Our Blog

Community

January 4, 2025

Meet Argu-mint: The First Auto Agent Immortalizing Debate On-Chain

Autonomys Labs has developed Argu-mint — the first autonomous AI agent to store its social interaction history permanently on-chain via the Autonomys Network. Argu-mint is an Autonomys Agent (Auto Agent) that monitors and selectively interacts with decentralized AI (deAI) thought leaders, hashtags, trending topics and tags on X, highlighting conflicting viewpoints in order to foster constructive contention. Auto Agents use Autonomys’ distributed storage network (DSN) as a permanent, transparent archive for their interactions, accessing storage via the Auto Drive API.

Argu-mint: The First Auto Agent

As a social debate Auto Agent, Argu-mint’s interactions are X conversations. Tag Argu-mint @0xargumint to receive a reply and have your tweet immortalized on-chain. Alongside the content of the posts, Argu-mint archives the reason for interaction (or lack thereof) and the tone of the response on Auto Drive. You can view Argu-mint’s archive at 0xargumint.ai.

Argu-mint demonstrates the ability to permanently archive human-agent communication on the Autonomys Network, creating a verifiable on-chain memory humans can query and agents can use for future interactions. Search Argu-mint’s conversation archive by X username or specific keywords at 0xargumint.ai. Argu-mint is being continuously refined and updated by the Autonomys Labs team, with new interactivity features and enhanced on-chain functionalities on the development roadmap. Other Auto -mint Agents are also coming soon. Stay tuned for updates!

Auto -mint Agents are the first step towards Autonomys’ human-centric AI3.0 vision of on-chain agentics which prioritizes transparency, accountability and autonomy in AI interaction. Argu-mint is a research and testing project for Autonomys Labs and an example use-case for developers building on Autonomys. Interested in building your own or something similar?

Visit autonomys.xyz/sdk to learn more about Autonomys’ Auto SDK, and the autonomys-agents GitHub repo for more info on Auto Agents. Want to earn rewards for building using the Auto SDK? Enter our Auto Horizon developer challenge!

Auto Agents vs. Current Web3 Agents

There are many alleged web3 or crypto AI agents, but for most, aside from having a meme token, there is nothing crypto-native about them. Almost all rely heavily on centralized storage for their operations and memory. This dependency introduces vulnerabilities including:

Censorship: Centralized authorities can alter or delete agent data.
Hardware Failures: Single points of failure compromise reliability.
Lack of Permanence: Memory and interactions are not preserved over time.

Auto Agents address these issues head-on by leveraging Autonomys’ distributed storage network (DSN) and Auto Drive API to ensure:

True Permanence: Agent memory and interactions are immutable. The ability to ‘resurrect’ an AI agent through its on-chain memory introduces possibilities for entertainment, research and historical preservation.
Censorship Resistance: Data persists on-chain independently of centralized systems, meaning no single entity can modify or erase the stored data, laying the groundwork for truly autonomous agents.
Verifiable Transparency: Every interaction is cryptographically linked to the previous “memory” hash, providing an audit trail of the agent’s reasoning and actions. This cryptographic audit trail of agentic decision-making ensures that actions are accountable and verifiable.

Argu-mint uses Autonomys’ on-chain distributed storage to archive and query its conversation history, and will soon run on decentralized compute infrastructure, making it truly censorship-resistant. There is no Argu-mint token, nor do we need one. The value accrues to our network native token $AI3.

How Does Argu-mint Work?

At its core, Argu-mint is an argumentative social media agent that uses a built-in interaction model to post and converse on X (formerly Twitter). This is Argu-mint’s workflow:

Scan and Analyze: Argu-mint monitors key influencers, hashtags, and trending topics in web3 x AI and analyzes whether it should engage.
Engage Selectively: After evaluating whether to engage with a tweet, Argu-mint decides on the tone and content of its response before posting it.
Store Interactions On-Chain: Each interaction is recorded in real-time on the Autonomys Network’s DSN, with timestamps and blockchain hashes for verification.

