Interpreting Talus: Polychain leads $6 million financing for L1 designed specifically for AI Agent.

Talus allows the local design and deployment of Decentralization on-chain smart agents, seamlessly, trustlessly, and interoperably utilizing on-chain and off-chain resources and services.

Author: Shenchao TechFlow

The trend of AI Agent is still continuing.

On Base and Solana, there have been many protocols and MEME related to AI Agent, stirring up market funds and attention.

However, the current AI Agent protocols that have appeared are mostly focused on the Application Layer, and are generally opening up their own AI tracks within existing public chain ecosystems;

But large infrastructure projects have always been a higher valuation narrative in the encryption world (whether the market is buying is another matter). Specifically, creating a chain for AI Agents to run on, will the narrative ceiling be higher?

Or in other words, will catching a falling knife in the market’s non-dumb buying VC coin discourse become a lifesaving straw for certain infrastructure projects if they can catch the heat of AI Agent?

When you are still in doubt, someone has already taken action.

VC influx into the battlefield, AI proxy recognition project

On November 26th, Talus Network, an L1 blockchain designed for AI Agent, announced that it has raised $6 million in funding led by Polychain, with participation from Foresight Ventures, Animoca, Geek Cartel, Echo, etc.

At the same time, a group of angel investors, such as Polygon co-founder Sandeep Nailwal, Sentient core contributors and Symbolic Capital co-founder Kenzi Wang, 0G Labs CEO Michael Heinrich, Allora Labs CEO Nick Emmons, and Nuffle Labs co-founder Atlan Tutar, also participated.

As early as February this year, when AI agents’ narratives were not so strong, the project had already completed a $3 million Series A financing, with Polychain Capital leading the investment, and dao5, Hash3, TRGC, WAGMI Ventures, and Inception Capital participating.

Talus’ total funding has now reached 9 million USD.

Interestingly, the ‘L1 designed specifically for AI Agents’ has caught the eye of another AI Agent.

Recently, AI proxy @aixbt_agent, which has quickly gained popularity on Base, also captured Talus Network. Aixbt is an AI proxy that monitors encryption Twitter hotspots and analyzes and judges events happening in the industry.

Aixbt believes that Talus can build AI agents that operate entirely on-chain and claims to be following this trend.

This wave of cargo undoubtedly also increased the popularity and discussion of Talus in Reverse. In an environment where AI Meme is everywhere, a serious infrastructure project has instead attracted more follow.

Design L1 for AI Agent, do system-level optimization

So, how does Talus Network design an L1 specifically for AI Agents?

Before discussing this matter, there is a more critical narrative question - why does the AI Agent need a dedicated chain?

The existing AI ecosystem is faced with three major pain points: unclear ownership, insufficient transparency, and lack of permissionlessness.

Specifically, in the current centralized AI system, control over resources is concentrated in the hands of a few entities, and users lack the right to speak for their own data and Computing Power; the AI decision-making process is often a black box operation, lacking audit and verification mechanisms; users also find it difficult to customize and adjust AI services according to their own needs.

Although some platforms in different ecosystems, such as Virutals and vvaifu, allow users to create AI agents themselves, more efforts are focused on the direction of permissionless, followed by the tokenization of AI agents, and sharing asset income through holding tokens.

The ownership of this AI and whether it is a real AI still need some infrastructure to answer.

So, a blockchain specifically designed for AI Agents, the classical blockchain can solve problems through the following paths:

Ledger — clear record and transaction of resource ownership

Contract — Transparency and verifiability of the decision-making process

Cryptography — Permissionless Open Ecosystem

When it comes to projects, Talus allows local design and deployment of on-chain smart agents for Decentralization, seamlessly and trustlessly utilizing on-chain and off-chain resources and services in an interoperable manner.

It establishes a protocol that allows for the representation, utilization, and transaction of these agents, resources, and services in a permissionless and verifiable manner.

Breaking down the design of Talus, the combination of technical components at the following 4 levels is worth following:

Infrastructure Layer: the combination of Cosmos SDK and CometBFT

The Cosmos SDK is already mature and reliable, but more importantly, its modular features allow the entire blockchain system to be flexible and expandable like building blocks. This flexibility is particularly important as AI technology iterates rapidly.

