THE CHEAT SHEET . What Apple Card’s sexism row tells us about algorithms

Forgive my ignorance, what’s an Apple Card?

Well, it’s a credit card from Apple, backed by financial services giant Goldman Sachs. The tech giant calls it a new kind of credit card. It is stored in your iPhone, and gives you a unique identity number, and allots you money in credit, which you can use for shopping, dining and similar expenses anywhere the card is accepted, just as you use a credit card from, say, American Express or HDFC Bank, in India.

Now, what’s the scandal all about?

On November 7, social media witnessed a heated debate on Apple’s alleged discrimination against women customers after David Heinemeier Hansson — a Danish computer scientist who developed the famous computer application framework, Ruby on Rails — alleged in a series of Tweets that Apple Card gave him a credit limit that was 20 times that of his wife’s credit limits even though both shared the same financial and social conditions. An infuriated Hansson called it a “sexist program”. He tweeted: “No appeals work. Even when she pays off her ridiculously low limit in full, the card won’t approve any spending until the next billing period. Women apparently aren’t good credit risks even when they pay off the... balance in advance and in full.”

Ugh! That’s sexism, plain and simple!

Yup! These allegations were soon endorsed by none other than Apple co-founder Steve Wozniak, who said he could borrow 10 times as much as his wife (whom other cards treat equally) on their Apple Cards. And Hansson tweeted that at Apple’s customer support, nobody was “authorised to discuss the credit assessment process”. A response he received was “...I swear we’re not discriminating, it’s just the algorithm.”

The algorithm, really?

Yes, and no. In case you’re a newbie to the world of algorithmic discrimination, an issue that’s fodder for raging debates across the globe now, where rights and privacy activists, academics and data scientists expose (human) biases that creep into codes that form the backbone of computing programs that governments, companies and other institutions use to find, profile and filter candidates for various schemes.

Interesting!

The old maxim that the code is neutral and algorithms and computers won’t be affected by human biases do not hold water now. As mathematician and data scientist Cathy O’Neil has explored in her book Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy , there is ample proof now, reported and researched out from across the globe, that algorithm-based systems are inclined to discriminate against the poor and the vulnerable, thanks to the way prejudices are built into the systems that produce these programs.

Can’t agree more.

In such systems, interestingly, lower class people, lower caste and minority populations in the Indian scenario, women, children, sexual minorities, and their ilk get filtered out of the ambit of business products and services such as credit cards, loans or venture funds and government welfare doles (pensions, social security monies, unemployment funds, etc). These discriminatory practices trigger and aggravate the gaps between the rich and the poor, the privileged and the underprivileged in society, as Virginia Eubanks observed in her book, Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor .

That’s shocking!

What’s even more shocking is the fact that Apple Card is not the first high-profile product to be called out for sexism or algorithmic discriminations. A report from Reuters showed that products and services such as Facebook Ads, Amazon’s recruiting services, digital assistants such as Apple’s own Siri, Google’s Assistant and Microsoft’s Cortana, Google Images and many others have been accused of being sexist and biased against women. And that shows it’s high time users and consumers demanded more transparency in the way these digital products design their screening systems and processes.

We must, as more and more agencies, governments and companies are using them to evaluate us!

You said it. The learnings from the Apple Card controversy are important for India as well, since we are fast incorporating “digital tools” into our screening and filtering processes for finding beneficiaries of various welfare programs. Even though the New York Department of Financial Services are probing the Apple Card row and Apple Card later changed the credit limit of Hansson’s wife to match his limit, the jury is still out on whether tech companies will be open to the idea of more transparency in their assessment codes and similar processes. To be frank, that’s a billion-dollar question.

A weekly column that helps you ask the right questions

COMMENT NOW