Applying for Your Next Job May Be an Automated Nightmare

Applying for Your Next Job May Be an Automated Nightmare

If you think looking for a job is already daunting, anxiety-riddled, and unpleasant, just wait until the algorithms take over the hiring process. When they do, a newfangled “digital recruiter” like VCV, which just received $1.7 million in early investment, hopes it will look something like this:

First, a search bot will be used to scan CVs by the thousands, yours presumably among them. If it’s picked out of the haystack, you will be contacted by a chatbot. Over SMS, the bot will set an appointment for a phone interview, which will be conducted by an automated system enabled by voice recognition AI. Next, the system will ask you, the applicant, to record video responses to a set of predetermined interview questions. Finally, the program can use facial recognition and predictive analytics to complete the screening, algorithmically determining whether the nervousness, mood, and behavior patterns you exhibit make you a fit for the company.

If you pass all that, then you will be recommended for an in-person job interview.

It should go without saying that the process described above—which required no exaggeration and is a straight description of the four-step process outlined on VCV’s website, under the “How the magic is being done” section—probably sounds like a certain kind of hell for the vast majority of job seekers.

For one thing, a lot of facial and voice recognition software is notoriously faulty and bias-prone. Some researchers and critics argue that any claims that algorithms can make judgements based on perceived mood or facial reaction (sometimes called “affective computing”) amount to little more than modern day phrenology. For another, all of this serves to render the job application process even more opaque, impersonal, and impenetrable to the average job hunter. VCV already counts large companies like Citibank, l’Oreal, Danone, and PricewaterhouseCoopers among its clients.

“Human biases have long plagued hiring, and any claim that machine learning algorithms alone can fix that is bogus,” Aaron Rieke, the co-author of a report published in December 2018 called Help Wanted: An Examination of Hiring Algorithms, Equity, and Bias, told me in an email. “It has been reported that VCV uses facial recognition to identify candidates’ “mood” and “behavior patterns” to help recruiters assess ‘cultural fit.’ This raises all kind of red flags.”

Illustration for article titled Applying for Your Next Job May Be an Automated Nightmare
Screenshot: VCV

VCV, which did not respond to a request for comment, is far from alone here. A growing suite of startups is pitching AI-driven recruitment services, promising to save corporations millions of dollars throughout the hiring process by reducing overhead, to pluck more ideal candidates out of obscurity, and to *reduce* bias in the hiring process. Most offer little to no evidence of how they actually do so. VCV’s much-larger competitor, HireVue, which has raked in a staggering $93 million in funding and is backed by top-tier Silicon Valley venture capital firms like Sequoia, is hocking many of the same services. It counts 700 companies as its clients, including, it says, Urban Outfitters, Intel, Honeywell, and Unilever. AllyO, which was founded in 2015, and “utilizes deep workflow conversational AI to fully automate end to end recruiting workflow” has $19 million in backing.

Last year, in an article titled Recruiting Generation Z? You’ll Want to Use These, Inc. Magazine rounded up the new companies using “AI, gamification, machine learning” and other technologies to hire the youngest job seekers. Ryan Jenkins, a self-described “Millennial and Generation Z speaker and generations expert” ranks HireVue as the #2 new hiring tool, just behind Pymetrics, which “uses neuroscience games and bias-free artificial intelligence (AI) to predictively match people with jobs where they’ll perform at the highest levels.” #4 is Mya / Wade & Wendy, which “offer chatbots that automate the process from resume to interview.”

Just how expansive these operations are right now is unclear, despite the fact that these digital recruiters tout contracts and partnerships with Fortune 500 companies. Rieke says information is scarce about how widely major recruiters have embraced automated hiring practices, bots, and facial recognition. But it’s clear the trend is attracting ample interest, investment, and encouragement from the business community.

“AI in human resources is cost-effective and better for business overall,” Barbara Van Pay, the CEO of SmartHR Consultancy, writes in Entrepreneur Magazine. Van Pay points to a 2016 Society of Human Resource Management survey that found the average cost-per-hire was $4,129. AI, she reasons, could whittle that figure away. “With many of the AI recruitment and Human Resources programs available offering tailor-made packages on a monthly, quarterly, and yearly subscription basis, it’s not hard to see that you can save a pretty hefty penny by transitioning to AI technology solutions.” To wit, HireVue’s website claims that “Hilton cut time-to-hire nearly 90% with Hirevue assessments,” ostensibly saving it a great deal of money.

The goal here is obvious enough—promise corporations they can spend less money on human headhunters and recruiters, less time manually poring over applications, and generally fewer resources on the hiring process by outsourcing all of the above to machines. But these startups risk offering a prime example of shitty automation—an automated product adopted in the name of saving money, that risks, in the end, just making everybody’s lives worse. In this case, the automation is designed to benefit one side of the equation almost exclusively: the employer.

Illustration for article titled Applying for Your Next Job May Be an Automated Nightmare
Screenshot: HireVue

The added hoops and inscrutable, bot-addled process will probably make matters worse for applicants. Think, for a second, about how much you appreciated the last robocall you accidentally picked up and suffered through 2.5 seconds of. Now consider what it would be like if that robocaller were conducting your preliminary job interview. Job interviews are about to be laced with a lot more expletives, at least. But bias and discrimination remain a distinct problem area.

“There are layers of concerns here,” Rieke tells me. “Facial recognition technology is often less accurate for women and darker-skinned people. Even assuming companies like VCV can evaluate ‘mood’ or ‘behavior,’ it’s not clear how that should help a recruiter assess candidates. If the ‘right kind’ of mannerisms are derived from a company’s current, homogeneous workforce, diverse candidates could be penalized.”

Not only is that hugely problematic with regards to people who might be passed over for work by biased AI—but it’s not hard to imagine a future in which predictive hiring software gives way to some hefty discrimination lawsuits, too. Founders and developers can claim their algorithms reduce bias all they want; until they have transparently demonstrated proof, we should default to skepticism. (Last year, the founders of Predictim, an AI-based platform designed to screen applicants for babysitting jobs, said their product was absolutely not biased, yet when I used it myself, the preliminary results looked pretty racist to me.)

Finally, it’s not really clear that the service works. Yes, there may be short term gains as companies cut operating costs by offloading tasks to an AI service—but there’s a reason the hiring process has typically been hands on; it’s a pretty good way for the two parties to tell if they’ll enjoy working together. Dubiously derived data points about performance and mood seem like a poor substitute, though time will tell.

“We need a lot more information about how systems like these are designed and tested,” Rieke says. “Until that happens, I’m extremely skeptical. The hype is way ahead of the facts.”


So now job seekers are going to have to pretend to be enthusiastic and happy to talk to a f’ing robot?

Using a face-scanning algorithm to determine if someone will be a good “cultural” fit is pretty damn dystopian. 

And get a load of this jargon-vomiting grifter: