Disclaimer: The opinions expressed by our writers are their own and do not represent the views of U.Today. The financial and market information provided on U.Today is intended for informational purposes only. U.Today is not liable for any financial losses incurred while trading cryptocurrencies. Conduct your own research by contacting financial experts before making any investment decisions. We believe that all content is accurate as of the date of publication, but certain offers mentioned may no longer be available.
Incubated by Metis, LazAI’s latest project introduces a novel framework designed to make AI more open, more human and more rewarding for everyone who contributes to its training and development.
LazAI, Metis work together to reconsider how AI uses data
Decentralized AI innovators LazAI are focused on building a new cohort of artificial intelligence and machine learning tooling. As of 2025, large-scale mainstream large language models (LLMs) have become too generic and too centralized to serve the growing needs of the digital economy.
Most of today’s AI is built on huge datasets scraped from the internet, designed to serve everyone but often fitting no one particularly well. These big models struggle to adapt to local cultures, niche industries or personal needs, and because they are created by a handful of companies, only a few elite actors can benefit from their income streams.
Backed by Metis, LazAI attempts to solve these problems in an entirely different manner. Their team is betting big on smaller, specialized AIs built from verified, high-quality human data. Also, they prioritize designing a system that rewards the people who provide it.
Through Data Anchoring Tokens (DATs), LazAI tokenizes datasets on-chain with traceable provenance (e.g., medical images restricted to ethical AI models). Contributors earn rewards proportional to the data’s impact. It can be described as an analog of liquidity mining in DeFi, but participants source liquidity instead of crypto.
Instead of judging AI by hype, LazAI tracks real-world performance. Models can earn rewards based on how well they actually solve problems, and developers can stake on their success.
Fair and inclusive LLMs are key for decentralized AI
LazAI introduces a way to measure how reliable AI agents are, solely based on the work they have done. It is a step toward making AI more transparent and accountable, not just a black box we hope for the best with.
iDAOs (individual-centric DAOs) empower communities to govern AI behavior and profit-sharing. For instance, a language-preservation iDAO trains niche dialect models and earns fees whenever their DATs are used.
To keep this concept working and democratic for various classes of participants, LazAI introduces the concept of Total Data Value Locked (TDVL).
This new-gen metric quantifies the market value of on-chain datasets, weighted by quality and utility. In turn, it relies on alignment yield and an agent utility score.
Alignment yield ratio is set to reward users who refine AI models through feedback, turning data contributors into stakeholders. An agent utility score rates AI agents’ reliability on-chain, akin to a FICO score for autonomous algorithms.
Addressing most dangerous roadblocks for AI progress
Contributors to the LazAI team highlight the importance of such developments for the progress of AI usage in crypto and far beyond:
LazAI knows it won’t be easy. They’re up against the same challenges that DeFi faced early on — preventing bad actors, making sure rewards stay tied to real value, and protecting users' privacy. But the vision is clear: we need a more open, human-centered AI economy where anyone can participate, where data is valued fairly, and where better AI comes from better collaboration — not just bigger machines.
Unlike traditional AI systems that centralize data, computing and alignment behind corporate walls, LazAI prioritizes data sovereignty, verifiable AI execution and individual-powered governance through iDAO, DAT and a verified computing framework integrated with Quorum-based Consensus.
LazAI enables a scalable, incentive-aligned AI ecosystem where contributors are fairly rewarded and AI outputs remain accountable. LazAI published a research article titled “Beyond TVL: Defining New Primitives in an AI-Native Economy” with more unique insights.
Vladislav Sopov