Latest / The Science of Leading / AI in Hiring: Balancing Efficiency and the Human Touch

Transcript

AI in Hiring: Balancing Efficiency and the Human Touch · The Science of Leading

• 0:00
  Welcome back to The Science of Leading,
• 0:01
  everyone.
• 0:02
  I'm Claire Monroe, and Edwin Carrington is here with me,
• 0:05
  as always.
• 0:05
  Today, we're diving into something that's...
• 0:08
  kind of everywhere right now-AI in hiring.
• 0:11
  And Edwin, I keep seeing two totally different takes on this.
• 0:14
  Some people are like, "Finally!
• 0:16
  A way to remove bias," and others are just,
• 0:19
  like, deeply anxious that algorithms are making things worse.
• 0:22
  You've seen both sides play out,
• 0:23
  right?
• 0:23
  Absolutely.
• 0:25
  And there's new research that captures that exact tension.
• 0:28
  One study-a multidisciplinary survey-really lays it out.
• 0:33
  There's hope, real hope, that algorithmic hiring could reduce bias.
• 0:37
  The idea is, machines don't bring gut instinct or unconscious prejudice
• 0:42
  into the process.
• 0:43
  In theory, it's more objective.
• 0:43
  Right...
• 0:47
  so, like, no more "this candidate just feels right" based on nothing?
• 0:51
  We've talked about that kind of bias before-like,
• 0:53
  hiring based on vibes.
• 0:53
  That's the theory,
• 0:56
  yes.
• 0:57
  But it's not the whole story.
• 0:59
  The same research shows a serious risk: if the data used to train these
• 1:03
  tools is biased-which, let's be honest,
• 1:06
  it often is-then the algorithm doesn't erase bias.
• 1:09
  It learns it.
• 1:10
  And then...
• 1:11
  it scales it.
• 1:13
  You're not solving discrimination.
• 1:15
  You're automating it.
• 1:15
  Oof.
• 1:18
  Yeah, it's like-bad data in, bad decisions out...
• 1:21
  but faster and everywhere.
• 1:23
  Why do you think we keep hearing these two-kind of competing-narratives?
• 1:26
  Like, "AI will save hiring," vs "AI is just a bias machine"?It's human
• 1:31
  nature to look for simple answers.
• 1:33
  We want to know: is AI good or bad?
• 1:36
  But this is one of those "it depends" situations.
• 1:39
  AI isn't inherently fair-or unfair.
• 1:43
  It depends entirely on how it's trained,
• 1:46
  deployed, and monitored.
• 1:47
  And that's where hiring leaders really need to focus.
• 1:47
  So,
• 1:51
  okay-given all the promise and the pitfalls,
• 1:53
  how do you think about it?
• 1:55
  Like, where do you land on this?
• 1:55
  I see massive potential-when it's done
• 2:00
  right.
• 2:00
  AI can make hiring more efficient,
• 2:03
  more consistent, and in some cases,
• 2:05
  less biased.
• 2:06
  But we need clear guardrails.
• 2:08
  We can't just assume the tech knows what "fair" looks like.
• 2:11
  If we don't stay actively involved-checking,
• 2:15
  adjusting, asking questions-we risk doing real harm.
• 2:15
  Yeah...
• 2:20
  it's like we're walking a tightrope.
• 2:21
  Promise on one side, danger on the other.
• 2:24
  And that sets us up perfectly for this next point,
• 2:26
  which I love-because it's where the real nuance lives.
• 2:29
  Like, how do you actually blend AI's strengths with the human side that
• 2:33
  makes hiring...
• 2:34
  you know, actually work?
• 2:34
  That's the key question.
• 2:38
  Most of the smart guidance out there-from Forbes HR Council to SocialTalent
• 2:43
  to the practitioners in the trenches-says the same thing: let AI handle
• 2:48
  the repetitive, time-consuming stuff.
• 2:50
  But keep the meaningful moments human.
• 2:50
  Right,
• 2:54
  like-okay-scheduling interviews,
• 2:56
  screening CVs, maybe answering FAQs through a chatbot.
• 2:59
  That stuff can eat up a recruiter's entire day.
• 3:02
  But if you automate it, you actually free up time for the parts that
• 3:05
  matter.
• 3:05
  Real convos.
• 3:06
  Real connection.
• 3:06
  Exactly.
• 3:09
  And here's what's important: research shows that what candidates want
• 3:13
  most is feedback, transparency,
• 3:16
  and personalization.
• 3:17
  That doesn't come from a bot.
• 3:20
  Automation should create the space for those high-impact moments-not
• 3:25
  eliminate them.
• 3:25
  Ugh-okay, story time.
• 3:28
  Early in my career, I worked somewhere that tried to automate,
• 3:32
  like, everything.
• 3:33
  No calls, no updates, just templated emails.
• 3:37
  It was efficient...
• 3:38
  but also kinda brutal.
• 3:40
  People started saying stuff like,
• 3:42
