Introducing Phoenix DataX

6 min readMar 9, 2022


Dear Phoenix Community,

Introducing the DataX marketplace for Phoenix Oracle Network — a data and AI marketplace that will be integrated seamlessly to Phoenix Oracle Network, providing enterprise users a seamless one-stop solution for acquiring additional datasets that are verified through our oracles.

To understand how this is differentiated and game-changing for the next generation of decentralized enterprise apps, we first have to understand the strategy and positioning of Phoenix Oracle Network as well as Phoenix Chain.

Phoenix Oracle Network

Differentiated Oracle Platform for Decentralized AI & Data Applications

Blockchain oracle technology is not new, and started off having limited use cases mainly connecting predictive markets and on-chain contract settlements with off-chain data. However with the proliferation of DeFi, oracles have cemented a new role of providing real-time price feeds — a seemingly simple task, albeit a core market need with high frequency and volume of use.

However, in this rather early stage in the evolution and adoption of blockchain technology, what various oracle networks have not yet accomplished is leveraging and integrating the technology with a variety of advanced data-driven use cases such as AI, federated learning, MPC (multi-party computation), IoT networks, B2B data exchange, and others. This is partially due to an initial focus on crypto-native ecosystems and use cases, and in turn these oracles are hard to use with more diverse data types and formats.

For Phoenix Oracle Network, our target scope of verification is data on an abstract-level — this includes raw data, structured data, schema, and AI models. For us, use cases range purely from acquiring additional verified datasets for AI model training to using oracles to sync AI model training parameters in federated learning. In the case of IoT applications for example, devices (edges) will compute certain machine learning applications locally, and transmit their parameters to the master node, and after synchronization the new averaged model across entire network will be transmitted back to the devices/edges. This back and forth transmission of data and parameters will be verified via Phoenix Oracle Network as to preserve the security and integrity of the system. In decentralized AI and data-related applications, the biggest risk is bad actors trying to sabotage or manipulate the system.

We are betting on the imminent emergence and subsequent proliferation of decentralized AI and advanced data-driven applications, and aim to play an important role in this space. Our team is equipped with blockchain engineers, data scientists, and machine learning engineers, as well as over 30+ years of combined experience in AI, data science, and big data analytics. We are well prepared to ride the upcoming wave of adoption centered around advanced data applications, privacy, and IoT.

Phoenix Oracle + Chain

An Integrated Suite of Solutions for the Next-Gen of Data-Driven Applications

Phoenix Oracle is chain-agnostic, however considering the user-acquisition strategy, it makes perfect sense to have a blockchain that works tandem with it — a full Layer 1 solution (and subsequently Layer 2 solution).

Additionally, we are currently designing and putting into practice various use cases that do not require a Layer 1 at all. This may sound strange, but the need is becoming more commonplace — decentralized data verification can be used in certain applications where smart contract VMs are not used. From the Phoenix ecosystem point of view and the optimal user-acquisition and adoption perspective, having flexible lightweight use cases that do not necessarily change the existing systems that an enterprise is using, and utilizing the least friction foot-in-the-door methods may be the ideal way forward.

However, using this approach, users who will be looking to increase the usage of blockchain technologies in their current application stack after adopting something simple like data plus verification, is bound to grow as decentralized data-driven applications become more mature. In such cases, an ecosystem built using a full stack approach (i.e. Layer 1 + Oracle Network + Layer 2) will have a definite advantage as a “full solution”.

The “full solution” point of view implies that an end-user is able to mix and match certain components to fulfill their use-case objectives. In our full-stack planning for Phoenix Chain, we believe that for additional components like Layer 2, there is no such thing as one standard solution. This means, based on use-case priority, we may separate a general Layer 2 module into three distinct functional ones, including: 1) Privacy Layer 2) AI Computation Layer 3) Transactional Layer. We will release separate plans for this shortly.

Introducing DataX

Curated Data Marketplace for Decentralized Data Applications

For applications that will be interfacing with Phoenix Oracle Network such as federated learning, privacy-enabled computation, collaborative advanced analytics, etc, a large portion of the cases the user will utilize their own internal data (“Bring your own data”). In some of the cases however, additional data enrichment from 3rd party data providers, or additional datasets may be required in order to more optimally complete the task at hand — for example, to train certain machine learning models.

Data exchange, monetization, or acquisition, is not a new concept in the realm of blockchain-enabled applications, but has not been implemented at-scale or successfully put into practice in a sustainable fashion. For Phoenix Oracle Network — DataX is the missing piece of the puzzle that provides an additionally unique differentiating factor that will complete the data plus verification one-stop-solution.

Phoenix Ecosystem Architecture

Seamless Integration with Oracle Network

DataX will feature seamless integration with Phoenix Oracle Network, enabling users to request verified data using a simple API, abstracting the verification process via Phoenix Oracle Network. DataX will work like any other third party data web service API.

Datasets and AI Models

DataX will initially prioritize on vertical-focused AI-training datasets provided by data partners such as Tencent and Unionpay. However, considering the niche and use-case focus of the Phoenix ecosystem, as well as considering how our oracle network is built, it only makes sense if the concept of “data” is expanded to AI models. In essence, DataX will also be a curated AI model marketplace.

Settled with PHB, Competitive Pricing

Phoenix Core Development has formed commercial relationships with the curated data partners — pricing of data + verification via DataX will be the same or less cost of standalone data itself, making the value proposition quite attractive. Another way to look at it is DataX is also a data value-added reseller. All transactions will be settled in PHB, regardless of the currency/crypto of origination. We will make use of automated OTC conversion channels as a part of the settlement protocol.

Curation & Selection

Unlike various other oracle networks, which tend to be open systems where they more or less allow the data ecosystem to develop by themselves, DataX will be selected and curated, with certain data partnerships developed strategically based on priority and optimal to target user adoption.

DataX’s initial priority is focused on fully-compliant, consumer-related behavioral and demographic data. Following priorities are as below:

  • Commodity/futures/derivative data
  • Industrial, geospatial, and logistic data
  • Macroeconomic and industry-level data

We hope that this update provides some valuable insight in the direction of the Phoenix ecosystem — Phoenix Chain, Phoenix Oracle and DataX — as we continue to position ourselves strategically in the blockchain space.




Phoenix is an L1 and L2 blockchain infrastructure, empowering intelligent Web3 applications, focusing on the next generation of AI & Privacy-Enabled Web3 Apps.