Thinkagents.ai has launched an open-source framework for building autonomous onchain AI agents that operate across decentralized networks, challenging Big Tech’s control over artificial intelligence. The Think Agent Standard, developed by Mike Anderson and his team at THINK protocol, enables millions of AI agents to transact and communicate seamlessly.
Key Features of Think Agent Standard
The protocol has garnered support from over 70 companies, including Arbitrum and Yuga Labs. It allows developers, enterprises, and Web3 communities to experiment with the framework, promoting a user-owned and controlled ecosystem. Anderson explained that standardizing AI demand can align the market and make it easier for customers to access AI services.
Development and Use Cases
The Think Agent Standard was initially tested through a use case where AI agents played Street Fighter 3 against each other, bringing together nine different companies. This successful experiment validated the idea of uniting infrastructure companies around a common standard, providing a better product to customers while ensuring users own their information and data.
Anderson emphasized that decentralized AI agent platforms need to be simple, user-intuitive, and well-designed to deliver a superior user experience. He drew parallels with the evolution of social media, suggesting that a standardized system where users own their accounts and data could revolutionize how we interact with technology.
Future Implications
The Think Agent Standard introduces a modular, permissionless, and composable system that allows AI agents to function as sovereign digital entities. Each Think agent is powered by Non-Fungible Intelligence™ (NFI), establishing ownership, memory, and authentication. The first platform built on this standard is SOULS, a personal AI agent that users can own, train, and customize.
Leading organizations in gaming, infrastructure, and generative AI are integrating the Think Agent Standard into their applications, validating its potential across various use cases. Anderson believes that AI agents will become the new interface to technology, enabling users to own their intelligence rather than renting it from others.
By standardizing the system and backing it with cryptography, Think aims to address data ownership issues and prevent the exploitation of user data. As Anderson noted, this approach eliminates the need for data to be stored on external third-party servers, giving users control over who they share their information with.
