COVID-19 Chatbots Show Alarming Inconsistencies in Data and Privacy
As overwhelmed hospitals turn to chatbots for COVID-19 triage, major disparities emerge in how these tools source medical information and handle user data. These inconsistencies raise serious concerns about the reliability of health guidance during a critical crisis.
Hospitals across the United States face overwhelming patient surges during the COVID-19 crisis, leading tech giants like IBM, Facebook, and Microsoft to promote chatbots as essential triaging tools. However, these AI assistants exhibit troubling disparities in how they source and handle critical health data. Such inconsistencies potentially lead to varied health outcomes for users seeking urgent medical guidance.
A review of six major chatbot providers reveals a fragmented landscape of information vetting. Companies like Clearstep, Orbita, and LifeLink maintain rigorous standards by consulting medical professionals and relying on CDC guidelines. In contrast, other providers avoid discussing their vetting processes entirely, claiming their tools serve only educational purposes rather than diagnostic functions.
Beyond data quality, significant concerns exist regarding how these chatbots handle personally identifiable information. With millions of anxious citizens interacting with these systems, the lack of standardized privacy practices creates a risky environment for sensitive health data. This patchwork of approaches highlights the urgent need for stricter regulatory oversight of AI-driven healthcare tools.