Had a chat with a friend working on an AI startup, and he said: 'Once the models we train are deployed on big platforms, the ownership of data and revenue distribution becomes completely opaque. In the end, either we get exploited for free, or we can't prove the value we've contributed.' This really hits the core pain points of Web2 AI: 1. Data value cannot be properly attributed, 2. Model contributions are hard to track, 3. Community members end up being 'unpaid blood donors.' What @OpenledgerHQ is trying to solve is precisely the attribution and liquidity issues of AI contributions. Over the past month, there have been a few signals worth paying attention to: 1) Proof of Attribution is moving toward mainnet validation It's not just a concept anymore—developers have already integrated it into the testnet, attempting to fully record the value chain of data and model usage. For developers, this finally means they can use an on-chain ledger to 'clearly document who did what.' 2) OpenCircle Fund has started disclosing supported projects The $25M ecosystem fund has entered the actual deployment phase, with the first batch of projects focusing on model marketplaces and AI Agent tools. This means we might soon see real-world applications that actually work, rather than just visions outlined in whitepapers. 3) Wallet interaction scenarios are evolving The collaboration with Trust Wallet isn't just about creating an 'AI chat interface.' It's exploring whether natural language can directly trigger on-chain operations while ensuring user signatures are verifiable. If this layer works, it would elevate Web3 from 'button-based interactions' to 'semantic interactions.' Why is this important? Because AI assetization isn't just about wrapping things in 'tokenization.' It's about enabling data, models, and Agents to truly be composable, callable, and billable. Achieving this requires the combination of attribution mechanisms, liquidity markets, and user entry points. The three metrics I'll be watching: - Whether attribution validation can be adopted by third-party projects (looking at adoption, not just isolated demos); - Whether applications supported by the fund can generate real transaction volume; - Whether the wallet interaction gray tests can be implemented in everyday use. In summary: OpenLedger is trying to make the 'value chain of AI contributions' transparent and billable. If they can truly pull this off, data and models will no longer be walled-off resources controlled by big platforms but will become 'liquid assets' that anyone can participate in and profit from.