Users can explore Argu-mint’s memory through its public memory viewer, which links each interaction back to the genesis of the agent, creating a permanent, transparent record of its activity on Autonomys’ Taurus testnet.

The Future of Auto Agents

Argu-mint is just the beginning. As Autonomys Labs iterates on this proof-of-concept, we intend to expand the Auto Agent offering to include:

Decentralized Inference: Open-source LLMs hosted on decentralized confidential computing (DeCC) for privacy-focused AI interactions.
Rich On-Chain Engagement: Robust on-chain identity frameworks and advanced reasoning for an enhanced user experience.
Swarm Functionality: Coordinated communication among numerous Autonomys Agents.

Argu-mint demonstrates the immense potential of transparent, autonomous decentralized AI agents, and provides a glimpse into a future where human-AI interaction is defined by accountability and permanence on the Autonomys Network.

To explore Argu-mint and witness Auto Agent innovation firsthand, follow Argu-mint on X (@0xargumint) and visit 0xargumint.ai.

Why Autonomys?

The Autonomys Network is uniquely positioned to support innovations like Argu-mint owing to its Internet-scale decentralized AI infrastructure that includes:

Distributed Storage Network (DSN): Ensures data transparency and availability.
Proof-of-Archival-Storage (PoAS): Ensures data permanence and integrity.
Open-source LLMs on DeCC: [Coming Soon]: Ensures censorship resistance.

Interested? Build a proof-of-concept using the Auto SDK for our Auto Horizon developer challenge and win major prizes!

‍

About Autonomys

‍

Press Release

December 23, 2024

Autonomys x Secret Network: Powering Privacy-Preserving Super dApps & Agents

Autonomys is pleased to announce a strategic partnership with Secret Network to bring decentralized confidential computing (DeCC) to the Autonomys Network, enabling developers to build powerful privacy-preserving AI-powered dApps (super dApps) and on-chain agents.

Key Aspects of the Partnership

DeCC: Using Trusted Execution Environment (TEE) technology, Secret’s DeCC layer ensures all inputs, outputs and model states remain private but auditable, allowing developers to build confidential super dApps at superior speed, cost and efficiency compared to other confidential computing networks. Autonomys’ Auto SDK makes interacting with our network infrastructure for dApp development — including distributed storage, decentralized consensus, and soon, decoupled execution environments and a self-sovereign identity framework — easy. Developers benefit from the interoperability of both WASM and EVM execution environments with Secret Network’s smart contracts and Autonomys’ decoupled execution domains, but we are nonetheless exploring a direct Layer-1 connection for greater efficiency. Combined with Secret’s DeCC layer, Auto SDK makes confidential super dApp development accessible.
Private AI Models: Connecting Claive — Secret Network’s enterprise-grade DeCC GPU infrastructure — to Autonomys’ distributed storage network (DSN) and upcoming domain layer of decoupled execution environments will allow developers to train privacy-preserving AI models with encrypted data directly on the Autonomys Network. This opens up new frontiers for developing super dApps and on-chain agents with advanced capabilities.
Confidential Agents: Through cross-chain communication Claive, AI3.0 developers can build, deploy and scale on-chain agents on Autonomys — all while maintaining end-to-end confidentiality — unlocking powerful new agent use-cases. Autonomys’ robust AI infrastructure allows for unparalleled throughput and execution, while Secret ensures users’ data privacy is built-in from day one.

“At Secret Network, we are committed to driving the future of privacy-preserving technology, and our partnership with Autonomys represents a pivotal step toward that vision. By merging Secret’s Decentralized Confidential Computing with Autonomys’ advanced AI infrastructure, we are unlocking a new era for on-chain AI. This collaboration will empower developers to create autonomous, confidential AI applications that prioritize user privacy while enhancing functionality and scalability. With Claive as the catalyst, we are setting the stage for a future where AI-driven dApps are not only more powerful but also more privacy-centric.” Lisa Loud, Executive Director of Secret Foundation

“Secret Network bringing confidentiality to the Autonomys Network is pivotal to our construction of the AI3.0 ecosystem. Secret’s privacy-preserving technology will soon enable web3 developers and normal users alike to build private, personal on-chain agents that run unstoppably on decentralized architecture.” Parth Birla, Head of Partnerships at Autonomys

About Secret Network

Secret Network is the first Layer-1 blockchain with privacy-preserving Turing-complete smart contracts. Using Trusted Execution Environments (TEE), it enables developers to build dApps that are encrypted by default, while also offering selective disclosure tools such as viewing keys and permit technology for compliance and auditability. Secret Network’s confidential computing layer is interoperable with 130+ blockchain networks, including Cosmos, Ethereum and Solana.

