$ETH Chicken plague Ethereum can't move much before it dips again, just shorted it at the same spot.

$ETH Didn't hold onto the bottom, 1.2 at 2925 was sold, 0.4 at 2940 was sold, sigh, lost 150u, so tragic, held the position for 2 days and still lost money.

$METIS ## Metis makes a bold move into AI: LazAI launches its mainnet Alpha version, offering an all-in-one solution The differentiation competition among Ethereum L2s is becoming increasingly fierce. While many rollup projects focus on optimizing transaction speed, Metis has chosen a different path: positioning itself as the next-generation AI infrastructure. In December last year, LazAI, a project incubated by Metis, officially announced the launch of its mainnet Alpha version, marking an important step for Metis in becoming the preferred L2 for AI applications. **Background: Why AI is the Next Frontier for L2** This decision was not made on a whim. Earlier this year, Metis formulated a comprehensive AI-centered strategy, primarily reflected in two aspects: Hyperion as a high-performance L2 supporting parallel execution of AI applications, and LazAI as a data and artificial intelligence network covering the entire model lifecycle. The underlying logic is profound. In the near future, AI will not only perform tasks within applications but also become the main interface for users. When AI receives human instructions and decides which application to use to complete tasks, the competition for “traffic capture” will undergo a fundamental change. L2s that control this AI input gateway will have an irreversible structural advantage. Meanwhile, Metis’s token (METIS) is currently priced at $4.92, down -5.29% in 24 hours, reflecting market volatility, but the project continues to steadily advance on its technical roadmap. ## LazAI: Not just an AI application, but the infrastructure for AI A common misconception is to confuse LazAI with other AI tokens or application platforms. In fact, it is quite the opposite: a complete network covering data, training, and deployment, built within a verifiable ecosystem. **Three main technological pillars:** **iDAO**: The minimal participation unit and consensus node. It can be any role in the AI lifecycle: data provider, model trainer, computing power contributor, or application developer. This modular design breaks down the AI ecosystem into interchangeable roles with consistent incentive mechanisms. **DAT (Data Anchoring Tokens)**: The core innovation. Unlike standard fungible tokens, DAT encodes three attributes simultaneously: proof of ownership and copyright (ownership and copyright proof), usage rights (access quotas, inference counts), and value participation (automatic generation of revenue linked to future usage). This allows data contributors to earn not only one-time transaction rewards but also permanent income. **Verifiable Computing Framework**: Solves the “black box” problem in AI. Combining TEE (Trusted Execution Environments) to ensure privacy, ZKP (Zero-Knowledge Proofs) to enable verification without exposing data, and OP (Optimistic Proofs) to improve speed. This hybrid system is similar to ZK Rollups but specifically designed for AI, balancing privacy, efficiency, and on-chain verifiability. ## Operational Process: From Encrypted Data to Verified DAT The process begins with users sending encrypted data to iDAO, which packages it into LazAI Flow and sends it to Quorum via VSC (Verifiable Service Coordinator, experts verifying data authenticity). LazChain’s consensus mechanism uses TEE/ZKP to verify and anchor hashes on-chain. After verification, a DAT is generated, registering metadata and ownership rights. When users transfer DATs to request services, TEE executes off-chain, and results are verified via ZKP/OP. The entire on-chain proof process is traceable and verifiable. ## Testnet achievements and mainnet Alpha release Before the official launch, LazAI has demonstrated tangible results. The testnet has accumulated nearly 140,000 active users. The advanced AI project Lazbubu, launched by the official team, attracted nearly 15,000 users, showing good user engagement. A typical application case is ROVR Network, which transforms ordinary vehicles into 3D map data collectors in the real world, using LazAI solutions. ROVR continuously generates high-precision geospatial data, input into the LazAI ecosystem as part of iDAO. These data are minted into DATs, forming DePIN and RWA databases, which future autonomous driving tools can utilize for self-optimization. **Developer Incentives:** With the launch of the mainnet Alpha, LazAI will not issue tokens for now but will introduce a developer incentive program worth 10,000 METIS. Covering everything from early prototypes to mature applications, including social media promotion and user growth funds. For developers and model contributors with AI expertise, this is an excellent opportunity to monetize through future airdrops. ## Academic support and development culture The technical credibility is further reinforced by academic backing. In June, Dr. Zehao Wang—core member of the Blockchain Research Center at the University of British Columbia and associate professor of Electronics and Computer Engineering—joined as a technical advisor. His expertise in multi-agent systems, edge trusted AI, and blockchain security adds depth to LazAI’s architecture. Metis values developer culture intentionally: first launching its pioneering decentralized scheduler in the L2 space, and now this incentivized ecosystem. This continuous effort attracts long-term talent. Infrastructure is “difficult, but right.” Early Web3+AI projects in the market generally fall into three categories: compute networks, launchpads for incentive models, and data annotation platforms. The first two fail to address actual needs, merely using blockchain as a tool; the third has limited scope. LazAI differs: it is a native Web3 infrastructure that solves specific problems through innovative mechanisms. It is not just a simple token airdrop patch but encodes logic directly into the protocol. Metis’s strategy is dual: expanding the on-chain application ecosystem on one hand, and positioning as a settlement layer on the other. As more non-AI application chains are built on mature L2s, the demand for Metis as a cross-ecosystem infrastructure will continue to grow. In the future, the value of blockchain tokens will increasingly depend on actual adoption and Gas demand rather than narrative speculation. Metis is building exactly such a product: supporting AI full lifecycle launchpad. For developers and users, the core products of the ecosystem strategy are always worth participating in.