Trusted Data Intermediary for UK Choral AI Dataset

Pioneering a Trusted Data Intermediary for AI Model Training in the Cultural Sector

Future Art Ecosystems has developed the first trusted data intermediary within a cultural organisation to enable responsible data stewardship for AI training. Through Serpentine Arts Technologies' collaboration with artists Holly Herndon & Mat Dryhurst, and partnerships with the Centre for Data Futures, King's College London, RadicalxChange, AWO and Keystone Law, we have created a legal and governance framework to steward a dataset of UK choir audio recordings for AI model training.

This groundbreaking experiment has yielded valuable insights into developing collective governance for training data, particularly regarding the challenges of participatory decision-making and legal constraints. These insights fall into two main categories:

Best practices in developing AI training datasets, including hiring a Data Steward to facilitate participatory governance for the dataset's use in The Call and future AI development;
Data rights, IP, and intermediating entities as tools for collective empowerment in the AI ecosystem.

Choral Data 'Trust' Experiment White Paper

The best way to keep up-to-date with new publications, events and general updates on this project is to sign up to our newsletter and/or join our Telegram.

<aside> 💬 All media is now training data.

Herndon Dryhurst Studio

</aside>

<aside> 💬 Data coalitions can become places for learning new ways of participating.

Matt Prewitt, RadicalxChange

</aside>

<aside> 💬 Data as a resource defies all traditional legal and economic tools of analysis. We’ll need creative proposals for the new frameworks we create to govern it.

Pioneering a Trusted Data Intermediary for AI Model Training in the Cultural Sector

Project Background

What is a ‘Trusted Data Intermediary’?

Project Materials

Further Reading

Roadmap

Key Roles & Terms

Partners