Incentive-Based
AI Alignment

EC'26 Workshop - Rome, July 6, 2026

About the Workshop

AI alignment, defined broadly as the correspondence between AI behavior and human goals, is emerging as one of the central challenges in the deployment of AI systems. This workshop focuses on incentive-based approaches to AI alignment, drawing on ideas at the intersection of economics, theoretical computer science, and machine learning. We plan a program featuring plenary talks, short spotlight talks, a poster session, and a discussion of key research directions.

Call for Submissions

We welcome submissions of working papers or recently published work (especially in venues beyond EC). The workshop will be non-archival, and accepted papers will be given spotlight presentations or poster slots. We especially welcome work on topics such as:

Understanding the incentive structure of AI alignment problems: AI alignment problems may emerge as a result of incentive misalignment. What could be the possible sources of such misalignments, and how to formalize them?
Topics include (but not limited to): economic models of human and AI decision-making, econometric estimation of AI valuations, information asymmetry and computational gaps, economic role of AI evaluation, implications of of model misspecification, and models of AI adoption.
Economic failure modes and inefficiencies: How can markets or economic interactions fail as a result of AI misalignment?
Topics include (but not limited to): strategic response, strategic algorithm selection, moral hazard, market unraveling, competitive exclusion, algorithmic monoculture, inefficiencies induced by performative feedback loops, algorithmic collusion, bias amplification, and language model arbitrage.
Economic solutions to alignment challenges: How can economic tools be leveraged to promote alignment within and around AI systems?
Topics include (but not limited to): mechanism design, contract design, market design, pricing schemes, monitoring, robustness, incentivizing debate, bargaining, social choice, and incentive-oriented AI evaluation.