Collaborative Causal Sensemaking: Closing the Complementarity Gap in Human-AI Decision Support

Researchers propose Collaborative Causal Sensemaking (CCS) as a new framework to improve human-AI collaboration in high-stakes decision making. The study identifies a 'complementarity gap' where current AI agents function as answer engines rather than true collaborative partners, limiting the effectiveness of human-AI teams.

• →Current LLM-based agents fail to reliably improve human-AI team performance compared to individual experts in high-stakes scenarios.
• →The complementarity gap exists because AI systems are trained as answer engines rather than collaborative reasoning partners.
• →Sensemaking capabilities - co-constructing causal explanations and surfacing uncertainties - are missing from current AI training pipelines.
• →CCS framework proposes new training environments that reward collaborative thinking and shared mental models.
• →The research agenda shifts focus from oracle-like AI systems to AI teammates that co-reason with humans.