Table of Contents
About Scout Article: The Scout Article is a series by Four Pillars that aims to identify promising early-stage blockchain projects through active research and monitoring of emerging sectors. Four Pillars' researchers continuously scout different fields to uncover new and innovative blockchain projects that have not yet received widespread coverage from other firms. Projects featured in the Scout Report are expected to demonstrate novel ideas or narratives. Each month, selected projects identified through an internal voting process among researchers will be included in that month's official Scout Report publication. This provides paid subscribers with early access to blockchain opportunities with future potential.
1. Gevulot
"Gevulot is a general L1 network specialized in generating and verifying for Zero-Knowledge Proofs (ZKP), innovatively combining the strengths of previous attempts to decentralize ZKP generation (e.g., Proof Market, Prover Network) and presenting a new paradigm in the ZKP proving market." - by 100y
1.1 Current Status of the ZKP Mining Market
In the blockchain industry, the influence of Zero-Knowledge (ZK) technology is becoming increasingly important. Starting with ZK rollups, which aim to improve the scalability of the base layer through ZK technology, recent developments include zkVM, zkBridge, and zkML in various sectors. In applications that adopt ZK technology, the proving process for generating ZKP is essential.
Unfortunately, most ZK applications currently rely on a single entity for proving. Major ZK rollups like zkSync Era, Polygon zkEVM, Linea, and Starknet all have a single entity performing the generation of ZKP. Although there are no risks related to validity or funds being stolen because execution can be proven through ZKP, there is a liveness risk that the system could stop due to a single point of failure. Therefore, there are increasing attempts to decentralize the prover role, mainly in two ways: 1) Proof Market and 2) Prover Network.
Proof Market is an open market where users who need ZKP generation place orders, and provers capable of generating ZKP are matched with these orders. This allows easy outsourcing of ZKP generation, and Proof Market can handle various applications, but it does not guarantee liveness. Examples of Proof Market include =nil; Foundation and marlin.
A Prover Network is a method where applications build their own decentralized prover network. Provers in the Prover Network generate ZKP only for specific applications, which is efficient and has no liveness issues. However, building a Prover Network is challenging and cannot handle multiple applications. Examples of applications attempting to build a Prover Network include Taiko, and Aleo.
1.2 Gevulot, suggest a middle ground
Source: Gevulot
Gevulot is a product built by the Equilibrium Group, a permissionless Layer 1 blockchain for ZKP systems. Unlike previous attempts limited to Proof Market and Proving Network, Gevulot resolves these issues by introducing a specialized L1 blockchain for ZKP systems. Unlike Proof Market and Proving Network, Gevulot has an incentive system, and being a universal blockchain system, it guarantees high liveness and can intervene in ZKP generation for various applications. This represents a compromise of previous attempts to decentralize ZKP generation.
Programs that can be deployed on Gevulot include 1) Prover and 2) Verifier. Users and applications that need ZKP generation send a 'Run' transaction containing the desired Prover program ID to Gevulot, and the Prover in the Gevulot network generates ZKP using that Prover program. The generated ZKP is then verified by a Validator using the Verifier program. If more than 2/3 of the Validator nodes participate in the verification, the ZKP is included in the next block.
When a ZKP generation request comes in, the task is assigned to a small group of random Provers through VRF. Since it's not assigned to just one Prover, it solves the liveness issue, and by not assigning it to too many Provers, it reduces the waste in computation related to ZKP generation.
1.3 Gevulot Use Cases
Due to its ability to easily prove complex computational processes, ZK technology is increasingly used in the blockchain field. All applications using ZK need ZKP generation, which means various ZK applications can utilize Gevulot.
Source: Gevulot
The most notable example is ZK rollups. ZK rollups perform computations off-chain to solve the scalability of L1 and prove the validity of these executions on L1 through ZKP. Sequencers in ZK rollups can gather transactions submitted by users, determine their order, and submit this data to Prover programs deployed on Gevulot. A small group of Provers determined by VRF generates ZKP for this. In ZK rollups, generated ZKP existing in the Gevulot mempool can be verified internally for finality, or they can wait for Gevulot's Validators to verify the ZKP before granting finality. Other applications like zkVM, zkBridge, and zkML that require ZKP generation can also outsource to Gevulot.
1.4 Future: Four Pillars’ Thoughts
1.4.1 xpara
Looking back a year ago, the implementation of zk (zero-knowledge) was considered something far away. Now, there are many zkEVM, zkVM, and zkBridge protocols being launched in production and used by numerous users. Due to rapid development and ongoing progress, the prover (the entity that generates zk proof) has various specifications, and its incentive system is not yet well-developed.
This is where a zk marketplace comes into play. It serves as an open marketplace for provers to generate zk proof for other parties. Similar to how the Keep3r network functions as a marketplace for bot jobs, projects like Gevulot are positioned to be the marketplace for provers, enabling these tasks to be performed in a more trustless manner.
There are other projects, such as =nil; Foundation, that are also building this zk marketplace. However, Gevulot stands out by having an architecture that supports diverse zk proof technologies compared to other players. Additionally, the developers of the Gevulot team have extensive experience as they come from Equilibrium Labs, which has contributed to the development of various zk projects like Starknet and Aleo.
2. Other Notable Projects
2.1 Ritual
"Ritual is an on-chain AI inference layer that provides verifiable AI model estimation and modularity for integration with other layers." by pillarbear
In the past year, AI has experienced a Cambrian explosion with a rapid increase in new models and applications. However, a few corporations monopolize current AI models and applications, and their learning or inference processes operate in an unverifiable manner. The key issues identified in the operation of AI are as follows:
Lack of Service Standards and Verification: Platforms developing AI-related models currently do not offer any guarantees regarding the accuracy of model estimations, the protection of personal information during the model's learning process, or resistance to censorship.
