Interactive AI and Human Behavior: Challenges and Pathways for AI Governance

TL;DR

This paper explores the complexities of human-AI interactions in the era of generative AI, emphasizing the need for new governance methods that incorporate behavioral insights to address evolving risks and benefits.

Contribution

It introduces a framework for human-centric AI governance that integrates behavioral insights and discusses methodological opportunities for studying dynamic human-AI relationships.

Findings

Identifies challenges in regulating adaptive, relational AI systems.

Proposes a regulatory approach incorporating behavioral insights.

Recommends new research methods for studying fluid human-AI interactions.

Abstract

As Generative AI systems increasingly engage in long-term, personal, and relational interactions, human-AI engagements are becoming significantly complex, making them more challenging to understand and govern. These Interactive AI systems adapt to users over time, build ongoing relationships, and even can take proactive actions on behalf of users. This new paradigm requires us to rethink how such human-AI interactions can be studied effectively to inform governance and policy development. In this paper, we draw on insights from a collaborative interdisciplinary workshop with policymakers, behavioral scientists, Human-Computer Interaction researchers, and civil society practitioners, to identify challenges and methodological opportunities arising within new forms of human-AI interactions. Based on these insights, we discuss an outcome-focused regulatory approach that integrates behavioral insights to address both the risks and benefits of emerging human-AI relationships. In particular, we emphasize the need for new methods to study the fluid, dynamic, and context-dependent nature of these interactions. We provide practical recommendations for developing human-centric AI governance, informed by behavioral insights, that can respond to the complexities of Interactive AI systems.