For the sake of multifacetedness. Why artificial intelligence patient preference prediction systems shouldn’t be for next of kin

AI knows best? Avoiding the traps of paternalism and other pitfalls of AI-based patient preference prediction

“…Referring to Goffman’s theory of ‘theatre’,8 they state that the different roles individuals play in their lives cannot be encoded appropriately in an AI. They argue that only next-of-kin can take into consideration this multifaceted nature 7. Although we do agree that humanity is multifaceted, we argue that next-of-kin who predict preferences of incapacitated patients are affected, in principle, by some of the same limitations that AI has.…”

“…Finally, the authors highlight the risk that next-of-kin and clinicians may ‘automatically’ rely on AI predictions 7. We agree that next-of-kin and clinicians may sometimes be affected by ‘AI conformity’ 7.…”

“…Finally, the authors highlight the risk that next-of-kin and clinicians may ‘automatically’ rely on AI predictions 7. We agree that next-of-kin and clinicians may sometimes be affected by ‘AI conformity’ 7. Therefore, it is necessary to instruct clinicians on how to guide difficult discussions, promoting the idea that the resolution of conflicts does not necessarily imply ‘blindly’ accepting the AI predictions 1.…”

“…Tretter and Samhammer reflect on how multifaceted our humanity is and the impossibility for the AI tools proposed to capture this complexity 7. Referring to Goffman’s theory of ‘theatre’,8 they state that the different roles individuals play in their lives cannot be encoded appropriately in an AI.…”