High Computing Costs and Limited Accessibility: AI hardware is becoming increasingly difficult to obtain, and the costs for hardware and computation are rising to excessive levels for developers. The current AI infrastructure is operated by a few corporations, limiting developers and users from freely integrating and offering their services.
Monopoly and Structural Mismatch: Corporations seek to maintain their models privately, suppressing innovation and strengthening monopolies. Even when models are open-sourced, appropriate infrastructure is lacking to reward contributions. Additionally, users have little or no say in the governance and ownership of AI.
Ritual has emerged to address these issues. It is an open, modular, sovereign computational layer for AI, offering a decentralized network that provides accessibility to AI operations and models. This allows users to access any model on the network via API, assuring accuracy and privacy in AI model inference.
The first execution layer of Ritual, 'Infernet', provides accessibility and an interface for AI model inference to applications on the chain. Infernet allows anyone to build models on Ritual, ensuring permissionless accessibility for model computation. Ritual has announced plans to develop Infernet into a modular layer interoperable with other blockchain layers, allowing different protocols to freely use it as an AI Co-processor regardless of the chain they use.
Ritual has recently completed a successful $25 million fundraising round led by Archetype. Key investors included Accomplice, Robot Ventures, and dao5, with angel investors like Balaji Srinivasan, Nicola Greco, and Keone Hon of Monad participating.
2.2 Nocturne
“Nocturne is a blessing of privacy, completely wrapping the journey of a blockchain transaction.” by Jay
Transparency and openness are some of the key attributes of blockchain that contribute to its ability to achieve trust, but ironically, they have also been major barriers to its adoption. This is because, in practice, if every transaction (i.e., financial activity) between individuals (or entities especially at the corporate level) is traceable, the reason to use blockchain becomes quite diluted.
In order to simultaneously ensure these double-edged swords and privacy, the blockchain ecosystem strives to introduce various solutions such as zero-knowledge proofs, homomorphic encryption, multiparty computation, and trusted execution environments (TEEs). However, due to various limitations such as the increasing complexity of the network, imperfections in the technology, high gas fees, bootstrapping of the ecosystem, etc., it is unlikely that users would be able to enjoy full privacy in the experience of many existing applications at the present time.
In this context, stealth address is currently gaining traction for its ability to be applied fairly quickly to a wide variety of services (especially within Ethereum's rich ecosystem) while ensuring a high level of privacy - a stealth address is, simply put, a one-time, temporary address created to ensure the privacy of a transaction.
Several ideas for implementations of stealth addresses have been proposed (e.g., EIP-5564, An incomplete guide to stealth addresses, etc.), but Nocturne adopts the most flexible and efficient structure among them. Notably, it incorporates the concepts of zero-knowledge proofs and shielded pools into its architecture, allowing it to simultaneously provide UX simplicity, compatibility, and enough privacy to completely hide the link between the user and the stealth address.
With so many different privacy solutions evolving, it's hard to say which one will ultimately win. One thing is for sure, though: Nocturne is raising the bar on the level of interactivity of the underlying accounts on the blockchain.
2.3 Superstate
"Trillions of RWA is coming on-chain, starting with the US Government Bond.” by xpara
SuperState is building a tokenized fund that aims to provide on-chain investors get exposure to US Treasury bill yields while addressing concerns around custody and shutdown risks. The fund seeks to bring traditionally off-chain assets like Treasury bills onto the blockchain to leverage the composability of decentralized finance (DeFi) protocols.
A draft prospectus has been filed to create the Superstate Short-Term Government Bond Fund, an open-ended mutual fund with a secondary record of ownership tracked on the Ethereum blockchain. This will allow the fund to be natively integrated into DeFi applications for enhanced access and functionality.
The SuperState team includes members from Compound Labs, most notably its CEO who co-founded Compound. As an early pioneer of lending protocols, Compound has previously deployed their on-chain “Compound Treasury” to off-chain assets like Treasuries. This provides the SuperState team with extensive experience in building products that utilize DeFi infrastructure to generate returns from real-world assets.
I believe the strength of DeFi lies in offering programmable and composable finance through shared blockchain networks. As security improves and liquidity increases, traditional assets like Treasuries and corporate bonds can potentially leverage the growing DeFi ecosystem. SuperState aims to bridge real-world assets and blockchain-based financial applications to create new opportunities for investors.
2.4 Jokerace
"Jokerace is a great tool for communities to make, run, and reward decisions." by moyed
If I had to describe Jokerace in one word, I'd say it's an "on-chain contest" service. With Jokerace, communities can easily create, run, and reward contests on-chain. This allows communities and protocols to incentivize users to take certain actions. For example, with Jokerace, you can do the following.
Create a contest for a feature we'd like to see in our project to gauge interest in advance.
Create a "create content related to our project" contest with a prize for the first-place winner to encourage secondary creation.
A "try our service and leave feedback" contest is a great way to get quality feedback from early users.
If you're a tokenized community, you can limit the contest to users with a certain amount of tokens, or if you don't have tokens, you can customize it with different criteria, such as following a certain account or completing a Zealy Quest. In other words, you can decide who can participate in the contest, who will judge it, how to set the rewards, and it all happens on-chain. They recently raised a seed round from 1kxnetwork and a bunch of angels, and they've been getting a lot of attention lately, with Eigenlayer planning their next LST to be onboarded through Jokerace.
Thanks to Kate for designing the graphics for this article.
