An AI ethics committee is reviewing a proposal to use facial recognition AI to verify customer identity for bank account access. Which risk requires HIGHEST priority in the ethics review?
Select an answer to reveal the explanation.
Short Explanation
Here's the deal — you need to prioritize the systematic lower accuracy for darker-skinned individuals, because it can lock people out of their bank accounts based on race. That's a discrimination and regulatory risk, not just a minor technical issue.
Full Explanation
The correct answer is the systematic lower accuracy for darker-skinned individuals due to training data that skewed toward lighter-skinned faces, leading to discriminatory account access outcomes. Documented performance disparities of facial recognition systems across skin tones create discriminatory access outcomes: darker-skinned customers may be denied access at higher rates, constituting discrimination in the provision of financial services. This is the highest-priority ethics and regulatory risk. The false rejection rate for customers in poor lighting conditions affects all users and is a usability concern. The computational cost of real-time facial recognition at scale is an operational concern. The customer preference for traditional PIN-based authentication over biometric methods is a user experience consideration. Racial performance disparities in a financial access tool combine discrimination risk, regulatory risk, and fundamental rights concerns.