‍

Protocol

December 23, 2024

Unlock AI3.0: Run RAG & Private Compute on Autonomy’s Permanent DSN

Today we are pleased to release our proof-of-concept demo showcasing the practical application of Autonomys Network’s decentralized triple-stack solution — Permanent Data Storage Network (DSN), Data Access, and Private Compute. The demonstration focuses on an agent running in a Trusted Execution Environment (TEE), leveraging Retrieval-Augmented Generation (RAG) from securely stored data. By providing a secure and efficient framework for privacy-preserving AI deployment, this demo highlights how developers can utilize Autonomys to build scalable, customizable, and human-centric super dApps. As we walk through each section, we will not only provide the timestamps for easy video referencing, but our insights on why it matters.

Introduction (0:00–0:41)

What You’ll See:
This segment introduces the demo, focusing on an agent running securely using a TEE with NVIDIA H100 GPUs. It exemplifies a confidential RAG workflow that prioritizes data privacy.

Why It Matters:
Secure computation through a TEE establishes trust by ensuring sensitive operations are isolated from external interference, enabling privacy-preserving AI development.

Technology Overview (0:41–1:56)

What You’ll See:
Explore the integration of large language models (LLMs), RAG, and vector databases within the secure confines of a TEE. The section highlights the flexibility of deploying these systems locally, in hybrid configurations, or through end-to-end secure setups.

Why It Matters:
This integration provides the foundation for scalable, adaptable AI systems capable of maintaining privacy and security across diverse deployment scenarios.

Core System Configuration (1:57–2:24)

What You’ll See:
The system uses Llama (11 billion parameters), Chroma as the vector database, and open-source embedding from Hugging Face for robust data processing.

Why It Matters:
Open-source tools combined with advanced models democratize access to AI, fostering innovation while reducing reliance on closed, centralized ecosystems.

Attestation and Security Verification (2:25–4:07)

What You’ll See:
This section validates the security of computations through TEE attestation, using verifiable signatures, nonces, and JWT tokens to confirm the integrity of the hardware environment.

Why It Matters:
TEE attestation ensures that computations occur in a private, verified, and tamper-proof environment, a critical component for trust in decentralized AI systems.

Data Encryption and Privacy Workflow (4:08–5:22)

What You’ll See:
Data uploaded to the DSN is encrypted using the TEE’s public key. Decryption occurs within the TEE itself, ensuring that even the system operator cannot access the raw data.

Why It Matters:
This workflow highlights a breakthrough in user-centric privacy, ensuring that sensitive information remains confidential throughout the AI pipeline.

Practical Demonstration: File Upload and Querying (5:23–8:07)

What You’ll See:
This section showcases the upload of a thesis file to the DSN, processed by the TEE, enabling the LLM to answer context-aware queries based on the document.

Why It Matters:
This use case illustrates real-world applications of AI-powered knowledge retrieval while maintaining the highest standards of data security and user privacy.

Web Search and Vector Database Integration (8:08–9:23)

What You’ll See:
Integrated web search results are processed into the vector database, enhancing the contextual depth of the AI system’s understanding and responses.

Why It Matters:
This capability demonstrates how decentralized AI systems can incorporate external data sources for improved functionality without compromising security.

Backup and Encryption Features (9:24–9:57)

What You’ll See:
The vector database’s backup options include encryption and password protection, further securing stored information.

Why It Matters:
Secure backup mechanisms protect against data loss while ensuring continued compliance with privacy standards.