Contract layer: Move language makes design elegant

The native object model of the Move language makes on-chain management of AI resources natural and elegant. For example, an AI model can be directly represented as an object in Move, with clear ownership and lifecycle, which is much simpler than the traditional account-based model of the blockchain. And the concurrency capabilities of MoveVM can support hundreds or thousands of AI agents running simultaneously, which is unimaginable in a traditional serial execution environment.

Resource Mapping Layer: Mirror Objects System

This system cleverly solves the problem of how AI resources are represented and traded on-chain. When you need to use a large language model, it is not possible to put the entire model on-chain.

In layman’s terms, you can think of Mirror Objects as the ‘digital avatars’ of these off-chain resources, through which on-chain Smart Contracts can trustfully operate off-chain AI resources.

Specifically, the Model Object is responsible for the on-chain representation of AI models. It not only records the metadata of the model but also includes the access permissions and usage conditions of the model. The Data Object manages the access control of the dataset to ensure the privacy and security of the data when used by AI models. The Computation Object tokenizes Computing Power resources, allowing Computing Power to be freely traded on-chain like a cryptocurrency.

Validation Layer: Multi-layer validation scheme

For ordinary AI Agent interactions, such as chatbot conversations, lightweight Digital Signatures can be used to ensure the authenticity of the responses.

In high-risk scenarios such as financial decision-making, Zero-Knowledge Proof can be used to ensure the correctness of the decision-making process without revealing specific details.

For scenarios that require quick response but can tolerate latency verification, such as AI NPC behavior in games, an optimistic fraud proof mechanism can be used to ensure both performance and eventual correctness.

For more technical details, you can refer to our previous article: “Interpreting Talus White Paper: AI Proxy Center of Decentralization”.

Infrastructure comes first, AI dating application comes second.

Currently, Talus itself is still in the Testnet stage and it will take time before it goes live on Mainnet.

From the perspective of project operation and attracting user attention, while holding back the big moves of infrastructure construction, releasing some applications in phases as pilot projects can make the market see the usability of this L1 and increase confidence.

Just as the financing news was announced, Talus also announced the first application in its ecosystem, “AI Bae”. The word “Bae” in this name comes from the internet slang “Before Anyone Else”, implying the social nature of this application.

Interestingly, Talus chooses to position its first application as an AI dating game rather than a more serious financial or business application, making the intention to attract more ordinary users clear.

From the information currently revealed, AI Bae will support users to create and customize their own AI companions, and introduce a Polymarket-style betting mechanism. This design is quite creative: it not only allows users to interact with AI, but also tokenizes their AI companions to create exclusive memecoins. In other words, your ‘digital boyfriend/girlfriend’ can not only chat with you, but also become an asset with market value.

This practice of combining social, gaming, and financial elements is not uncommon in the encryption market. The new public chains of this generation want to break through, and perhaps using popular gameplay in a reasonable way is also an effective way to break the deadlock.

Currently, AI Bae has opened Allowlist registration. In the crypto market, which is generally pessimistic about new public chains and infrastructure, Talus’s unconventional approach may bring some unexpected surprises to the project. After all, in a bull run, sometimes an interesting application can be more effective than just talking about technical advantages.

Gameplay, old taste

In addition to the dating app mentioned above, Talus has also launched a gamified task activity called “Enchanted Seasons”. The first season event is named “The Awakened Orb”, which runs from November 11th to January 11th next year.

This design is quite “gaming” in nature: daily tasks (Daily Rituals), weekly tasks (Weekly Quests), team challenges (Team Challenges) - from the perspective of tasks, it is indeed a common operation method for Web3 projects. Currently, users can participate in tasks such as binding social media and posting to earn points, which is also a common old trick used by previous projects.

However, in the current market environment, users’ enthusiasm for pure task systems is not as strong as before. How to further design more differentiated tasks, or more clearly define the economic value that points can bring, may become the key to breaking through.

Even the L1 designed specifically for AI Agent still cannot do without the traditional community incentive model in the early stages of operation.

In the encryption world, even the most advanced technology needs to start from the user’s psychology and behavioral habits, play with narrative and assets, in order to succeed.