Archive — US — techcrunch
People struggle to get useful health advice from chatbots, study finds
A recent Oxford study reveals that while many adults use AI chatbots for health advice, reliance on these tools can lead to poor health decision-making due to unclear communication. Nearly 1 in 6 American adults use chatbots monthly, but the study shows they often fail to diagnose conditions accurately compared to traditional methods. Experts urge caution as AI in healthcare grows.
Key points
- 1. Many individuals turn to AI chatbots for health advice due to long wait times and high costs in healthcare.
- 2. 1 in 6 American adults use chatbots for health guidance monthly, according to a survey.
- 3. An Oxford study involving 1,300 UK participants showed chatbots could hinder more effective health decisions.
- 4. Participants using chatbots had difficulty identifying health conditions and often underestimated their severity.
- 5. Key information was often omitted when querying chatbots, leading to poor recommendations.
- 6. The study highlighted a communication breakdown between users and chatbots, affecting health outcomes.
- 7. Experts advise using trusted sources for health decisions rather than AI output.
Why it matters
- The increase in AI application in healthcare parallels trends towards digital health tools amid service accessibility issues.
- Tech companies like Apple, Amazon, and Microsoft are exploring AI solutions for health-related applications, but the effectiveness remains in question.
- Concerns about AI in healthcare reflect broader debates on the readiness of technology for high-stakes advice, similar to ongoing discussions about generative AI ethics in various fields.
The findings of the Oxford study underscore the risks associated with using AI chatbots for health advice, revealing critical flaws in communication and decision-making. As healthcare technology evolves, reliance on chatbots must be cautiously addressed to prevent potential health risks.