Black Panther Face Scorecard
February 2018 Results
The Gender Shades research used 1270 faces to reveal IBM, Miscrosoft, and Face++ were better at guessing the gender of male faces than female faces and especially struggled on darker female faces.
The release of the Black Panther movie which centers a cast of dark-skinned men and women provided an opportunity to see how well Artificial Intelligence software from IBM, Face++, Microsoft, Kairos, and Clarifai detected the faces and predicted the age, gender, and race of the main characters of the movie on opening week.
While 9 images are not enough to draw scientific conclusions, we hope the results encourage more research on using artificial intelligence for facial analysis across a range of human traits. View Results Below.
*Age is considered correct if the prediction is within 2 years of age of actor in photo