Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Viral Tweet About Apple Card Leads to Goldman Sachs Probe

bloomberg.com

Post reply on HN
Print view View original

51–60 of 180 posts

Re: Viral Tweet About Apple Card Leads to Goldman Sachs Probe

#51

Post by jacquesm »

School-Cotton wrote: post #31
jacquesm wrote: post #20

Earlier quoted context omitted.

No, but neither is there anything that would justify giving a spouse such a very different limit especially considering that they file joint tax returns and her credit score is better than his (see article). It would seem that his conclusion is warranted absent evidence to the contrary, the difference is too large to explain in ways that make any sense.

Maybe I am missing something, but wouldn't it make perfect sense if they have dramatically different income? I think even if two people have their property in common (and if the algorithm even knows about that), it is still not unreasonable to believe that there is a higher probability of the one with higher income paying off his or her loans.

I don't think DHH or his wife are a risk with respect to being able to pay off their loans. Any reasonably adept credit scoring algorithm should be able to pick up on that.

Re: Viral Tweet About Apple Card Leads to Goldman Sachs Probe

#52

Post by xenihn »

resonantjacket5 wrote: post #44
xenihn wrote: post #39

Earlier quoted context omitted.

What we really need is the ability to force organizations to expose their models.

Problem is once a model is exposed it can be easily gamed

You don't need to expose the model to figure out how to game it. It's going to be gamed anyways.

Re: Viral Tweet About Apple Card Leads to Goldman Sachs Probe

#53

Post by JumpCrisscross »

School-Cotton wrote: post #31
jacquesm wrote: post #20

Earlier quoted context omitted.

No, but neither is there anything that would justify giving a spouse such a very different limit especially considering that they file joint tax returns and her credit score is better than his (see article). It would seem that his conclusion is warranted absent evidence to the contrary, the difference is too large to explain in ways that make any sense.

Maybe I am missing something, but wouldn't it make perfect sense if they have dramatically different income? I think even if two people have their property in common (and if the algorithm even knows about that), it is still not unreasonable to believe that there is a higher probability of the one with higher income paying off his or her loans.

> wouldn't it make perfect sense if they have dramatically different income?

The implication in citing their joint filing status is they submitted identical incomes. They have the same address, assets, and she had a better credit score. She is also a woman.

Re: Viral Tweet About Apple Card Leads to Goldman Sachs Probe

#54

Post by esotericn »

It strikes me that longer term, we're just going to end up playing a cat and mouse game with people continually inventing different categories that businesses aren't permitted to 'discriminate' on.

Today it might be gender and race. That makes a lot of sense, because the alternative is to further entrench what are basically inheritances.

But aren't we just going to rattle on through and have the algorithms discover (whether we actually realise it or not) that, say, someone diagnosed with X is less creditworthy than someone diagnosed with Y, or that someone bullied in school is less creditworthy, or whatever else?

The whole point of ML is to extract this sort of information from a dataset.

Is it even possible or meaningful to create an unbiased model? Doesn't a model's profit imply bias, whether we currently consider it morally correct or not?

I'd be interested in an argument to convince me otherwise. My view at the moment is basically 'we spend all of this time building models, and then we have to stop using them because they're socially negative/immoral, but for a brief period shareholder value was maximised'?

Re: Viral Tweet About Apple Card Leads to Goldman Sachs Probe

#55

Post by zrail »

School-Cotton wrote: post #31
jacquesm wrote: post #20

Earlier quoted context omitted.

No, but neither is there anything that would justify giving a spouse such a very different limit especially considering that they file joint tax returns and her credit score is better than his (see article). It would seem that his conclusion is warranted absent evidence to the contrary, the difference is too large to explain in ways that make any sense.

Maybe I am missing something, but wouldn't it make perfect sense if they have dramatically different income? I think even if two people have their property in common (and if the algorithm even knows about that), it is still not unreasonable to believe that there is a higher probability of the one with higher income paying off his or her loans.

Credit card limits are usually based on household income.

Re: Viral Tweet About Apple Card Leads to Goldman Sachs Probe

#56

Post by xenihn »

toomuchtodo wrote: post #43
xenihn wrote: post #39

Earlier quoted context omitted.

What we really need is the ability to force organizations to expose their models.

To regulators, sure. Entirely reasonable. To your average Joe? No. This is already done to prevent redlining when originating mortgages. Disclaimer: Work in financial services in risk management, interface with regulators. Opinions are my own.

What's your argument for opposing full exposure, other than the potential impact on your personal livelihood?

Re: Viral Tweet About Apple Card Leads to Goldman Sachs Probe

#57

Post by kyrieeschaton »

I guess it worked for him to throw a hissy fit, but there's absolutely no reason to actually believe they're conditioning on gender.

It could be "it's a new product, we randomly assign credit limits to see how it affects behavior".

It could be "it's a community property state and we're overexposed to this household if we give the second card the same limit as the first".

It could, realistically, be almost anything except for evil bankers deciding to use illegal criteria to underwrite that has a side effect of limiting the amount that can be charged to the account each month (you know, how they actually make money).

Oh, random CSRs don't get a pithy explanation of a multivariate nonlinear underwriting decision to poorly convey to customers? That's kind of precedented!

Re: Viral Tweet About Apple Card Leads to Goldman Sachs Probe

#58

Post by zrail »

xenihn wrote: post #21
zrail wrote: post #14

Earlier quoted context omitted.

Car insurers can point at actuarial data to back up their rating decisions.

Are you saying that there are insurers who don't base their rating decisions on actuarial data? Isn't that a requirement?

Parent comment said car insurer.

Re: Viral Tweet About Apple Card Leads to Goldman Sachs Probe

#60

Post by tempsy »

bobobob420 wrote: post #47

Highly doubt Goldman Sachs included gender discrimination in their risk model...do regulators get to see the risk models credit companies use to determine how much credit one can use?

It's well known there are proxies are you can't use when building a credit risk model e.g. zip codes are off limits because it can be used to a proxy for ethnicity.

Intentional or not, it's possible they could have used something that proxied for gender.

Post reply on HN