Awesome, not awesome.
#Awesome
"For example, grocers can apply analytics to granular club card data in ways that inform, reward, or incentivize healthier food purchasing options. Doing so can increase access to essential needs, of particular value to vulnerable populations living in food deserts. Analogously, insurers can use policyholder data not only to build actuarial models, but also to use pricing to incent safer risk behaviors and risk management on the part of policyholders. Online retail platforms can use their data to inform customers when prices for essential goods are excessive. These are all examples of using data and AI in ways that help rather than exploit customers." - Mark E. Bergen, Shantanu Dutta, James Guszcza, and Mark J. Zbaracki Learn More from Harvard Business Review >
#Not Awesome
"Unfortunately, however, bias not only affects how individuals perceive the world around them, it also influences the datasets we use to build models. Observational datasets that store patient features and outcomes often reflect the underlying bias of health care providers; e.g., certain treatments may be preferentially offered to those who have high socioeconomic status. In short, algorithms can inherit our own biases. Making personalized medicine a reality is therefore predicated on our ability to develop and deploy unbiased tools that learn the patient-specific decisions from observational clinical data." - Collin Stultz Learn More from MIT News >
|
