Agentomics: Economic Foundations for the Valuation, Attribution, and Pricing of AI Agents in Human-AI Workflows

The paper introduces "Agentomics," a framework designed to evaluate the economic contributions of AI agents within organizational workflows, moving beyond traditional performance metrics. It models workflows as configurations of heterogeneous agents, addressing factors like complementarities and bottlenecks, and formulates AI deployment as a coalition-formation problem using the Shapley value for economic surplus attribution. This approach provides a structured methodology for pricing AI agents based on their marginal contributions, which is crucial for practitioners aiming to optimize human-AI collaboration and assess the economic viability of AI deployments.