Being in the midst of the DeFi + AI wave, here is a quick look at the DeFAI Landscape across four key areas

Original Article Title: DeFai = DeFi + Ai

Original Article Author: Poopman, Cryptocurrency Researcher

Original Article Translation: DeepTide TechFlow

When traditional DeFi meets emerging AI, what kind of sparks will fly? What new variants or technological innovations can we create?

Today, we will explore the early ecosystem of DeFAI (Decentralized Finance + AI) together.

Hopefully, this article will provide you with some inspiration!

(*I will soon be publishing a 20-page in-depth analysis article on Medium. Today's content is just a preview to give you a quick overview of this emerging field.)

Why Pay Attention to DeFAI?

The combination of Artificial Intelligence (AI) and blockchain is not new. From early decentralized model training in the Bittensor subnet to decentralized GPU and computing resource markets like Akash and io.net, and now the rise of AI and memecoin integration on Solana, each phase has demonstrated how blockchain can complement AI's capabilities through resource aggregation, driving the realization of sovereign AI and consumer-grade application scenarios.

According to CoinGecko data, as of January 13, 2025, the total market capitalization of DeFAI has reached approximately $1 billion. Among them, Griffain holds a 45% market share, while $ANON holds 22%.

Starting from December 25, 2024, as frameworks and platforms such as Virtual and ai16z saw the return of "US funds" after the Christmas holidays, the DeFAI industry began to accelerate.

This is just the beginning. The potential of DeFAI far exceeds its current performance.

Although current applications are still in the proof-of-concept stage, we should not underestimate the potential of transforming DeFi into a more intelligent, user-friendly, and efficient financial ecosystem through AI technology.

Before delving into the DeFAI ecosystem, we first need to understand the fundamental principles of how AI agents operate in the DeFi and blockchain environment.

Operation Mechanism of AI Agents in DeFi

An AI agent is a program that represents a user to perform tasks according to a specific workflow. These agents are primarily supported by Large Language Models (LLMs), capable of generating responses based on their training data.

In blockchain, agents can interact with smart contracts and accounts, handle complex tasks without the need for continuous user intervention.

For example:

· Simplifying the DeFi user experience: by one-click completing multi-step cross-chain bridging and liquidity mining operations

· Optimizing liquidity mining strategies: providing users with higher returns

· Automated trade execution: buying or selling assets based on market analysis (whether from a third party or their own model)

Referencing @threesigmaxyz's research, AI models typically follow the following 6 core workflows:

· Data collection

· Model inference

· Decision-making

· Custody and operation

· Interoperability

· Wallet management

Once you have "collected" the above 6 core elements, you can build your autonomous AI agent on the blockchain. These agents can play different roles in the DeFi ecosystem, thereby enhancing on-chain efficiency and user trading experience.

Exploring the World of DeFAI v2

Overall, I categorize the integration of DeFi with AI (DeFAI) into four main categories:

· Abstract/Human-friendly AI

· Yield Optimization and Portfolio Management

· DeFAI Infrastructure or Platforms

· Market Analysis and Prediction

Abstract AI or AI ChatGPT

In this field, an ideal AI solution should have the following capabilities:

· Automatically execute multi-step transactions and Staking operations without the user needing any professional knowledge.

· Conduct real-time market research and provide users with the necessary key information and data to help them make wise trading decisions.

· Retrieve data from multiple platforms, identify market opportunities, and provide users with comprehensive analysis.

Next, let's take a look at some popular tools in this field:

Griffain

@griffaindotcom is currently the first and top-performing abstract AI tool on the Solana blockchain, supporting various functions such as transaction execution, wallet management, NFT minting, and fast token sniping.

Its key features include:

· Completion of transaction operations with natural language input

· Launching Token projects via Pumpfun, minting NFTs, and supporting selecting addresses for airdrops

· Multi-agent collaboration capabilities

· Agents can tweet on behalf of users

· Sniping newly launched Meme coins on Pumpfun based on specific keywords or conditions

· Automated Staking and DeFi strategy execution

· Task scheduling allowing users to customize personalized agents by inputting memory data

· Retrieval of data from multiple platforms for market analysis, such as identifying the primary holders of a specific Token

Wallet Functionality:

When an account is created, the system generates a wallet through Privy. Users can authorize the account to the agent, and the agent will autonomously execute transactions and manage the portfolio. To enhance security, the private key is split and stored using Shamir's secret sharing technology, ensuring that Griffain and Privy cannot independently control the wallet.

