Today, Ritual and Celestia are excited to announce that users of Ritual will now be able tap into Celestia’s high-throughput data availability (DA) layer for both Ritual’s Chain and Infernet.
Modularity Meets Ritual’s AI Sovereignty
Ritual is building the first decentralized execution layer for AI. The first phase of Ritual is Infernet, a decentralized oracle network (DON) that allows smart contracts across any blockchain to natively access AI models for a variety of on-chain use cases and tasks. Over the next few months, we will be rolling out the second phase of Ritual: a sovereign chain built with a custom VM optimized for AI-native operations across models. You can read more on our product page here.
Celestia has been a trailblazer in the blockchain space, advocating for a modular architecture that separates consensus from execution to massively increase blockchain scalability while ensuring verifiability for any user. Developers can securely deploy their own high-throughput blockspace in any ecosystem using Blobstream, enabled by a zk attestation over Celestia’s data root. Celestia natively enables a new kind of paradigm, allowing developers to deploy self-governing, user-verifiable blockspace the way they would a smart contract.
This design philosophy aligns seamlessly with Ritual’s mission to bring AI on-chain with computational integrity, privacy, censorship-resistance, and decentralized incentives, with the opportunity to constantly improve via a modular architecture. We aim to foster a decentralized ecosystem where AI and crypto not only coexist but thrive together, helping usher in an entirely new wave of companies and protocols.
By combining forces, we aim to create a symbiotic ecosystem where AI and blockchain not only coexist but thrive together, fostering the development of entirely new companies and protocols.
Modularity for the Ritual Network
There are several ways in which Ritual’s users can take advantage of Celestia’s DA layer. Below, we outline a few.
Ritual Chain: The Sovereign AI Execution Layer
The Ritual Chain will be a sovereign execution layer powered by a VM custom-built to support AI-native operations. Model operations like inference and fine-tuning, are enshrined as opcodes and precompiles on the chain, with nodes that are running execution clients providing the models and compute necessary to serve these operations.
Celestia plugs into the Ritual Chain in several ways. From providing a DA option for the chain itself to publishing integrity proofs, Celestia offers users of Ritual best-in-class security and availability guarantees around their data, model workflows, and transactions.
Infernet
Ritual’s Infernet is the first product of its kind to enable users to access AI models both on and off-chain. Users can query models and return their outputs via on-chain smart contracts, do the same via off-chain REST APIs, or merge the flows and have part of sequence on-chain and part off-chain. Nodes with access to models and the compute to serve model operations (inference, finetuning, etc.) can permissionlessly join the network and begin fulfilling users requests. Infernet provides unmatched flexibility for coupling model workflows and on-chain workflows and enables a variety of novel use cases.
Users of Ritual can choose to use Celestia as a high-throughput DA layer for relevant transaction data around model workflows. Data created as a result of Ritual workflows, such as transactions posting model inputs/outputs or attached proofs, can be uploaded as blobs to Celestia for availability guarantees. This unlocks new possibilities and paradigms for developers. For example, being able to attest to the number of times a user interacted with a model or the quality of their prompts into a model on the Ritual network using Celestia can be used to create new reward and incentive mechanics for applications.
Computational Integrity Proofs
An important feature of both the Ritual Chain and Infernet is the ability to include proofs of computational integrity when desired. Different AI use cases necessitate different levels of integrity on the behalf of users and dApps, and Ritual’s infrastructure is built to be flexible in the integration of various proof systems.
The Ritual Chain and Infernet support ZK, OP, TEE, and other proof systems that nodes may choose to integrate. These techniques generate validity proofs, fraud proofs, attestations and other relevant pieces of information that need to be published somewhere where users have strong guarantees that they won’t be tampered with and can be accessed for verification. Celestia is a perfect fit for this, enabling users of Ritual to publish their proof data to Celestia’s DA layer and inherit the security guarantees Celestia provides to ensure proofs are always available for verification.
New Frontiers
Ritual’s integration with Celestia is just the beginning. We are excited to see more AI applications building on top of Ritual and Celestia, using our shared infrastructure. We’re excited to work with the community on ideas around how to best leverage this shared infrastructure and build the next generation of AI dApps.
For more updates on this front, follow Ritual and Celestia on Twitter. To learn how to develop an application or run a node on Ritual, check out our documentation here.
Disclaimer: This post is for general information purposes only. It does not constitute investment advice or a recommendation, offer or solicitation to buy or sell any investment and should not be used in the evaluation of the merits of making any investment decision. It should not be relied upon for accounting, legal or tax advice or investment recommendations. The information in this post should not be construed as a promise or guarantee in connection with the release or development of any future products, services or digital assets. This post reflects the current opinions of the authors and is not made on behalf of Ritual or its affiliates and does not necessarily reflect the opinions of Ritual, its affiliates or individuals associated with Ritual. All information in this post is provided without any representation or warranty of any kind. The opinions reflected herein are subject to change without being updated.