Designing for Constructive Civic Communication: A Framework for Human-AI Collaboration in Community Engagement Processes

TL;DR

The paper investigates how AI can support constructive civic communication within community engagement by analyzing two key pathways: between public leaders and community organizations and among community groups facilitated by public leaders. It highlights benefits such as streamlined workflows, improved sensemaking, accessible translations, and pattern discovery, while also warning of risks like hallucinations, opacity, bias, and erosion of human agency. Adapting Shneiderman’s human-centered AI framework, it proposes a two-dimensional control-automation grid with separate public-leader and community-organization controls to ensure high automation does not compromise agency. Practical design guidance emphasizes inclusive multi-channel participation, flexible interfaces, robust feedback loops, and deliberate facilitation of reflection and deliberation, with concrete tools like Coalesce, CommunityPulse, SenseMate, and the Feedback Map Tool to reduce burden and enhance transparency. Collectively, the paper argues that careful design can enable AI to amplify human connection in democratic governance while mitigating risks that threaten legitimacy and participation.

Abstract

Community engagement processes form a critical foundation of democratic governance, yet frequently struggle with resource constraints, sensemaking challenges, and barriers to inclusive participation. These processes rely on constructive communication between public leaders and community organizations characterized by understanding, trust, respect, legitimacy, and agency. As artificial intelligence (AI) technologies become increasingly integrated into civic contexts, they offer promising capabilities to streamline resource-intensive workflows, reveal new insights in community feedback, translate complex information into accessible formats, and facilitate reflection across social divides. However, these same systems risk undermining democratic processes through accuracy issues, transparency gaps, bias amplification, and threats to human agency. In this paper, we examine how human-AI collaboration might address these risks and transform civic communication dynamics by identifying key communication pathways and proposing design considerations that maintain a high level of control over decision-making for both public leaders and communities while leveraging computer automation. By thoughtfully integrating AI to amplify human connection and understanding while safeguarding agency, community engagement processes can utilize AI to promote more constructive communication in democratic governance.

Figures (2)

• Figure 1: A diagram of the key communication pathways in a typical community engagement process.
• Figure 2: Shneiderman’s human-AI collaboration framework