When it comes to the human vs. AI in recruiting question, intelligent tools haven't replaced the qualities only people bring. Both have a role.
AI tools, often referred to as intelligent tools, are marketed as a means to help recruiters choose the best candidate. As more companies use such tools, recruiters may wonder when to rely on them to help pick a candidate for a position and when to rely on personal gut instincts and experiences to select new employees.
To solve this conundrum, three analysts shared their views on humans vs. AI in recruiting.
AI will miss great candidates
Stacia Sherman Garr, co-founder and head analyst, RedThread Research
Gut instinct has the potential to pick up on things or measure things that you may not be able to measure through [technology alone].
Gut instinct is not just how you feel, but also what recruiters should be doing beyond just measurement, and that is looking at the bigger context of what's happening. [AI tools] give us an opportunity to understand and appreciate what comes to us, naturally -- or through gut instinct -- but then to question it and really dig into our assumption.
AI is an advanced mathematical model, and those models need lots of data to get trained on. They're going to come up with a recommendation for someone who would fit, most of the time.
But are you going to miss out on people who either have new or different capabilities than what you might have written in a job requisition? Yes. And that is where you have to have some of that balance and ability to choose somebody who's a nontraditional candidate fit. AI can be a really good and powerful tool, but we can't lose that which makes us human.
Using data and deep analysis can help us understand a highly complex situation, because people are only able to process a fairly limited amount of data. But we shouldn't be throwing out our gut instinct and what makes us human just because a mathematical model told us so, so I think it's a balance of the two.
AI tools are still immature
John Sumser, principal analyst, HR Examiner
What recruiters do in their job is just like anybody else in their job; they learn. They apply their learning into a more comprehensive picture, and that's the work that's called analysis. Just like if you were a tree cutter and went out into the woods to figure out which trees to cut, you would be able to tell which trees to cut based on look and feel and would develop that capacity to tell problems that result from a bad judgment.
In other words, [the ability to choose a good candidate is] a skill that can be developed through usage. What people do is they get better at making their decisions because they develop a skill and a capacity for doing it, and from the outside, it looks fast. From the inside, it wasn't fast to get there.
What intelligent tools do is they make some equation out of the job description and out of résumés and then they look for matches. If you have a recruiter who relied exclusively on an intelligent tool to make hiring decisions, it'd have to be a very unique case since most jobs are not a perfect match between a résumé and a job description. That's where bias starts to creep in when you start to believe that there's a best person for the job.
It's not that AI is bad; it's not that machine learning is bad; it's not that machines are not capable of a lot of things. They're just not as good as you'd think. AI can predict who would be good for the job 80% of the time. And so that's really good information to have.
[As a metaphor,] if there's an 80% chance of rain tomorrow, you take an umbrella. Yet you're not surprised when it doesn't rain. So we have to learn how to apply that way of thinking to decision-making about people.
The human vs. AI in recruiting question
Dean Da Costa, recruiter, Lockheed Martin
Technology [such as AI tools] can help us gather research, find candidates, scrape the candidates, enhance them and create good outreach.
The actual screening, which means talking to and interviewing [candidates], relies much more on gut, or an educated gut, given that we will have a lot of information already, and in an interview are really looking for validation and red flags.
If you're trying to decide which person to hire, you've got to just go with your gut: 'Which one did I like better?' But usually there's more to it; these are not pure emotions. Your gut reaction, no matter how hard you try to place it, is going to be impacted by your experiences.
AI provides you the data, but it does not provide you with [a holistic view of the candidate] behind the data.