Why algorithms won’t replace human interaction during the hiring process

We capture and take so many photos on our smartphones these days, most people have turned to some sort of app – from Apple’s native Photos to Google Photos – to help organize and store them all. Most of those apps also have some sort of AI software that tries to help make that easier.

Take Google Photos, for instance. The app’s Assistant makes suggestions about screenshots and photos to archive, so you can clean up your library, offers daily tips, and even puts together collages, animations, and short videos based on similar shots and content using geotagging information and facial recognition programming. Often, those little creations are a fun surprise. Or, in the case of reddit user Alex Harker, it can alter reality enough to turn your friend’s head into a giant, lurking forest monster.

Source: Alex Harker/Imgur

As Alex explained on reddit:

“I literally took like 3 pictures, one with them in, and two without them. And for some bizarre reason Google Assistant offered me a really strange panorama of the 3 photos spliced together.”

The original photos – source: Alex Harker/Imgur

Algorithms, computers, and machine learning still can’t match humans

As Harker’s reddit post quickly picked up steam online, commenters were quick to notice how well the app did it’s photoshop job — which illustrates perfectly the shortcomings of the technology when it compares to humans. Humans, after all, quickly notice the absurdity of the created image; the algorithm has no way to recognize or reconcile its mistake.

Over the past five years, the use of algorithms in recruiting and hiring has quickly grown into common practice, especially at larger corporations. And no wonder, when companies are overloaded with resumes and applications and sorting through them all manually takes an immense amount of time and effort.

Machine learning and AI algorithms can help sort through the immense amount of data that’s created during the hiring and recruitment process. Predictive workforce analytics can help you identify people who are the right cultural fits or the right personality for a particular role or team.

But at the end of the day, these machines are still machines and so much of hiring requires human analysis, human interaction, and human intuition.

The best role for these algorithms is not as replacement, but as enhancement – it is far more likely that these programs simply make human recruiters jobs easier, rather than replace them.

In between is the applicant tracking system with a lot of buzz words, categories and filters that don’t work.  Nothing can replace human interactions. There are no shortcuts when hiring qualified personnel, you have to do the work.” – Bridgitte Hyacinth, Author of The Future of Leadership: Rise of Automation, Robotics and Artificial Intelligence

The future of hiring algorithms is likely similar to that of marketing automation: these tools haven’t replaced the sales-and-marketing departments. They’ve simply helped make them more efficient and effective at their jobs.

By | 2018-01-31T20:58:10+00:00 January 31st, 2018|Uncategorized|Comments Off on Why algorithms won’t replace human interaction during the hiring process

About the Author: