When the HR department is a robotic phrenologist: "face-scanning algorithm" gains popularity as a job-applicant screener

Hirevue is an "AI" company that companies contract with to screen job applicants: it conducts an hour-long videoconference session with applicants, analyzing their facial expressions, word-choices and other factors (the company does not actually explain what these are, nor have they ever subjected their system to independent scrutiny) and makes recommendations about who should get the job.

The system is one giant red flag, from the junk science of microexpressions to the deployment of discredited "personality tests" as a way to evaluate candidates. The company claims that it uses (nonspecific) internal tests to ensure that racial bias (and other forms of bias) are not creeping into its models, but it does not say that it ever follows up to determine whether the candidates it rejects go on to be successful employees somewhere else, nor whether the candidates it recommends perform well once they're hired. This lack of followup is a bright line test for distinguishing rigorous machine learning from "weapons of math destruction."

Despite this, Hirevue has a ton of high-profile corporate customers, including Hilton, Unilever and Goldman Sachs.

Hirevue's statements in defense of their products read like a palmist or tarot reader explaining the mystical basis for their certainty that they are able to know your future. And since the corporations that use their service have no way to know whether they rejected a superior candidate on the basis of the black-box machine judgments doled out by Hirevue, they give it glowing reviews.

HireVue said its system dissects the tiniest details of candidates responses their facial expressions, their eye contact and perceived enthusiasm and compiles reports companies can use in deciding whom to hire or disregard.