  "I felt invisible." It was rough.
• 3:45
  So we added short check-in calls-just 5 minutes!-and the change in candidate
• 3:50
  feedback was unreal.
• 3:50
  That's a perfect example.
• 3:53
  And it's exactly what companies like Accenture and BCG are building into
• 3:57
  their models.
• 3:58
  They call it High Touch, High Tech.
• 4:00
  The tech helps with speed-but the touch is where the trust gets built.
• 4:04
  You map out the process so that every human moment delivers maximum impact.
• 4:04
  And
• 4:10
  it's not just touchy-feely stuff,
• 4:11
  either.
• 4:12
  That kind of attention drives better outcomes.
• 4:14
  Better referrals.
• 4:15
  Better offer acceptance.
• 4:17
  Like we said in our last episode-automation isn't the enemy.
• 4:20
  But if you're not intentional about what you automate,
• 4:23
  you're gonna miss the mark.
• 4:23
  Exactly.
• 4:25
  If empathy gets lost in the name of efficiency,
• 4:28
  the long-term costs are real.
• 4:30
  But with the right design?
• 4:32
  You get both.
• 4:33
  Speed and experience.
• 4:33
  Okay, but now comes the hard part-if you do scale
• 4:38
  up the tech, how do you keep things fair?
• 4:40
  And clear?
• 4:41
  Like...
• 4:41
  how do you stop it from turning into this mysterious black box?
• 4:41
  This is
• 4:46
  where most organizations get tripped up.
• 4:48
  It's not enough to just say, "We use AI,
• 4:51
  but trust us." Fairness, empathy,
• 4:53
  and transparency need to be built in-and checked for-constantly.
• 4:53
  Okay,
• 4:59
  so like, what does that actually look like?
• 5:00
  What should teams be focusing on,
• 5:02
  day to day?
• 5:02
  Three big things: First,
• 5:05
  bias auditing.
• 5:06
  You have to test your tools to make sure they're not replicating old
• 5:09
  patterns.
• 5:10
  The EU AI Act is really clear about this-it's not optional.
• 5:14
  Second, transparency.
• 5:16
  Candidates should know when AI is being used,
• 5:19
  and how.
• 5:20
  If a hiring manager can't explain it?
• 5:23
  That's a red flag.
• 5:24
  And third-explainability.
• 5:24
  That one's huge.
• 5:28
  Because even if you're not doing anything shady,
• 5:31
  it still feels shady when no one can explain what's happening.
• 5:34
  And the privacy side too-like,
• 5:36
  if people feel like their data's just...
• 5:38
  floating around?
• 5:39
  That's a trust killer.
• 5:39
  Absolutely.
• 5:42
  Strong organizations-Accenture,
• 5:45
  BCG, others-they don't leave that stuff to chance.
• 5:48
  They do bias checks.
• 5:50
  They communicate clearly.
• 5:51
  And they make sure there's always a human involved when it counts most.
• 5:51
  Yeah,
• 5:56
  but I feel like the temptation is real,
• 5:57
  though.
• 5:58
  Like, "Hey, this tool's fast-let's just roll with it." But if no one
• 6:01
  knows how it works...
• 6:02
  you lose trust.
• 6:04
  Fast.
• 6:04
  Exactly.
• 6:06
  When the process becomes a black box,
• 6:09
  people pull away-candidates, hiring teams,
• 6:12
  regulators.
• 6:12
  I've seen it.
• 6:13
  The organizations that lead are the ones who stay transparent,
• 6:17
  who keep humans in the loop, and who treat experience as a strategic
• 6:21
  asset-not just a checkbox.
• 6:21
  Yeah,
• 6:24
  because let's be real-you can have the flashiest tech stack in the world,
• 6:27
  but if people don't trust the process?
• 6:29
  It's game over.
• 6:30
  Same old problems, just in a shinier container.
• 6:30
  Exactly right.
• 6:35
  AI should amplify what people do best-not erase it.
• 6:38
  The companies who get this right don't just use AI.
• 6:41
  They design with intention.
• 6:43
  They lead with integrity.
• 6:43
  Okay,
• 6:46
  that feels like a solid place to land.
• 6:48
  If you're wondering how to actually put some of this into action-like,
• 6:52
  practically-OAD has free tools you can try.
• 6:54
  Behavioral assessments, bias-reducing strategies...
• 6:58
  all of it.
• 6:58
  Just head to o-a-d dot a-i.
• 7:01
  It's simple, and it really helps you get clearer on fit.
• 7:01
  It's been great
• 7:05
  as always, Claire.
• 7:06
  And thank you to everyone listening.
• 7:08
  Just by being curious about this stuff-you're already moving toward better
• 7:12
  decisions.
• 7:12
  We'll be back soon with more on the science-and the humanity-behind
• 7:17
  smarter hiring.
• 7:18
  See you next time on The Science of Leading.
• 7:18
  Take care,
• 7:21
  Claire.
• 7:21
  Take care, everyone.