PlayBase
Last updated
Last updated
PlayBase is a decentralized AI data processing and data storage network powered by modular nodes. PlayBase's modular node network operates to validate actions and outcomes that occur in live content, allowing rich experiences to be built on top of this verifiable data.
Game/sport/stream/show data is obtained through multiple capture methods including API, native capture and hardware-based capture. Developers, creators, players, viewers and other third parties may all submit captured data through active or passive means to The PlayFi Protocol. An API and SDK will be provided for technical parties although technical ability will not be required for all submitting parties.
In order to process, validate and format a wide range of data being streamed to The PlayFi Protocol, PlayFi utilizes a decentralised AI compute layer. This AI layer will have a wide and advanced understanding of multiple content formats but those who would like to further expand the scope of the AI's understanding will be granted access to an SDK to train additional ML models.
Trained ML models will be sent to the AI Processor nodes. These nodes are responsible for running the streamed data through the ML models while verifying their validity and configuring them in a standardised format. All modular nodes receive rewards for their contribution to PlayBase.
The formatted data will be sent to the PlayFi data processor which is responsible for hashing the data, updating the lookup tree of the data edge network and sending the data to it.
The data edge network is a decentralised network responsible for storing and distributing data. Data is stored and retrieved on this network utilizing various measures of proofs and verification to ensure only reliable data is handled at each step of the process.
Data held within PlayBase storage modules can be accessed by both the PlayChain and external sources via oracles built upon the PlayBase API. Oracles pay $PLAY to the node operators to access the data held within. This $PLAY is distributed as rewards among nodes, AI model trainers, and data uploaders depending on their contribution to the network and the accessed data.
Data nodes are kept aligned by an Actively Validated Service (AVS) developed by PlayFi on top of Eigenlayer. Nodes will be required to dual stake $PLAY and ETH to operate a node and be eligible for rewards. When nodes misbehave by altering the original data, going offline for too long when active, not storing data or not providing the expected data to the consumers they can be partially or completely slashed. Nodes who operate in good-faith and fufill their responsibilities will be rewarded with $PLAY.
