Trust in Generative AI for Health Information Consumption and the Effect of Learned Dependency: An Experimental Investigation
arXiv:2606.20605v2 Announce Type: replace-cross Abstract: Background: Generative artificial intelligence (GenAI) is increasingly used for health information, yet its influence on users' trust calibration remains unclear. Objective: This study examines whether learned dependency on GenAI influences trust in AI-generated health information and whether text highlighting reduces overreliance on incorrect outputs. Methods: Two randomized controlled experiments were conducted with 338 college students and 563 Amazon Mechanical Turk participants. Both experiments used a 2 by 2 between-subjects design
The proliferation of GenAI in consumer health applications has made understanding user trust and dependency a critical and immediate research area.
This research provides crucial insights into how users interact with and become dependent on AI for sensitive information, directly impacting safe and ethical AI deployment in health.
Understanding how learned dependency forms and how to mitigate overreliance on incorrect AI outputs will influence future GenAI interface design and regulatory guidance.
- · AI ethicists
- · Healthcare providers
- · AI developers focused on safety
- · Digital health platforms
- · Unregulated GenAI developers
- · Platforms promoting unverified health information
Increased pressure on GenAI developers to incorporate features that calibrate user trust and reduce overreliance, such as highlighting potential inaccuracies.
Development of industry standards and regulatory frameworks specifically addressing trust, dependency, and responsible disclosure in AI-generated health information.
A potential shift in health literacy education to include critical evaluation of AI-generated content, empowering individuals to use GenAI tools more safely and effectively.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI