ITIF Defends Amazon’s Rekognition Against ACLU Racial Bias Claims

The American Civil Liberties Union’s “flawed” methodology in criticizing Amazon’s Rekognition for racial bias was a setback for legitimate use of facial identification (see 1807270040), blogged Information Technology and Innovation Foundation Vice President Daniel Castro and Research Assistant Michael McLaughlin Monday. “It is unclear what error rate and level of bias groups such as the ACLU are willing to accept,” they wrote. “The standard should not be perfection, but rather better than the rates humans achieve today. And by that metric, facial recognition technology is clearly a positive step forward.” Amazon says ACLU’s results could be improved by following best practices on boosting confidence thresholds from 80 to 95 percent.