Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5

Predictive AI is bolstering physiognomy instead of discrediting it. Outrage begins...

#1
C C Offline
Is your face gay? Conservative? Criminal? AI researchers are opening a kettle of worms
https://thebulletin.org/2022/05/is-your-...st-heading


I likewise want to project something "insidious" into this area of research, because -- as a citizen of the West rather than Russia or China, I share similar ideological biases and worries. But I want to be on the side that is privately aware that the ostracism of it is potentially resting in engineered BS, done for "the good of society". Rather than figuratively being a member of the choir who totally buys that the ensuing damnation does not stem from protecting unalterable "Church" doctrine.

This is an example of how the science establishment, in conjunction with society, does indeed have its own preexisting preferences for how it wants things to be, and that these (at least in the short term) are more important to uphold than objectivity, neutrality, facts/reality, etc.

Physiognomy is supposed to be wholly pseudoscience, but predictive AI is undermining that traditional evaluation slash belief: "...facial recognition technology can judge a persons’ political orientation accurately 68 percent of the time even when controlling demographic factors [...] In 2017, Kosinski published another work showing that a neural network trained on facial images could be used to distinguish between gay and straight people. Surprisingly, the experiments using the system showed an accuracy of 81 percent for men and 74 percent in women. "

So using the motivated reasoning of its own cultural preferences to reach dice-loaded conclusions, the establishment is going all out to portray this as irrelevant for moral reasons ("but what was the value of the question the study asked?"). And if that doesn't succeed -- then interpret the data according to the establishment's own cognitive filters and in-house personal biases, as well as splash great hyperboles of unbridled skepticism upon it.

IOW, even if this work could be exposed as flawed through impartial scrutiny, we will never know that for sure because all the guns are out to kill it through any means possible. Self-righteousness and holy condemnation tops non-motivated procedure or due process.

Remember this the next time you see scientists stepping outside their own sphere of expertise, entering the sphere of policy to play philosopher kings (scientism), and issuing great proclamations about compliance to _X_ government's trendy norms or that we should conform to their own. They're human: they have personality flaws, career-serving agendas, and intellectual frailties. (They also have to cater to the policies of the administrators of their own departments and institutions, who uncritically jump on the bandwagons of every virtue-signaling and funding source bandwagon that comes along. And satisfy what science publishers most reward for in the "publish or perish" environment of the system).
Reply




Users browsing this thread: 1 Guest(s)