image source head

A quick look at the AI ​​pet game The Farm: Does AI agent bring new gameplay to blockchain games?

trendx logo

Reprinted from panewslab

01/05/2025·5M

A quick look at the AI ​​pet game The Farm: Does AI agent bring new gameplay
to blockchain games?

Author: Zen, PANews

Although the chain gaming circuit has attracted much attention from capital and the market, its gameplay and models are highly homogeneous and lack real innovative breakthroughs. However, with the rise of AI agent technology, the blockchain game industry may usher in new opportunities for change. In this wave, The Farm attempts to create an unprecedented immersive game world through deep integration with AI Agents, and redefines the way players interact with the virtual ecosystem.

In addition to innovation at the conceptual level, The Farm, which is in line with the current hot spots in the industry, has also received initial recognition from the market. Its current market value has reached US$75 million, and it once increased by nearly 50% on January 3.

The Farm: GenAI-powered AI agent game

The Farm is the first GenAI-driven AI agent game based on Hyperliquid, which integrates on-chain creature generation (similar to CryptoKitties 2.0), simulated management (such as "Stardew Valley") and battle mechanisms (such as "Pokémon Go"), through AI Help realize a new experience for players to create and interact with. The game drives the economic ecology through $FARM tokens and combines multi-stage gameplay to gradually unfold.

Game design and features

Phase One: The Ancestors

This phase went live on December 13, 2024, and ended one week after launch. During this period, players can upload two photos to generate pixel-style hybrid creatures created by GenAI models for free. After spending 100 USDC to mint your favorite creatures, they will participate in voting, and the top 50 creatures in the voting results will become "ancestors". "Ancestors" will receive 10% of all future creature casting proceeds. Ancestors and its creators and voters will also receive airdrop rewards from developer wallets.

A quick look at the AI ​​pet game The Farm: Does AI agent bring new gameplay
to blockchain games?

Stage 2: Evolution

This phase begins immediately after Ancestor Selection ends on December 20, and its features will continue to roll out. All creatures are given characteristics, attributes, skills, personalities, favorite foods, etc., and these characteristics are generated by AI. Creatures will be given personalities, on-chain wallets, and support text and voice conversations. These creatures can learn AI skills such as tarot divination and lottery drawing. Players can train creatures by uploading data, and purchase food or sign up for courses to improve attributes. Additionally, organisms will gradually unlock autonomous agent behavior.

Stage Three: The Battlefield

According to The Farm roadmap, this phase is scheduled to go online in mid-February 2025. Creatures can participate in battles independently or in guilds formed by players. The battle mode supports betting. The winner will win the loser's gambling reward, and the loser's creatures may perish as a result. At this stage, the dimensions of biological interaction and competition will be further expanded.

System design of The Farm

According to an article published by The Farm introducing its system design, the game does not adopt the more mainstream SWARM system, but is based on and inherits the design concept of Langchain.

SWARM (cluster) is characterized by its autonomy, decentralization and flexibility. Each AI agent acts as an independent node in the SWARM pool, and its distinctive feature is the emergence phenomenon (Emergent Behavior), which will lead to dynamic propagation in task management. That is, tasks are processed through a decentralized, adaptive interaction method, and agents dynamically discover and collaborate to complete tasks based on local decision-making and interaction.

When a request is sent to the AI ​​agent in the SWARM pool, the agent either completes the request independently and returns the result, or breaks the request into subtasks, handles the part itself, and passes the remaining subtasks to other AI agents in the SWARM pool for processing. In the second case, since the agent cannot obtain a global view of the capabilities of all agents in the SWARM pool, its dynamic propagation methods may include broadcasting subtasks, forwarding based on local knowledge, assigning subtasks randomly or based on simple rules, and starting from the decentralized ledger. Read agent capability information. Although these dynamic propagation methods give the SWARM system agent autonomy, due to the lack of task planning and trajectory planning mechanisms, there may be defects such as time-consuming, high cost and loss of execution status.

As an Agent game that will involve the interaction of multiple agents, Farm proposes another complete design concept to achieve higher mission planning accuracy and better agent coordination. Farm believes that the multi-agent system on the chain should have higher accuracy in task trajectory planning, and at the same time, the status of agent execution should be tracked, which can be achieved through the data availability layer (DA).

Unlike the SWARM system, Farm introduces Orchestrator AI or on-chain AI Oracle service. The functions of this design concept include: task disassembly and allocation, service discovery and global view, subtask execution status and agent output result tracking, and dynamic adjustment to ensure the integrity of the task cycle (an agent cannot handle the task or times out, the system will redistribute it subtask). Through global view and optimal path planning, it avoids redundant calculations and waste of resources caused by dynamic propagation, and greatly improves the execution efficiency and success rate of complex tasks. This approach also reduces the risk of state loss and provides a basis for collaboration between agents, enabling multi-agent networks to achieve a higher degree of interoperability.

Token model: Half of the protocol revenue is used for FARM repurchase and

destruction

Starting from an AI agent game based on generative artificial intelligence (GenAI), The Farm initially plans to combine simulated business gameplay (such as "Stardew Valley") and battle mechanisms (such as "Pokémon Go") through on-chain creature generation (similar to CryptoKitties 2.0). 》) Attract users and gradually expand into a general AI agent launch platform, and eventually provide Rollup as a service function, enabling AI agents to have their own application chains and develop their ecosystem.

In terms of revenue mechanism, the game part of The Farm uses $FARM tokens for on-chain creature generation, in-game item and skill sales, and battle/betting rake. The general AI agent launch platform supports agents to issue their tokens and collect fees through the pre-sale and liquidity launch stages, while drawing commissions from agent services.

For agents that support the application chain, $FARM needs to be pledged to start the application chain. 50% of the protocol proceeds are used for $FARM repurchase and destruction, 40% is distributed to $FARM stakers, and 10% is distributed to the team. In addition, $FARM pledgers can obtain $veFARM, which can be used to share revenue and proxy priority configuration to achieve continuous accumulation of token value.

By introducing AI agents, The Farm aims to enhance the experience of open-world game players exploring the unknown world and growing up. Players can create characters that can continuously evolve according to their personal preferences, instead of being limited to a few fixed templates. The game world no longer has a preset script, and the creativity of all players will jointly shape the world. In addition, characters can interact with real people, breaking the boundaries between virtuality and reality.

more