Bi-Directional Alignment

An approach to human-AI interaction design that addresses alignment from both directions: not only adapting AI systems to match human values and preferences, but also helping humans understand, direct, and correct AI system behaviour. Traditional AI alignment focuses solely on making systems behave as intended, but bi-directional alignment recognises that users also need transparency about system reasoning, scaffolding to provide effective input, and agency to approve or modify AI outputs. This is especially important for neurodivergent users whose needs may fall outside normative training data.