Goldman Sachs needs to change AI bias to win back Apple Card users’ trust

Goldman Sachs needs to invest in tools to mitigate bias in the Apple Card application process. Apple

Since the news of Apple Card Gender Discrimination broke last week Goldman Sachs has been in damage control mode.

The financial institute, which has partnered with Apple to issue the new credit card, insists that its approval process does not favor male candidates, as reported by users. It’s safe to say that the public isn’t buying Goldman’s claim of innocence; same co-founder of Apple Steve Wozniak noted his wife’s considerably lower Apple Card spending limit.

SEE ALSO: Amazon now offers a credit card for customers with bad credit

“What Goldman Sachs didn’t take into account is that machine learning algorithms excel at finding latent features in data,” explained Lux Search analyst Cole McCollum. “These are the features that are not directly used in training a machine learning model, but are inferred from other features that are.”

McCollum told Observer that this type of bias found in historical datasets is one of the biggest challenges in implementing machine learning. This is also why companies should explore artificial intelligence (AI) bias detection and explanation tools that can help mitigate potential discrimination. “In the case of Apple Card spending limits, even though gender was not specifically considered (as Goldman Sachs claims), other related features in the dataset may still embed these biases. and lead to unfair decisions,” he said.

Apple’s security-focused Titanium Card is meant to be an innovative consumer credit product. While Goldman Sachs’ application process still requires technical tweaks, the Apple Credit Card could come out the other end thanks to its enticing perks, like a daily cashback feature.

“This incident should serve as a wake-up call to companies to invest more in interpretability and algorithm testing in addition to training executives on the subtle ways bias can seep into AI projects. and machine learning,” McCollum concluded.

How Goldman Sachs Can Regain User Trust After Apple Card Discrimination

Gladys T. Hensley