LazAI Launches Mainnet Operation, Metis Competes in Data Infrastructure for the AI Era

Metis Launches LazAI, an AI Data Layer Project, Officially Initiates Alpha Mainnet on December 22 Night (GMT+8). Behind this release lies Metis’s deep contemplation on the future integration of Web3+AI.

Why Does Metis Stick to the AI Path?

Unlike other L2s that are shifting towards high-frequency trading and application chains, Metis has anchored its AI strategic direction since the beginning of the year. This is not following the trend but based on a calm market judgment.

While Web3 fills market gaps in regions with weak financial infrastructure, its advantages at the AI application layer are not obvious. However, at the infrastructure level, the situation is entirely different—cloud service providers like Alibaba Cloud and AWS have begun integrating L2 and Alt L1 tools, offering users more options, often at lower costs than Web3 solutions.

Metis leverages the computational speed and verification efficiency advantages of L2 to launch LazAI, an integrated solution. From engineering design to market adaptation, LazAI demonstrates originality—it is not merely a transplant of Web3 concepts.

LazAI’s Three-Layer Structural Design

The core innovation of LazAI focuses on three aspects:

iDAO and Participant Modularization

iDAO is the smallest network participation unit and also the consensus node. Data providers, AI model trainers, compute contributors, application developers—every role in the ecosystem can become an iDAO, enabling modular participation in the AI lifecycle.

DATs Data Token Standard

DATs (Data Anchoring Tokens) are the core innovation of the LazAI team, adopting a semi-homogeneous token standard. Each DAT encodes three key attributes: a “proof of ownership” certifying asset origin and author identity, a “usage right” defining call quotas, and a “value participation” that automatically grants holders profit sharing.

This mechanism allows data contributors and AI developers to continuously profit from their contributions, rather than earning a one-time reward.

Computational Verification Framework

LazAI employs a hybrid verification system combining TEE (Trusted Execution Environment), ZKP (Zero-Knowledge Proof), and OP (Optimistic Proof) to address the “black box” problem of AI computation. This system is similar to ZK Rollup but tailored for AI scenarios, balancing privacy, efficiency, and verifiability.

Ecosystem Layout During Mainnet Alpha Stage

LazAI’s mainnet Alpha currently adopts a phased strategy—no token issuance for now, but providing opportunities for professional contributors and AI developers to profit via airdrops.

Developer Incentive Program totals 10,000 METIS, covering the entire cycle from early prototypes to mature applications, with multi-tiered ecosystem support including social media promotion and user growth funds.

Initial Results on Testnet

LazAI has demonstrated market appeal during the testnet phase:

• Active User Base: Nearly 140,000 users
• Official AI Partner: Evolved AI partner Lazbubu has attracted close to 15,000 users
• Ecosystem Applications: ROVR Network (which converts daily vehicles into 3D physical space data mappers) has adopted LazAI solutions

ROVR continuously maps environments with its devices, generating geospatial datasets. These data are minted into DATs within the LazAI ecosystem, forming high-precision DePIN and RWA databases. This provides a self-optimizing foundation for future autonomous driving AI tools.

Metis’s Technical Accumulation and Talent Attraction

Metis was the first to implement decentralized sequencers on L2, reflecting its dedication to technological innovation. Its developer-focused culture has also attracted academic attention—Dr. Zehua Wang, core member of the Blockchain Research Center at the University of British Columbia and Associate Professor of Electronics and Computer Engineering, joined LazAI as a technical advisor in June this year.

Dr. Wang has deep expertise in decentralized multi-agent system collaboration and security, especially in trusted edge AI, blockchain security, and zero-knowledge proofs.

Why AI?

This question seems simple but has profound logic behind it. As Ethereum mainnet’s efficiency bottleneck is gradually overcome, many L2s face growth pressures. Some choose to build their own L1 or develop application chains, but these solutions also fall into homogeneity competition.

A key observation is: future application entry points may not be a competition among humans but among AIs. When ByteDance’s phone is equipped with Doubao AI, users no longer need to operate multiple apps individually; instead, they tell the AI their needs, and the AI orchestrates applications to achieve goals. This changes the underlying logic of internet traffic competition.

If L2s focus only on transactions, prediction markets, or RWA tokenization, but neglect AI as an entry point, these application chains may ultimately become AI’s workers rather than its masters. Metis clearly saw this a year ago.

In March this year, Metis announced a dual-chain strategy at ETHDenver: besides Metis, Hyperion is a high-performance L2 optimized for AI applications, supporting parallel execution and real-time feedback. Hyperion is deeply integrated with the Metis SDK, supporting modular application chain construction, optimized for high-frequency trading and real-time AI applications. LazAI is the flagship product under this strategy.

The True Logic of Long-Term Value

Metis and other L2s face the same dilemma: the efficiency advantage of L2 is gradually eroded by Ethereum mainnet, and they must dominate at least one domain with strong products to ensure stable on-chain usage and ecosystem operation.

While AI infrastructure is challenging, the right direction. Web3 solutions for AI data annotation have only just begun to emerge in recent months. Metis, as a pioneer, offers a truly Web3 Native solution, not just on-chain confirmation and token issuance.

Metis’s application ecosystem expansion and chain-as-settlement-layer strategy are complementary. In the future, token value will increasingly be linked to real applications, and network adoption and demand for Gas tokens will be the core factors determining token and network value.

LazAI, as a key part of this, not only enhances Metis’s own value but also validates a trend: if AI-unrelated application chains flood onto L2, the value of METIS will become even more prominent.

Blockchain products are penetrating all dimensions of internet applications, with outstanding performance in AI. Projects that only do “on-chain models” or “AI Launchpads” are unlikely to have long-term vitality, while products like LazAI that span the entire AI lifecycle are worth attention. For developers and users, products occupying the core of the ecosystem strategy are always worth participating in.

METIS Real-Time Data Reference