Explainable AI
Research into AI systems whose outputs can be understood, traced and audited.
Design Principle
Transparency
Inference processes should be inspectable: source, model, prompt, framework, output and evaluation.
Research foundations
Relevant research traditions and thinkers that inform this principle.
Algorithmic Accountability
The study of how automated systems can be made answerable for their effects on people.
AI Governance
The policies and practices through which AI systems are developed and held to account.
Provenance Systems
Methods for tracking the origin, lineage and transformation of data and outputs.