Focus On Humanizing AI
As organizations begin their journey into artificial intelligence (AI), ethics often enter the design process. While achieving a uniform set of ethics may seem insurmountable, some design points will help facilitate the humanization of artificial intelligence and provide appropriate checks and balances. Constellation has identified design pillars for Level 1 AI. Level 1 AI is defined as machine learning proficiency (see Figure 1)
Figure 1. Five Levels of Artificial Intelligence Requires Different Design Points
The five pillars include (see Figure 2):
- Transparent. Alogrithms, attributes, and correlations should be open to inspection for all participants.
- Explainable. Humans should be able to understand how AI systems come to their contextual decisions.
- Reversible. Organizations must be able to reverse the learnings and adjust as needed.
- Trainable. AI systems must have the ability to learn from humans and other systems.
- Human-led. All decisions should begin and end with human decision points.
Figure 2. Five Pillars For Level 1 AI Ethics Focus On Humanizing AI
The Bottom Line. Instill The Five Design Pillars For AI Ethics In All Projects
Prospects of universal AI ethics seem slim. However the five design pillars will serve organizations well beyond the social fads and fears. The goal – build controls that will identify biases, show attribution, and enable course correction as needed.