Human-AI Collaboration with Misaligned Preferences

Explore a 12-minute conference talk examining the challenges and dynamics of human-AI collaboration when preferences between humans and artificial intelligence systems are not aligned. Delve into research findings on how mismatched objectives and preferences impact collaborative decision-making processes between humans and AI systems. Learn about the theoretical frameworks and practical implications for designing effective human-AI partnerships despite conflicting goals or priorities. Discover insights into managing preference misalignment in collaborative scenarios and understand the broader implications for human-AI interaction design and implementation across various domains.