NASA’s new dark energy space telescope can also detect killer asteroids
technologyreview.com·2h ago
AI Research
arXiv CS.AI·23d ago·5 min read
LLMs for health: Perceived benefits, risks, intention to use AI chatbots, and willingness to self-disclose across sensitive health topics
TL;DR
Health-related AI chatbots, particularly large language models (LLMs), show promise in enhancing patient engagement and support. However, concerns about privacy and the accuracy of information persist.
✦ Why It Matters
Developers should prioritize user privacy and transparency in AI chatbot design to enhance trust and engagement.
Key Takeaways
- Design chatbots to address low-sensitive health topics effectively.
- Assess user characteristics to tailor chatbot interactions.
- Highlight benefits of using chatbots to encourage engagement.
- Monitor perceived risks to improve user trust in AI chatbots.
Full Summary
AI chatbots are increasingly utilized for addressing health-related inquiries, but their effectiveness can vary based on the topic and user characteristics. A study involving 1,388 participants in the Netherlands employed a mixed design to explore how topic type (physical vs. psychological) and sensitivity (low vs. high) affect perceived benefits, risks, and willingness to self-disclose health information.
Results indicated that perceived benefits positively influenced users' intentions to engage with chatbots, while perceived risks had a negative impact. Participants showed a greater intention to use chatbots for low-sensitive topics compared to high-sensitive ones.
Additionally, individual characteristics played a crucial role in shaping perceptions and willingness to disclose health information. These findings highlight the importance of understanding user perceptions in the design and deployment of AI chatbots for health.