Related Experiment Videos
Mapping Human-AI Teaming in Risk Analysis: Role Evolution, Thematic Landscape, and Governance Mechanisms From News
Cong Cheng1, Jian Dai1, Lulu Yan1
1School of Management, Zhejiang University of Technology, Hangzhou, China.
Summary
Public discourse reveals human-AI teams in risk analysis often fail due to poor oversight design. Effective governance requires deliberate interaction design beyond basic human-in-the-loop requirements for robust AI systems.
Area of Science:
- Artificial Intelligence
- Risk Analysis
- Human-Computer Interaction
Background:
- Human-artificial intelligence (AI) teams are increasingly integrated into risk analysis processes.
- Public discourse frequently highlights the fragility of oversight mechanisms in human-AI collaboration, particularly when review conditions are inadequately designed.
Purpose of the Study:
- To examine public discourse on the roles, failure modes, and governance of human-AI teaming in risk analysis.
- To identify key themes and dynamics in media portrayals of AI in risk contexts.
Main Methods:
- Analysis of 184,282 AI-related news articles from Factiva (1956-2025) using a funnel-shaped mixed-methods design.
- Structural topic modeling to identify 20 thematic clusters and their dynamics.
- Large language model (LLM)-assisted inductive thematic analysis to identify failure modes and governance mechanisms.
Main Results:
- Five mirror-mapped pairs of failure modes and governance mechanisms were identified: opacity/explainability, bias/auditing, agency erosion/cognitive friction, systemic fragility/oversight and liability, and accountability vacuums/institutional governance.
- Media discourse predominantly follows an AI-predicts-human-decides paradigm.
- Formal human-in-the-loop requirements are often insufficient without careful interaction design.
Conclusions:
- Public discourse suggests that effective human-AI teaming in risk analysis necessitates intentional interaction design to address identified failure modes.
- Governance mechanisms must be robust and adaptable to ensure meaningful oversight and mitigate AI-related risks.
Related Concept Videos
01:29Impression Management Techniques I: Managing Appearances
Appearance is a multidimensional aspect of self-presentation that encompasses observable attributes such as clothing, grooming, speech, and nonverbal behavior. These elements are often strategically managed to align with socially constructed expectations in different settings. For instance, individuals tailor their appearance during job interviews, social gatherings, or athletic events to meet the perceived norms of those environments.Contextual Adaptation and Social SignalsThe research...
01:29Impression Management Techniques III: Aligning Actions
Aligning actions are communicative strategies individuals employ to maintain social harmony and preserve personal identity in the face of potential disruptions to social norms. These actions are particularly important in managing social impressions when one's behavior might be seen as inappropriate, incompetent, or morally questionable.Types of Aligning ActionsThe three principal types of aligning actions are disclaimers, accounts, and apologies.DisclaimersDisclaimers are preventive; they are...