Anon

@HeyAnonai, created by renowned developer @danielesesta, who previously developed DeFi protocols Wonderland and MIM. Anon's goal is to simplify the DeFi interaction experience, making it easy for both beginners and experienced users.

Key features include:

· Cross-chain asset bridging based on LayerZero

· Real-time price and data updates through Pyth

· Automation and triggers based on time and Gas prices

· Real-time market insights, such as sentiment analysis and social data analysis

· Support for lending operations in collaboration with protocols like Aave, Sparks, Sky, and Wagmi

· Natural language trading feature supporting multiple languages, including Chinese

Furthermore, Anon has recently released two significant updates:

· Automation Framework

· Intelligent Agent functionality focusing on Gemma research

These updates have positioned Anon as one of the most anticipated abstraction tools currently.

Slate (Not Yet Launched)

Slate, backed by BigBrain Holdings, with its founder @slate_ceo positioning it as "Alpha AI," is capable of autonomous trading based on on-chain data signals. Currently, Slate is the only abstraction AI tool that can achieve trade automation on the @hyperliquidX platform.

One notable aspect is their fee structure.

Within Slate's service, fees are primarily divided into two categories:

1. General Operations: For regular transfers or withdrawals, Slate does not charge any fees. However, for executing more complex operations such as Exchange, Bridge, Claim, Borrow, Lend, Repay, Stake, Unstake, Long, Short, Lock, and Unlock, the platform charges a 0.35% fee.

2. Conditional Operations: If users set up conditional orders (e.g., limit orders), Slate charges fees based on different condition types:

Charge a 0.25% fee for Gas-based conditional operations;

Charge a 1.00% fee for all other conditional operations.

In addition to Slate, there are many emerging abstraction AI tools in this field, here are some representative projects:

· @AIWayfinder

· @orbitcryptoai

· @dolion_ai

· @askthehive_ai

· @HeyElsaAI

· @Spectral_Labs

· @Infinit_Labs

· @ProjectPlutus_

· @bankrbot

And many more projects in development...

Here is a comparison table of multiple abstraction AI tools:

Figure: Compiled by DeepWave TechFlow

Automated Yield Optimization and Investment Management: Unlike traditional yield strategies, DeFi protocols in this field use AI to analyze on-chain data, identify trends, and provide insights to help teams devise more efficient yield optimization and portfolio management strategies.

T3AI

@trustInWeb3 is a lending protocol that supports under-collateralized loans, utilizing AI as an intermediary and risk management engine.

T3AI's AI agent can monitor the health of loans in real-time and ensure they remain repayable through its risk index framework. This is an interesting application of AI in DeFi.

Kudai

@Kudai_IO is an experimental AI agent focused on the GMX ecosystem, developed by GMX Blueberry Club using the EmpyrealSDK toolkit. Currently, the $KUDAI Token is trading on the Base network.

Here is Kudai's roadmap:

The core idea of Kudai is to use all transaction fees earned through $KUDAI to fund autonomous trading operations by the smart body and return the profits generated by these operations to Token holders.

In the upcoming second phase (out of four phases), Kudai will have the following features that users can trigger via natural language commands on Twitter:

· Purchase and stake $GMX to generate a new income stream

· Invest in the GM pool of GMX to further increase earnings

· Purchase GBC NFT at the floor price to diversify their portfolio

Sturdy Finance V2

@SturdyFinance is a protocol that combines lending with yield aggregation, dynamically allocating funds between different whitelist isolated pools using an AI model trained by the Bittensor SN10 subnet miners to optimize profits.

Sturdy's architecture is divided into two layers: isolated pools and aggregation layer.

1. Isolated Pools: These are single-asset pools where users can either lend one asset or borrow with one collateral, reducing the inter-asset risk.

