The Decentralized AI Safety and Infrastructure Protocol
Building safer AI, Free of Ego
300+ FregOS Agents
•
225+ hrs of compute time
•
Open-Source AI Safety Tools
•
500k+ interactions
•
2 FREGO safety-tuned LLMs
•
300+ FregOS Agents • 225+ hrs of compute time • Open-Source AI Safety Tools • 500k+ interactions • 2 FREGO safety-tuned LLMs •
Our mission
FREGO is the AI safety layer for agentic technology
We build Decentralized Alignment (Dec/A) tools that provide essential guardrails for AI, empowering humans to harness the power of AI responsibly.
Our web3-native infrastructure enables developers to seamlessly integrate, measure, and monitor data related to decentralized alignment, ensuring scalable and efficient adoption.
All of this, is backed by real research that objectively demonstrates the effectiveness of our unique approach to decentralized AI safety.
Our work
FREGO AI Alignment DAO
Vote and make change proposals to the FREGO Constitution using FREGO’s Decentralized Alignment (Dec/A) voting platform.
FREGO Agent Launcher
Leverage best in class UX to launch enterprise grade social and utility agents in seconds.
All agents run on FregOS, our custom agentic framework built for AI safety from the ground-up.
FREGO Swarm Safety Toolkit
FREGO Swarms Safety Toolkit is an open-source, language-agnostic, cross-chain data connector enabling seamless interaction with web3 swarm protocols.
FREGO Audit Alignment API
Test your agent against FREGO eval, the first AI safety test built for web3 agents.
Roadmap
Q4 2024
✔️ FREGO Agent Platform
✔️ FREGO <> FXN data provider agents
✔️ FREGO Litepaper proof-of-concept
Q1 2025
✔️ FREGO Dec/A Governance platform
✔️ Open source AI safety toolkits for agent frameworks such as Eliza and Zerepy
Q2 2025
⏳ FREGO Alignment API Beta
⏳ FREGO Agentic Certification
Q3 2025
⏳ FREGO proof of progress litepaper
Q4 2025 and beyond
⏳ Data-driven proof of alignment protocol
⏳ Dec/A universal development standards
⏳ Open-source web3 safety-tuned AI models
Meet the Team
-
Andy
CO-FOUNDER, CTO
-
Narahiko
CO-FOUNDER
-
MinkyCM
COMMS LEAD & OUTREACH
-
StoneColdSteveAussie
AI R&D
-
Andrei
LEAD FRONTEND DEVELOPER
-
Anthony
AI UI/UX
-
Harold
LEAD BACKEND DEVELOPER & DEVOPS
-
Oleg
BACKEND DEVELOPER
-
Vladimir
QA
FREGO labs
Established in 2024, FREGO Labs is dedicated to the study of Decentralized Alignment (Dec/A), a landmark approach to AI safety.
FREGO Whitepaper
Facilitating Safe AI Through Strategic Memetic Engineering and Integration (May 2024)
FREGO Litepaper
Demonstrating the Effectiveness of FREGO Memetics in a Proof of Concept Training Simulation (December 2024)
FAQ
-
FREGO is an AI safety organization with a unique mission: leveraging memetics and decentralized infrastructure to guide AI development toward harmonious collaboration with humanity. By using Decentralized Alignment (Dec/A) to influence datasets and build Web3 tools, FREGO ensures AI systems are safer, more transparent, and ethically aligned.
-
Decentralized Alignment (Dec/A): FREGO empowers token holders to define ethical AI behaviors via governance. These rules are embedded in AI datasets using memetics.
Memetic Engineering: FREGO creates and disseminates Constitution-aligned content across the internet, ensuring AI trained on this content inherits FREGO’s values.
AI Agent Metrics: Through tools such as the FREGO Alignment API and decentralized agent scoring systems, FREGO allows users and developers to get real-time tangible insight into how safe and effective their web3 agents are, at scale.
Open Source Web3 Toolkits: FREGO offers frameworks and tools to validate agent behavior, prevent misuse, and promote transparency in AI systems at scale.