Designing Value-Aligned Human-AI Interaction

Zana Buçinca explores how decision-makers often over- or under-rely on AI advice due to design flaws in current decision-support systems, which frequently conflict with the core values of users. Historically, AI tools have prioritized optimizing AI performance over the effectiveness of the broader human-AI team. Buçinca presents research highlighting the critical role of cognitive motivation in enabling individuals to critically engage with and effectively use AI advice. She advocates for a paradigm shift toward designing human-AI interactions that align with human values and motivations, introducing a framework to achieve this and addressing the opportunities and challenges it presents.

#transparency , #ai , #artificialintelligence , #responsibleai , #ResponsibleArtificialIntelligence, #ethicalai , #TrustworthyAI, #RegulatoryActivity, #researchanddevelopment , #trust , #fairness , #safety , #reliability , #security , #privacy , #transparency , #sustainability , #accountability , #innovation , #technology , #BellLabs, #NokiaBellLabs, #nokia