Phoenix Oracle Network — Beta Launch!
Dear Phoenix Global Community,
We are glad to announce that the Beta version of Phoenix Oracle is live! Check out the GUI to see it in action here https://oracle.phoenix.global.
The vision of Phoenix Global has been very clear: to create a comprehensive, blockchain-based solution for the next generation of data-intensive and AI-driven enterprise apps. Within that vision, both a Layer 1 blockchain and an oracle data service are crucial in providing the differentiated and focused value for applications in areas of data sovereignty and decentralized AI (federated learning).
Phoenix Global blockchain provides the core mechanism for trust, governance, and management of the decentralized AI process, and Phoenix Oracles serves as a secure gateway to streamline data, AI models, computational resources, and privacy-preserving logic with on-chain smart contract processes.
For Phoenix Oracle Network we have the following strategic value drivers that’s differentiated from other generalized oracle platforms:
- The notion of “data” as an abstract concept that include raw data, structured data, schema, objects, models, and logic. Through our oracles, combined with our focus on decentralized AI, we are able to harness the value and potential of blockchain oracles to a much deeper extent. Phoenix Oracle Network serves as a secure and trusted gateway to facilitate and coordinate off-chain AI computation, privacy-preserving business logic, and key data itself with a on-chain governance. This means that Phoenix Oracle Network is not only a secure gateway for data itself, but also AI models, parameters, logic, and 3rd party data exchange.
- Most current use cases of blockchain oracles are transactional, or involve the smart contracts being the endpoint of the oracle data request. For example, in cases of DeFi, from an off-chain data source. In the case of Phoenix Global specific use cases, such as Federated learning, the use of the oracle is not limited to data requests (ie. Price lookups) but rather exchange of information, models, and logic between participants of the same dApp as well, not necessarily needing outside data in order to create value. One simple example – the oracle will assist in the synchronizing the parameters of a trained machine learning model across computation nodes, which an iterative process that happens continuously.
Phoenix Oracle Network was built to be chain-agnostic for the most flexibility and optimal ecosystem development – for this Beta implementation the demo GUI is being run on Ethereum Testnet. We will also implement Phoenix Oracles on Binance Smart Chain and Phoenix Chain, as well as subsequently strategically picked ecosystems such as Klaytn and Nervos.
Along with this Beta release, we are also happy to announce a new core extension module of Phoenix Oracle Network: DataX Data Exchange Protocol. DataX is a protocol for enhancing and enriching data for decentralized data and AI applications via Phoenix Chain and Phoenix Oracle Network, and harnesses two main capabilities:
- Enable Phoenix Oracle users to easily access proprietary datasets for enterprise data use cases (AI, marketing, risk management, etc) that are rare and cannot be accessed by other oracle solutions. DataX will enable users to pay using a pay-as-you-go model (using the Phoenix Global token PHB) and the data to be verified via Phoenix Oracle Network (most of these data would require expensive bulk enterprise contracts otherwise). Current data partners include Unionpay, China Telecom, WeChat, and various commodity and economic data vendors.
- Enable Phoenix Oracle users and federated learning participants to cross-monetize and trade datasets. DataX will also enable participants to “trade” their datasets in a decentralized manner in order to have enriched datasets for machine learning for various use cases, including consumer insights and CRM on a B2B/P2P basis. For example, two enterprises may have different data on the same consumer/customers – DataX will facilitate a decentralized value-weighted exchange of such data, which would ultimately be Oracle-verified and can directly be used in federated learning. This type value exchange would have transformative value.
We will release separate detailed information and specifications for DataX shortly.