Human-AI collaboration success depends on quality and cost balance
From Task Success to Productive Success: Evaluating Human-AI Collaboration by Quality and Cost
Computation and LanguageArtificial IntelligenceHuman-Computer Interaction
Summary
Measuring how well people and AI work together usually looks at how fast tasks get done or how good the results are. This paper shows that's not enough because it ignores how much effort or interaction it takes to get there. The authors created a way to measure both the quality of the outcome and the cost of interaction, finding that some tasks benefit from more back-and-forth while others don’t. They also found that user opinions don’t always match actual productivity. This method helps make the hidden effort visible and can guide better AI system design.
What this means in practice
- •For ai system designers: Evaluate and improve AI interfaces by measuring both quality and user interaction cost to enhance overall productivity.
- •For customer support teams: Analyze chat interactions to reduce effort and improve outcome quality when using AI assistants during customer service sessions.
Authors
Saki Imai, Mert İnan, Malihe Alikhani
Abstract
AI productivity is often measured by task completion time, economic value, or improvements in outcome quality. However, these measures usually treat collaboration as a black box where they capture what output was produced, but not the interaction cost required to produce it. Motivated by economics literature, we introduce a productivity-oriented framework for evaluating human-AI collaboration as outcome quality relative to interaction cost. Across two datasets spanning four tasks, we show that: (1) sessions with identical quality ratings can differ by up to 70 times in interaction cost; (2) quality-cost relationships vary by task, with some tasks rewarding extended interaction and others favoring fast convergence; (3) subjective user ratings are not reliable substitutes for productivity; and (4) productive sessions are characterized by agents probing earlier and users spending less effort repairing the interaction. By distinguishing productive success from costly success, our framework makes interactional cost visible and shows how dialogue analysis can inform the evaluation and design of AI systems.