This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
The World Health Organization's AI health advisor, SARAH, has been criticized for providing inconsistent and unreliable health information. Investigations revealed that the bot frequently gave contradictory answers, failed to provide relevant healthcare provider contact information during medical queries, and offered inadequate support for users experiencing mental health crises. Advocacy groups have called for the tool to be taken down due to these reliability concerns.
The WHO's AI-powered health advisor, S.A.R.A.H. (Smart AI Resource Assistant for Health), is alleged to provide inconsistent and inadequate health information. The bot reportedly gives contradictory responses to the same queries, fails to offer specific contact details for healthcare providers, and inadequately handles severe mental health crises, often giving irrelevant or unhelpful advice.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
AI systems that fail to perform reliably or effectively under varying conditions, exposing them to errors and failures that can have significant consequences, especially in critical applications or areas that require moral reasoning.
Also classified as
Users experiencing severe medical symptoms or mental health crises may unsafely rely on the chatbot instead of seeking professional medical care.
The SARAH chatbot exhibits a clear lack of capability and robustness by providing contradictory answers, broken links, and geographically inappropriate crisis resources in a critical health context.
AI system
Due to a decision or action made by an AI system
Unintentional
Due to an unexpected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The risk of providing inconsistent and potentially dangerous health advice was an unexpected outcome of deploying the SARAH AI chatbot to assist the public.
Added categories peak at 2: Minor (Epistemic Harm, Direct)
The World Health Organization's AI health advisor, SARAH, exhibited significant reliability issues, including providing contradictory health advice and geographically inappropriate crisis hotlines. While this presents public health safety concerns, it does not pose a direct threat to national security, critical infrastructure, or sovereignty.
The system is described as an AI health advisor designed to provide advice and suggestions to the public based on WHO guidance, requiring human interaction and not taking independent actions.
Limited Risk: The system is a chatbot designed to interact with humans and provide health promotion advice, which falls under the category of chatbots requiring transparency.
The system is explicitly deployed by the WHO as a smart resource assistant to provide public health advice and healthy living guidance.
The report clearly outlines the specific failures of the SARAH chatbot, including contradictory answers and geographic limitations of its crisis support. While it lacks quantitative data on actual user harm, the nature of the AI's limitations is well-documented by testing.
The system could be viewed as a simple informational tool (Level 4) rather than a chatbot with moderate risk (Level 3), but its role in crisis support elevates its risk profile.
Technology
Inconsistent LLM Outputs
The bot provides contradictory answers to identical queries.
Hallucinated or Broken Links
The chatbot dispenses broken links and poor-quality answers.
Context Switching Failure
The bot abruptly switches from provider search to tobacco cessation.
Data Inputs
US-Centric Crisis Data
Suicide helpline advice defaults to US-only numbers for global users.
Incomplete Provider Directory
The system failed to retrieve or share local provider contact details.
Human Factors
User Reliance on Critical Advice
Users seeking mental health support receive inadequate global resources.
Process and Methods
Inadequate Testing and Validation
The chatbot was deployed with critical flaws in crisis response.
Management
Premature Deployment
WHO released the bot despite critical shortcomings in mental health advice.
Regulatory Environment
Lack of Chatbot Health Standards
No clear regulatory standards govern medical advice provided by chatbots.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
