In the midst of the DeFi + AI trend, a quick look at the Panorama of the Four Major DeFAI Areas

Original Title: DeFai = DeFi + Ai

Original Author: Poopman, Cryptocurrency Researcher

Original Translation: DeepTech TechFlow

When traditional DeFi meets emerging AI, what kind of spark will be ignited? 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 am about to release 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 Focus on DeFAI?

The combination of Artificial Intelligence (AI) and blockchain is not a new concept. From early decentralized model training in the Bittensor subnet to decentralized GPU and compute resource markets like Akash and io.net, to the current 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-level application scenarios.

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

Starting from December 25, 2024, with frameworks and platforms like Virtual and ai16z witnessing the return of "US funds" after the Christmas holidays, the DeFAI industry has begun 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 ecosystem of DeFAI, 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

AI agents are programs that represent users to perform tasks according to a specific workflow. At the core of these agents is support provided by Large Language Models (LLMs), capable of generating responses based on their training data.

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

For example:

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

· Optimizing liquidity mining strategies: Providing users with higher returns

· Automating trade execution: Buying or selling assets based on market analysis (whether third-party or proprietary models)

Referring to @threesigmaxyz's research, AI models typically follow the following six core workflows:

· Data Collection

· Model Inference

· Decision Making

· Custodianship & Operation

· Interoperability

· Wallet Management

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

Exploring the World of DeFAI v2

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

· Abstracted/User-Friendly AI

· Yield Optimization and Portfolio Management

· DeFAI Infrastructure or Platform

· Market Analysis and Prediction

Abstracted 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 specialized knowledge.

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

· Fetch 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 rapid token sniping.

Its main features include:

· Transaction operations can be completed with natural language input.

· Launch Token projects, mint NFTs, and support airdrops to selected addresses through Pumpfun.

· Multi-agent collaboration features.

· Agents can autonomously post tweets on behalf of users.

· Snipe new meme coins listed on Pumpfun based on specific keywords or conditions.

· Automated Staking and DeFi strategy execution.

· Task scheduling, allowing users to customize personalized agents by inputting mnemonic data.

· Fetch data from multiple platforms for market analysis, such as identifying the primary holder of a particular token.

Wallet Functionality:

Upon creating an account, a wallet is generated by Privy within the system. Users can authorize their accounts to the agents, which will autonomously execute transactions and manage the portfolio. To enhance security, the private key is split and stored using Shamir's secret sharing, ensuring that Griffain and Privy cannot independently control the wallet.

Anon

@HeyAnonai, developed by renowned developer @danielesesta, who created DeFi protocols Wonderland and MIM. Anon aims to simplify the DeFi interaction experience, making it easy for both beginners and experienced users to get started.

Main features include:

· Cross-chain asset bridging based on LayerZero

· Real-time price feed and data updates via Pyth

· Automation operations and triggers based on time and Gas price

· 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

Additionally, Anon has recently released two significant updates:

· Automation Framework

· Smart body 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, and founded by @slate_ceo, is positioned as "Alpha AI," capable of autonomous trading based on on-chain data signals. Currently, Slate is the only abstract AI tool capable of achieving transaction automation on the @hyperliquidX platform.

One notable aspect is their fee structure.

Within Slate's services, fees are mainly divided into two categories:

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

2. Conditional Operations: If a user sets up a conditional order (e.g., limit order), Slate charges fees based on different condition types:

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

Charge a 1.00% fee for all other conditional actions.

In addition to Slate, there are many other 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 Deep Tide 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 create more efficient yield optimization and portfolio management strategies.

T3AI

@trustInWeb3 is a lending protocol that supports undercollateralized loans, using AI as an intermediary and risk management engine.

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

Kudai

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

Here is Kudai's development 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 GMX's GM Pool to further increase earnings

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

Sturdy Finance V2

@SturdyFinance is a protocol that combines lending with yield aggregation functionality. It dynamically allocates funds between different whitelist isolated pools using an AI model trained by Bittensor SN10 subnet miners to optimize earnings.

Sturdy's architecture is divided into two layers: Isolated Pools and Aggregation Layer.

1. Isolated Pools: These are single-asset pools where users can only lend one asset or borrow against one collateral type, 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. The Bittensor subnet provides the aggregation layer with the optimal allocation strategy. When users lend their assets to the aggregation layer, their risk is limited to the chosen collateral type, avoiding risks from other lending pools or collateral assets.

Representative projects in other earnings optimization and investment management areas 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 14.76% of market attention, making it one of the most influential agents in the ecosystem.

AIXBT's functionality extends beyond providing market insights; it is also interactive, capable of answering user queries and even issuing tokens via the Twitter platform. For example, the $CHAOS token was created through 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 deeper into the following topics on the Medium platform:

· How to create models

· Provision of data and computing resources

· Role of validation mechanisms

· Functioning of Trusted Execution Environments (TEEs)

Due to the extensive content, please refer to the articles 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

Frameworks

· @arcdotfun

· @ai16zdao

Platforms / Integrated Solutions

· @virtuals_io

· @aisweatshop

· @Almanak__

· @autonolas

· @Cod3xOrg

· @crestalnetwork

· @CreatorBid

· @openservai

· @WaveformBackup

· @getaxal

· @EmpyrealSDK

General Infrastructure

· @joinFXN

· @TheoriqAI

· @hyperbolic_labs

· @BagelOpenAI

· @Hive_Intel

Toolkits

· @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.

Phase One: Focused on improving efficiency and introducing tools that simplify complex DeFi operations. For example:

· AI capable of understanding imperfect inputs

· Tools for swift transaction completion

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

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

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

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

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

· Many current tools are merely a superficial ChatGPT wrapper, lacking clear evaluation standards.

· The fragmented nature of on-chain data may lead AI models to lean towards centralization rather than decentralization, with no clear solution at present.

Original Article Link