free and open to everyone thanks to the support from our sponsors:

https://olas.network/

https://venice.ai/

https://near.org/

https://mira.foundation/

https://www.theoriq.ai/ --- Follow the Speakers: - Luke Saunders on Twitter/X ► https://x.com/lukedelphi - Travis Good on Twitter/X ► https://x.com/IridiumEagle - Jeremy on Twitter/X ► https://x.com/immorriv - Colin Gagich on Twitter/X ► https://x.com/colingagich - Travis Good on Twitter/X ► https://x.com/ryanmcnutty33 --- Chapters 00:00 Introduction to Verifiable Inference 03:46 Defining Verifiable Inference 05:13 Use Cases for Verifiable Inference 10:01 Real-World Applications and Innovations 16:38 Different Approaches to Verification 24:18 Exploring Zero-Knowledge Proofs 28:08 Proof of Logics and Its Implications 34:02 Multi-Agent Systems and Transaction Verification 37:46 Challenges in Optimistic Approaches 41:36 Determinism and Model Reproducibility 45:30 The Balance of Open and Closed Source Models 50:25 The Future of Edge Computing and Inference 57:49 Decentralization and Government Control of AI Disclaimer All statements and/or opinions expressed in this interview are the personal opinions and responsibility of the respective guests, who may personally hold material positions in companies or assets mentioned or discussed. The content does not necessarily reflect the opinion of Delphi Citadel Partners, LLC or its affiliates (collectively, “Delphi Ventures”), which makes no representations or warranties of any kind in connection with the contained subject matter. Delphi Ventures may hold investments in assets or protocols mentioned or discussed in this interview. This content is provided for informational purposes only and should not be misconstrued for investment advice or as a recommendation to purchase or sell any token or to use any protocol.