- cross-posted to:
- [email protected]
Racial profiling keeps getting reinvented.
Fuck that.
They then used data on these individuals’ labour-market outcomes to see whether the Photo Big Five had any predictive power. The answer, they conclude, is yes: facial analysis has useful things to say about a person’s post-mba earnings and propensity to move jobs, among other things.
Correlation vs causation. More attractive people will be defaulted to better negotiating positions. People with richer backgrounds will probably look healthier. People from high stress environments will show signs of stress through skin wrinkles and resting muscles.
This is going to do nothing but enforce systemic biases, but in a kafkaesque Gattica way.
And then of course you have the garden of forking paths.
These models have zero restraint on their features, so we have an extremely large feature space, and we train the model to pick features predictive of the outcome. Even the process of training, evaluating, then selecting the best model at this scale ends up being essentially P hacking.
This is just phrenology with extra steps
Bro has the brainpan of a stagecoach tilter
"Imagine appearing for a job interview and, without saying a single word, being told that you are not getting the role because your face didn’t fit. You would assume discrimination, and might even contemplate litigation. But what if bias was not the reason?
Uh… guys…
Discrimination: the act, practice, or an instance of unfairly treating a person or group differently from other people or groups on a class or categorical basis
Prejudice: an adverse opinion or leaning formed without just grounds or before sufficient knowledge
Bias: to give a settled and often prejudiced outlook to
Judging someone’s ability without knowing them, based solely on their appearance, is, like, kinda the definition of bias, discrimination, and prejudice. I think their stupid angle is “it’s not unfair because what if this time it really worked though!” 😅
I know this is the point, but there’s no way this could possibly end up with anything other than a lazily written, comically clichéd, Sci Fi future where there’s an underclass of like “class gammas” who have gamma face, and then the betas that blah blah. Whereas the alphas are the most perfect ughhhhh. It’s not even a huge leap; it’s fucking inevitable. That’s the outcome of this.
I should watch Gattaca again…
Like every corporate entity, they’re trying to redefine what those words mean. See, it’s not “insufficient knowledge” if they’re using an AI powered facial recognition program to get an objective prediction, right? Right?
People see me in cargo pants, polo shirt, a smartphone in my shirt pocket, and sometimes tech stuff in my (cargo) pants pockets and they assume that I am good at computers. I have an IT background and have been on the Internet since March of 1993 so they are correct. I call it the tech support uniform. However, people could dress similarly to try to fool people.
People will find ways, maybe makeup and prosthetics or AI modifications, to try to fool this system. Maybe they will learn to fake emotions. This system is a tool, not a solution.
Goodhart’s law: “When a measure becomes a target, it ceases to be a good measure”
TLDR as soon as you have a system like this people will game it…

Well, that sounds insane.
Woaw, we skipped right from diversity hiring to phrenology hiring without wasting a single beat. Boy has the modern world become efreceint.
Wow. If a black box analysis of arbitrary facial characteristics is more meritocratic than the status quo, that speaks volumes about the nightmare hellscape shitshow of policy, procedure and discretion that resides behind the current set of ‘metrics’ being used.
The gamification of hiring is largely a result of businesses de-institutionalizing Human Resources. If you were hired on at a company like Exxon or IBM in the 1980s, there was an enormous professionalized team dedicated to sourcing prospective hires, vetting them, and negotiating their employment.
Now, we’ve automated so much of the process and gutted so much of the actual professionalized vetting and onboarding that its a total crap shoot as to whom you’re getting. Applicants aren’t trying to impress a recruiter, they’re just aiming to win the keyword search lottery. Businesses aren’t looking to cultivate talent long term, just fill contract positions at below-contractor rates.
So we get an influx of pseudo-science to substitute for what had been a real sociological science of hiring. People promising quick and easy answers to complex and difficult questions, on the premise that they can accelerate the churn of staff without driving up cost of doing business.
Gotcha. This is replacing one nonsense black box with a different one, then. That makes a depressing kind of sense. No evidence needed, either!
All of that being typed, I’m aware that the ‘If’ in my initial response is doing the same amount of heavy lifting as the ‘Some might argue’ in the article. Barring the revelation of some truly extraordinary evidence, I don’t accept the premise.
A primary application of “AI” is providing blackboxes that enable the extremely privileged to wield arbitrary control with impunity.
“okay, okay, hear me out: what if nazi methods, but for getting a job. we could even tattoo their number on their arms. it’s only consequent, we already devide by skin colour”
WTF
It’s completely normal for fascists to promote pseudo-science. Always had been.
Indeed their publication is named after one of the worst pseudo-sciences.
Wait a minute, this sounds suspiciously familiar… I think I heard about the psudoscience of a government measuring the size of people’s heads to try to find out their ethnicity… somewhere in Rwanda… 🧐
I looked for the original article, abstract:
Human capital—encompassing cognitive skills and personality traits—is critical for labor market success, yet the personality component remains difficult to measure at scale. Leveraging advances in artificial intelligence and comprehensive LinkedIn data, we extract the Big 5 personality traits from facial images of 96,000 MBA graduates, and demonstrate that this novel" Photo Big 5" predicts school rank, compensation, job seniority, industry choice, job transitions, and career advancement. Using administrative records from top-tier MBA programs, we find that the Photo Big 5 exhibits only modest correlations with cognitive measures like GPA and standardized test scores, yet offers comparable incremental predictive power for labor outcomes. Unlike traditional survey-based personality measures, the Photo Big 5 is readily accessible and potentially less susceptible to manipulation, making it suitable for wide adoption in academic research and hiring processes. However, its use in labor market screening raises ethical concerns regarding statistical discrimination and individual autonomy.
The PDF is downloadable here: https://scholar.google.com/citations?view_op=view_citation&hl=en&user=2eia4X4AAAAJ&sortby=pubdate&citation_for_view=2eia4X4AAAAJ%3A_FxGoFyzp5QC
I don’t have the time nor the expertise to read everything to understand how they take into account the bias that good looking white men with educated parents are way more likely to succeed at life.
I’m wondering if things like FAS (which can have certain facial characteristics) are muddling the results as well.
one can also get the full paper directly from yale here without needing to solve a google captcha:
I don’t have the time nor the expertise to read everything to understand how they take into account the bias that good looking white men with educated parents are way more likely to succeed at life.
i admittedly did not read the entire 61 pages but i read enough to answer this:
spoiler
they don’t
How many times has Big 5 been debunked yet employers still like it for reasons?
Last time did not end well for about 6 million people…
Boeing CEO: “We’re always innovating, and sometimes we need to boldly embrace the wisdom of the past if it can be re-examined in light of current technology. From now on, our airplane navigation systems will be based on the Flat Earth model. This makes navigation so much more computationally efficient, guys.”
So, the US Supreme Court model then.
Not April fool’s or the onion? What the fuck?
Everyone is kind of focusing on the hiring part, which is incredibly nazi already, but they’re saying for lending too. Fucking yikes.
How revolutionary. They’re going to upend the world of finance by trying to give loans to people who can’t structurally afford to pay them back in all likelihood? Isn’t that the entire playbook for wage slaves to those who are otherwise of sound mind and body?
Maybe they try to just continue with the current (awful)status quo without the psuedoscience? Course then they’d lose out on the fascism.















