Get the latest tech news

Just add humans: Oxford medical study underscores the missing link in chatbot testing


Patients using chatbots to assess their own medical conditions may end up with worse outcomes than conventional methods, according to a new Oxford study.

It includes important medical details (it’s painful to look down) and red herrings (he’s a regular drinker, shares an apartment with six friends, and just finished some stressful exams). This study is useful, but not surprising, according to Nathalie Volkheimer, a user experience specialist at the Renaissance Computing Institute (RENCI), University of North Carolina at Chapel Hill. This study serves as a critical reminder for AI engineers and orchestration specialists: if an LLM is designed to interact with humans, relying solely on non-interactive benchmarks can create a dangerous false sense of security about its real-world capabilities.

Get the Android app

Or read this on Venture Beat

Read more on:

Photo of Humans

Humans

Photo of Oxford

Oxford

Photo of missing link

missing link

Related news:

News photo

Humans have nasal respiratory fingerprints

News photo

Shot to the eye brings back vision in mice – humans next | The researchers hope to begin human clinical trials of their antibody technique by 2028, offering hope to thousands who suffer from retinal disease

News photo

Half of businesses rethink ditching humans for customer service bots