News Technology Media Telecoms

Nokia launches blockchain data marketplace for AI models


Yesterday Nokia announced a new blockchain powered offering to enable data to be sold and used for AI and machine learning. Called Nokia Data Marketplace, the objective is to motivate enterprises and telecoms firms to share their data by offering them a monetary incentive.

The solution uses the Hyperledger Fabric enterprise blockchain and its permissioning to ensure the security and integrity of the data that’s exchanged. Ultimately the objective is to aggregate data from distributed sources for machine learning. It also enables federated learning in which data stays local, runs a machine learning model, and feeds back updates to improve the model without sharing the underlying data.

In many cases, the data might come from IoT devices. Nokia envisions the solution being used in numerous sectors from connected cars and aerospace to ports, energy, smart cities and healthcare.

Connected cars and data sharing is a hot topic. Denso and General Motors recently led a working group to create the latest blockchain mobility data sharing standards from the MOBI Alliance. And China’s auto manufacturers have launched a blockchain solution for connected vehicle data sharing.

Public blockchains have proved popular for driving data monetization. For example, Daimler is running a pilot with Ocean Protocol. Given corporates often don’t want to share the underlying data, analytics and AI can be run on their data locally with the key metrics communicated back. 

Another public blockchain AI solution is Fetch.AI which recently partnered with Bosch. It’s keen to enable collective learning to train AIs used within its devices.