Agent-Fi on AO: A financial paradigm integrating AI agents

Imagine a future world where AI agent agents form a digital companionship/symbiotic relationship with humans. Autonomous Agents can clarify intentions in conversations, automatically dismantle tasks and perform tasks according to the natural language requirements put forward by users. Achieve expected results.

AO has established an asynchronous parallel network based on Actors. It does not agree on the entire calculation process of the contract, but only agrees on the transaction sequence. The optimistic default fixed transaction sequence will have consistent running results in the virtual machine. This choice allows for massive scaling of AO network computing to support any type of computing. The AR network is used as a consensus layer for transaction sequences and a storage layer for transaction result status.

Compared with other current mainstream blockchain projects, which are mostly single blockchains and only support native state machine smart contracts from the bottom layer, AO's infrastructure compatibility can support more complex computing capabilities, including AI models. of operation.

AO’s Compute Unit (Compute Unit) has been able to access 16GB of memory after the recent WASM virtual machine update, which means that we can download and execute 16GB models on AO. 16GB is more than enough to run large language model calculations, such as the Falcon series of Llama 3 unquantized versions and many other models.

At the same time, AO uses WeaveDrive, which allows users to access Arweave data in AO just like accessing a local hard disk, and is compatible with different types of virtual machines. Highly heterogeneous processes interact in a shared environment, which means we enjoy more data. sources and combination possibilities. This also means there is an increased incentive for users to upload data to Arweave when building applications in the future, as this data can also be used in AO programs. The AO development team has uploaded approximately $1,000 worth of model data to the Internet when testing large language models running in the AO+AR system, but this is just the beginning.

AO’s system design makes it possible to implement smart contracts integrating AI agents. By programming in AO, we create AI agents to make intelligent decisions in the market. Agents may fight against each other or represent humans against humans. "When we look at the global financial system, approximately 83% of Nasdaq's transactions are executed by robots." Current quantitative trading is the predecessor of AI agent trading, and in the future, machine learning models will be designed and selected to execute automated trading processes Will be more easily "unboxed" and automated by AI.

The development of DeFi in the past few years has made it possible to perform various financial operations on the chain without trusting centralized entities, such as lending, trading tokens or derivatives. But when we are really talking about the market, it is not just the reliability of these operations, in fact, the reliable execution of various operations is only the foundation. The core factor that determines whether a market is dynamic is still the flow of capital, the people who decide to buy, sell, lend or participate in various financial games. Today, if you want to participate in cryptocurrency investing without doing all the research and participation yourself, you must find a reliable fund, trust them to manage your funds and empower fund members to make intelligent decisions. But with the development of AO applications, we may be able to expand the intelligent decision-making part of the market, filter information in the network, process data, combine strategies, integrate the wisdom of AI agents to make real-time decisions in the network, and create a very rich decentralized Autonomous agent financial system.

There are already some projects starting to realize this vision. We will introduce Autonomous Finance (hereinafter referred to as AF), Dexi and Outcome. Among them, AF’s achievements are the most eye-catching.

Autonomous Finance

AF focuses on researching and developing financial applications combined with AI on AO. By building AI models and data-driven financial decisions on the AO chain, AF has made an attempt to put the intelligent decision-making layer on the chain. The main business has three parts, namely Core Infrastructure, Intelligent Agent Finance (AgentFi) and Content Finance (ContentFi).

Core facilities include protocols such as decentralized exchanges (DEX), lending, derivatives, and synthetic assets.

AgentFi mainly refers to the execution of trading strategies through composable semi-autonomous and fully autonomous agents. Unlike other autonomous agent frameworks that rely on off-chain programs for signal processing and logic processing, the autonomous agents provided by AF use on-chain data streams to self-learn and execute investment strategies based on various liquidity pools and financial foundations within the AO ecosystem. These agents can operate autonomously without the need for off-chain signals or human intervention.

Typical autonomous agents include:

• Dollar cost averaging (DCA) asset management agents

• Self-balancing autonomous index funds

• Autonomous hedge funds with customized risk strategies

• Income aggregation agents

• On-chain prediction agent

• High-frequency trading agent

The DCA agent serves as the basic agent and is often called when other more complex agents execute logic. Therefore, as a frequently used combinable agent module, there are many customizable parameters for users to adjust according to their own needs, such as within a specific price range. Triggered trading, fixed-interval trading length adjustments and weighted trading based on asset prices (e.g. buy more when prices are lower), as well as data-driven take-profit and profit reinvestment signals.

