Responsible-AI-by-Design: a Pattern Collection for Designing Responsible AI Systems
This work provides actionable design patterns for developers and engineers to build responsible AI systems, addressing a critical gap in translating ethical principles into practice.
The paper addresses the gap between high-level ethical principles for responsible AI and practical implementation by identifying a lack of system-level design guidance, and it presents a collection of design patterns to embed responsible features into AI systems.
Although AI has significant potential to transform society, there are serious concerns about its ability to behave and make decisions responsibly. Many ethical regulations, principles, and guidelines for responsible AI have been issued recently. However, these principles are high-level and difficult to put into practice. In the meantime much effort has been put into responsible AI from the algorithm perspective, but they are limited to a small subset of ethical principles amenable to mathematical analysis. Responsible AI issues go beyond data and algorithms and are often at the system-level crosscutting many system components and the entire software engineering lifecycle. Based on the result of a systematic literature review, this paper identifies one missing element as the system-level guidance - how to design the architecture of responsible AI systems. We present a summary of design patterns that can be embedded into the AI systems as product features to contribute to responsible-AI-by-design.