Yuyue
Yuyue|Jun 20, 2025 06:56
These days, the Little Octopus OpenLedger @ Openledger HQ China Tour has started, and its voice on the @ KaitoAI list is also moving forward. Kaito's new algorithm of "quality over quantity" provides positive feedback to bloggers who insist on manual combat, and also highlights the fairness of the mouth massage project that is worth discussing Speaking of OpenLedger, before their trip to China, they conducted a series of activities in Korea. I saw that many Korean bloggers were concerned about the issue of "AI data profitability fairness" that the project was solving. The PoA (Proof of Attribution) mechanism of OpenLedger: 👉 PoA solves the problem of fairness in AI profitability: a three piece set of "value consensus+verifiable attribution+automatic distribution of profits" The focus of OpenLedger is on the "contribution as a reward engine" in the model layer: it revolves around whoever creates traceable value in model calls receiving the money. The specific profit mechanism is data upload=ownership confirmation, data operation=value evaluation, training logs=contribution traceability, quality governance=punishment mechanism, and on this basis, token profits are allocated 👉 Professional models, the PoA+SLM approach is where the momentum lies When major manufacturers are focusing on universal large models (LLM), OpenLedger is targeting specialized models (SLM). The key point for OpenLedger to create SLM is vertical scene priority, which encourages domain bloggers and small teams to upload vertical data (legal/financial/medical, etc.) first. In this way, each SLM can be like an independent intelligent protocol that can be continuously called and profited by agents. In the future, multiple SLMs can run in series and be called by the Agent architecture, no longer isolated 👉 Other noteworthy points: -Led by Polychain, with participation from top venture capitalists and founders such as former Coinbase CTO and Polygon, raising $15 million in funding -There are already 22.18M total transactions on the test network, with 187.52K transactions per day. The travel schedule in China is very dense, and the portal guides the way ⬇️ https://(x.com)/OpenledgerHQ/status/1934948813892112441 Before TGE, it's just a matter of lip syncing. After TGE, it's important to focus on its SLM mechanism and the integration of agents to see if it can truly break through the call distribution loop, in order to determine if it's difficult to falsify
+4
Mentioned
Share To

Timeline

HotFlash

APP

X

Telegram

Facebook

Reddit

CopyLink

Hot Reads