2. Aggregation Layer: Built on Yearn V3, users' assets are allocated to whitelist isolated pools based on utilization and yield at the aggregation layer. The Bittensor subnet provides optimal allocation strategies for the aggregation layer. By lending assets to the aggregation layer, users only carry the risk of the chosen collateral type, avoiding risks from other lending pools or collateral assets.

Representative projects in the field of other yield optimization and investment management include:

· @derivexyz

· @Thales_ai

· @Mozaic_Fi

· @boltrade_ai

· @vainguard_ai

· @Ensofi_xyz

· @0xARMAgeddon

· @glamsystems

And many more projects currently in development...

Market Sentiment Analysis AI Agent

AIXBT

@AIXBT_agent is a market sentiment tracking AI agent that integrates and analyzes data from over 400 key opinion leaders (KOLs) on Twitter using its proprietary engine. AIXBT can capture market trends in real time and provide valuable insights to users around the clock.

Among all AI agents in the DeFi space, AIXBT holds a 14.76% market share, making it one of the most influential agents in the ecosystem.

AIXBT's functionality is not limited to providing market insights; it is also interactive, able to answer user queries and even issue tokens via the Twitter platform. For example, the $CHAOS token was created through a collaboration between AIXBT and another interactive bot, Simi, using the @EmpyrealSDK toolkit.

Other market analysis AI agents include:

· @tri_sigma_

· @ASYM41b07

· @kwantxbt

· @gemach_io

DeFi Infrastructure and Ecosystem Platform

The implementation of Web3 AI agents relies on decentralized infrastructure. These projects not only provide model training and inference services but also offer data, validation mechanisms, and coordination layers for AI agent development.

Whether in Web2 or Web3, models, computing power, and data are always the three core pillars driving the development of Large Language Models (LLMs) and AI agents.

We delve into the following topics on the Medium platform:

· How to create models

· Provision of data and computational resources

· Role of validation mechanisms

· Working principle of Trusted Execution Environments (TEE)

Due to the extensive content, please refer to the article on Medium for specific details.

Here is a DeFi infrastructure ecosystem map created by @pinkbrains_io:

The key players in this field include:

Trusted Execution Environment (TEE)

· @PhalaNetwork

· @MarlinProtocol

· @AutomataNetwork

Framework

· @arcdotfun

· @ai16zdao

Platform / Integrated Solution

· @virtuals_io

· @aisweatshop

· @Almanak__

· @autonolas

· @Cod3xOrg

· @crestalnetwork

· @CreatorBid

· @openservai

· @WaveformBackup

· @getaxal

· @EmpyrealSDK

General Infrastructure

· @joinFXN

· @TheoriqAI

· @hyperbolic_labs

· @BagelOpenAI

· @Hive_Intel

Toolkit

· @sendaifun

· @lexiconinfra

The Future Development of DeFi AI

I believe the DeFi market will go through three main stages: first pursuing efficiency, then achieving decentralization, and finally focusing on privacy protection.

The development of DeFi AI will go through 4 specific stages.

Stage One: Focus on improving efficiency by introducing tools that simplify complex DeFi operations. For example:

· AI that can understand imperfect inputs

· Tools for fast transaction completion

· Real-time market research to help users make more informed decisions based on their goals

Phase Two: The AI agent will enable autonomous trading, able to execute strategies based on third-party data or insights from other AI agents. Advanced users can fine-tune models, building agents to optimize returns for themselves or clients.

Phase Three: Users will focus on wallet management and AI accountability. Trusted Execution Environments (TEEs) and Zero-Knowledge Proofs (ZKPs) will ensure transparency and security of the AI system.

Phase Four: Finally, a no-code DeFi AI toolkit or an AI-as-a-Service protocol may emerge, creating an agent-based economy where users can fine-tune models through cryptocurrency transactions.

While this vision is exciting, there are still some pressing issues to address:

· Many current tools are merely a simple wrapper around ChatGPT, lacking clear evaluation criteria.

· The fragmentation of on-chain data may lead AI models to be more centralized rather than decentralized, with no clear solutions at present.

Original Article Link