AI could revolutionize healthcare, but it’s only as good as the quality of the data set it is given to learn; while AI is touted as objective, we can make it biased if the samples we give lack diversity. For example, men and women manifest symptoms differently for a variety of diseases, such as cardiovascular disease and diabetes; AI trained to spot symptoms common to men may make it misdiagnose women with the same disease. Read full article here
Education
Overcoming Bias
Kai Cheng’s professor had a brilliant scheme. In his first lecture, he promised that each lecture would feature a “Lie of the Day”. But why? It made his students more attentive and analytical, poring over every detail of his lecture and making sense of why things were true. It was such a powerful teaching method that his students digested his most technical lectures quite easily because they tried so hard to catch his lie. The kicker? There was no lie in that first lecture; he had lied about that too! Read full article here
Merv
August 17, 2022