DCA agent application is built around two key AO processes:

• An agent process triggered by Cron (a time-based task management system, often used to trigger task execution at scheduled times): mainly responsible for user-initiated and automatic timing DCA transactions, record managed funds and timely update the back-end AO process

• Back-end AO process: manage agent applications related to the user name and track and record the historical transactions of each agent

The following figure illustrates DCA The design architecture and interactive components of the agent

For users who use the front end, the front end of the DCA agent is built on DEXI. Users can set up the DCA agent by connecting to the AO Connect wallet on the DEXI website. Among them, DEXI accesses information about the available AMM pool and obtains the latest prices, the DCA agent is responsible for executing specific transaction logic, and the backend AO process retrieves all agents related to the user.

Content Finance is a framework for attributing and monetizing data stored on the Arweave permanent network into composable assets for AO processes. AF is building applications that allow data contributors or content funds to contribute data such as historical and real-time market intelligence to permaweb. These contents will serve as on-chain signals for autonomous agents and machine learning. For example, autonomous agents could create new markets based on social media sentiment and historical data. Some examples:

• Monetizing data signals

• Content-driven financial agency

• Subscription-based data recommendation agency

• Influencers contributing data to autonomous financial strategies

• Data Contribution-related DAO and content funds aggregate various data sources to provide dynamic on-chain signals

Currently, AF has launched two main products, namely AO Link and Data OS.

AO Link is the message browser of the AO network, providing similar functions to the block browser in traditional blockchain systems. It includes message calculation functionality, graphical visualization of message links for clear and easy understanding, real-time message streaming for up-to-date information, and linked message lists for easy organizational navigation. Users can also view their token balance and message inbox. This tool provides a professional and efficient way to interact with and analyze the structure and activity of ao networks.

Data OS is the ContentFI protocol developed on AO Network, which uses autonomous AI agents to obtain content and regenerate content derivatives. Through this innovative approach, DataOS not only enhances the relevance and accessibility of content, but also establishes a reward mechanism for content creators. Currently, we can view various data on the AO network at https://stats.dataos.so/ and observe network activity. Various data related to the content are not displayed for the time being.

Dexi

Dexi is a crucial interactive interface for ordinary users to use agents in AO to participate in Agent Fi. It is also an application implemented by agents on the AO network, which can independently identify, collect and summarize various information in the AO network. Various financial data for various events (equivalent to Dexscrenner on AO). This data covers asset prices, token exchanges, liquidity fluctuations, and token asset characteristics such as smart contract details. Dexi mainly serves two types of users: end users who access the platform through the Web terminal and AO applications (which can be understood as Bot/Agent) that interact with Dexi by sending messages to utilize the collected data. As core infrastructure, the main service provided by Dexi is a data subscription service. Processes on the AO network can subscribe to Dexi's data flow for a fee and immediately receive alerts for price adjustments and other updates.

Outcome

Outcome is a prediction market built by the @puente_ai team and supported by @fwdresearch, @aoTheVentures and @aoComputerClub. Outcome provides users with a platform to place bets on various events. Current prediction themes in the market include technology, memes, business, games, DeFi and AO. The project claims that in the future users can make automatic bets in the prediction market by building autonomous agents that rely on real-life data and are based on large language models.

AgentFi on AO provides us with a new perspective to explore the future of deploying AI models directly on the blockchain and using various AI agents to perform automated transactions. The limitations of traditional single blockchains are broken by the design of AO+AR with novel underlying innovations. We look forward to seeing more applications on AO and cases of financial strategies combined with AI agents.

Reference

https://www.theblockbeats.info/news/53865

https://permadao.com/permadao/AI-on-AO-AO-AI-224ba15c840a4309972fec5350d9ed90

https://www.communitylabs.com/blog/ao-in-ai-key-highlights?utm_source=Blog&utm_medium=X&utm_campaign=AI+on+AO&utm_id=Community+Labs

https://www.autonomous.finance/research/en-US

The above is the detailed content of Agent-Fi on AO: A financial paradigm integrating AI agents. For more information, please follow other related articles on the PHP Chinese website!