Exchange Ranking
Top Exchanges
24h Volume Ranking
Popularity Ranking
Exchange BTC Balance
Proof of Reserves
Decentralized Exchanges
Funding Rate
Funding Heatmap
Liquidation Data
Max Pain
Long/Short Ratio
Whale L/S Ratio
Binance/Okex/Huobi L/S
Bitfinex Margin L/S
ETF Tracker
Solana ETF
XRP ETF
Hong Kong ETF
Bitcoin Treasuries
Crypto Reversal
Ethereum Reserves
HyperLiquid Wallet Analysis
Hyperliquid Whale Watch
Large Transactions
On-chain Movement
Bitcoin ROI
Stablecoin Market Cap
Options Analysis
News
Articles
Economic Calendar
Features
Wallet
Contract Calculator
Security
Collections
Watchlist
Following
TAO Lu Zerui: Using Formula 1 racing as an analogy to AI business logic
Xavier Lyu used a Formula 1 racing race to explain his business model: "If your maintenance team is 2 seconds slow when tightening the wheels, even if you have the fastest engine, you may not be able to win the race." he said. In his analogy, the engine is the artificial intelligence model-a core asset that every laboratory is proud of; and the maintenance team is all the elements that make the model run cheaply and efficiently. Lu Zerui believes that the key to winning or losing the game now has shifted to the "maintenance team".
He is one of three founders who are building a business based on the Bittensor network. They showed the media how to use the network to do something little-known in the cryptocurrency world: find customers who are willing to pay.
First, explain its underlying architecture in plain language. Bittensor is a blockchain network driven by TAO tokens. It hosts independent projects called "subnets," each of which pays contributors-miners-from around the world to compete for a particular artificial intelligence task. Those who provide the best solution will be rewarded. The founders who operate these subnets see this setup as a global research engine: publishing a difficult problem that hundreds of competitors work on around the clock.
The cryptocurrency industry is often known for being more hype than actual revenue. But the three founders are putting forward different views.
AI model's "maintenance team"
吕泽瑞来自泛太资本（Pantera Capital），这是一家加密货币投资公司，他在那里从事量化和基本面研究工作。他的公司Pareton AI并不直接构建AI模型，而是致力于降低模型的运行成本并提高其速度。
吕泽瑞的商业逻辑是：选择合适的模型已不再是区分AI公司的关键因素，效率才是。整个行业的利润率都很薄，因此即使是单数字百分比的成本节省也能带来巨大的累积效应。Pareton利用Bittensor的矿工，比内部团队更快地搜索海量的优化可能性。“我们能够更快、更便宜地找到解决方案，”他说。他补充道，与早期设计合作伙伴进行的试点项目看起来非常有前景。
吕泽瑞认为，更大的机遇存在于科技行业之外：医院、学校和小企业希望使用私有、本地运行的AI。这些组织无法将敏感数据发送到云提供商，并且需要足够精简的模型以便在自己的机器上运行。
付费让人类教导机器人
王凯伦（Cameron Wang）从芝加哥大学毕业后，自2017年起一直在创业。他对机器人的看法是：身体已经准备好了，但大脑还不够成熟。
He said that the robot hardware is already quite good, but the internal intelligence level is closer to that of a two to three-year-old child. His company is filling the missing link. Regarding Open Roboto's work, he said: "You can understand it as an AI model of a robot."
The bottleneck lies in real-world data-robots learn by recording videos of humans actually doing physical work, but such data is never enough. Open Robot translates this scarcity into the business itself: contributors wear recording equipment while performing manual labor and pay for it through rewards earned by Bittensor. These footages are used to train the model.
Wang Kailun said that just two months after its launch, Open Robot has established partnerships with a number of companies. Its basic models have reached the most advanced level in robot benchmark testing, and the value of factory pilot projects reaches hundreds of thousands of dollars. "The demand there is huge," Wang Kailun said.
Global research force for sales teams
Gavin Zaentz worked on the stock exchange on Nasdaq before co-founding Leadpot, which automates lead mining and sales outreach for customers including Dropbox.
His use of Bittensor is the most direct of the three: He points the network to his products. As many as 256 miners competed to improve the performance of Leadpot sales agents, an arrangement he described as equivalent to allowing the company to "outsource our research and development." The result, he said, is "an agent system that we absolutely cannot build internally independently."
