The World Health Organization has a new chatbot prototype. Trained on OpenAI’s ChatGPT 3.5, it’s an artificial intelligence that goes by the name S.A.R.A.H. Short for Smart AI Resource Assistant for Health, the AI serves as a virtual health worker that’s available to speak with online visitors about everything from mental health to dieting in eight languages.

The virtual assistant, however, isn’t quite ready to solve the world’s health problems, because it “may not always be accurate.”

“These technologies are not at the point where they are substitutes for interacting with a professional or getting medical advice from an actual trained clinician or health provider,” Alain Labrique, the director of digital health and innovation at WHO, told Bloomberg.

Instead, the virtual assistant is part the organization’s campaign to further the development of technologies designed to help with the world’s health care worker shortage. On its website, the WHO welcomed researchers to use S.A.R.A.H. in their studies, and warned the virtual assistant is not always factual in its current state.

“WHO Sarah is using Generative AI to deliver health messages based on available data,” the WHO website said. “However, the answers may not always be accurate because they are based on patterns and probabilities in the available data. WHO takes no responsibility for any conversation content created by Generative AI.

“Furthermore, the conversation content created by Generative AI in no way represents or comprises the views or beliefs of WHO, and WHO does not warrant or guarantee the accuracy of any conversation content. Please check the WHO website for the most accurate information. By using WHO Sarah, you understand and agree that you should not rely on the answers generated as the sole source of truth or factual information, or as a substitute for professional advice.”