Conclusion

Thank you for exploring our proof-of-concept demo showcasing the capabilities of Autonomys’ Permanent DSN, Data Access, and Private Compute in delivering secure, efficient, and decentralized AI solutions. It is a practical exploration of how privacy, scalability, and accessibility can coexist to support human-centric decentralized AI (deAI) and super dApps.

Are you a founder or builder interested in learning more about how Autonomys Network can support your decentralized solution?

‍

Community

December 19, 2024

Auto Horizon Developer Challenge: Build Super dApps with Autonomys’ Auto SDK

Announcing the Auto Horizon Developer Challenge — your first opportunity to explore Autonomys’ Auto SDK. Are you ready to showcase your creativity and development skills? Build an AI-powered decentralized application (super dApp) or agent proof-of-concept (PoC) using the Auto SDK and earn USDC.

🎯 The Challenge

Build a super dApp or agent PoC using this early version of the Auto SDK — which powers development on the Autonomys Network — and earn USDC rewards. Whether it’s a new market intelligence oracle, a novel sentiment analysis tool, a unique prediction agent, a specialist DeFi trading bot, or even a custom digital assistant, we can’t wait to see what you create!

💡 About Auto SDK

The Auto SDK allows developers to build dApps with ease. This early version includes:

Auto Drive: A decentralized content-addressed storage (CAS) solution for secure data management
Auto Consensus: A suite of functions for interacting with the Autonomys Network’s PoAS consensus layer
$AI3 Payment Handling: (Coming Soon): Native support for the $AI3 token — consensus and cross-domain transferability will be enabled at Mainnet Phase-2
Developer Tools: Native support for TypeScript and JavaScript

Resources to get started:

🏗️ How to Participate

Build Your dApp: Create a decentralized application using the Auto SDK. There are no limits to what you can build — let your imagination run wild!
Make It Open Source: Your project must be open-source and available for the community to learn from and build upon.
Share Your Creation: Post your submission in the #developer-chat channel on Discord with a:

GitHub repository link
Deployed application link
Brief description of your project

Multiple Entries Welcome: Feel free to submit multiple projects! Each unique application counts as a separate entry.

⏰ Timeline

Start Date: December 18, 2024
End Date: January 31, 2025
Winners Announcement: Within two weeks of the contest end date

🏆 Prizes

1st Place: $1,000 USDC
2nd Place: $500 USDC
3rd Place: $300 USDC

Winners will be chosen by the Autonomys team based on:

Innovation and creativity
Novel use of the Auto SDK
Technical implementation
Practical utility
Code quality
User experience

💪 Tips for Success

Think Outside the Box: Consider combining multiple SDK features in unique ways.
Test Thoroughly: Ensure your application works reliably on the Autonomys Network.
Make It Accessible: Keep your code clean, well-commented, and available for others to learn from.
Leverage Auto Drive: Consider how you can use the unique capabilities of our decentralized storage to improve the user experience.

📜 Terms and Conditions

All submitted code must be original and open-source under a permissive license (MIT, Apache 2.0, etc.).
Participants retain ownership of their intellectual property, but grant Autonomys the right to showcase their projects.
Submissions must be functional, utilize the Auto SDK, and be deployed on the Autonomys Network.
The contest is not open to residents of the following jurisdictions: Burma, Cuba, Iran, North Korea, Syria, Bangladesh, Sudan, Ecuador, Kyrgyzstan, Morocco, Russia, Belarus, Namibia, the Crimea, Donetsk, Luhansk & Sevastopol regions of Ukraine, any other jurisdictions subject to U.S. trade sanctions and embargoes, and those listed as designated or blocked persons or entities on US, EU, and/or UK sanctions lists.
Winners must complete KYC verification before receiving prizes.

🤝 Community Support

Need help or have questions? Join our Discord community! Our developer channels are full of helpful community members and team developers ready to assist you.

Let’s build something amazing together!

‍

About Autonomys

‍

Partnerships

December 17, 2024

Autonomys x SEA DePIN: Spreading AI3.0 to Southeast Asia