October 27, 2025 • 37 mins
In this episode, guest hosts Pauline James and David Creelman welcome culture strategist and author Dr. Jessica Kriegel for a conversation on how leaders can use AI thoughtfully to enhance, not replace, the human side of work.
In this lively discussion, Jessica unpacks what’s really happening inside organizations as AI, culture, and change management collide. She explains why this is the “moment for change management,” how leaders can create experiences that shift employee perceptions, and why the illusion of control is the biggest barrier to transformation. Jessica also shares fascinating examples, from manufacturing firms using AI for data insight to companies experimenting with “culture-measuring” chatbots.
Key Themes
- How leaders can move beyond “the action trap” to shift employee perceptions through meaningful experiences
- Real-world AI use cases that reveal both opportunity and risk, from sales analytics to culture analytics
- The fine line between empowerment and surveillance in workplace technology
- Why “surrender” is the surprising secret to stronger leadership and sustainable change
Why HR Professionals Should Listen
For HR leaders navigating the dual revolutions of AI and culture, Jessica Kriegel offers a rare combination of candor, intelligence, and humor. She speaks the language of both strategy and humanity, reminding us that change doesn’t start with systems or data, but with what people believe.
Her upcoming book, her co-authored with Joe Terry, Surrender to Lead, which will be released in January 2026, expands on her insights. Jessica invites leaders to let go of the illusion of control and focus on what really drives results: their presence, mindset, and the ability to create meaningful experiences for employees which inspire alignment and action.
Feature Your Brand on the HRchat Podcast
The HRchat show has had 100,000s of downloads and is frequently listed as one of the most popular global podcasts for HR pros, Talent execs and leaders. It is ranked in the top ten in the world based on traffic, social media followers, domain authority & freshness. The podcast is also ranked as the Best Canadian HR Podcast by FeedSpot and one of the top 10% most popular shows by Listen Score.
Want to share the story of how your business is helping to shape the world of work? We offer sponsored episodes, audio adverts, email campaigns, and a host of other options. Check out packages here.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
SPEAKER_03 (00:02):
Welcome to the HR Chat Show, one of the world's
most downloaded and sharedpodcasts designed for HR pros,
talented defects, techenthusiasts, and business
leaders.
For hundreds more episodes andwhat's new in the world of work,
subscribe to the show, follow uson social media, and visit
hrgazette.com.
SPEAKER_04 (00:25):
Hello and welcome to the HR Chat Podcast.
I'm Pauline James, founder andCEO of Anchor HR and associate
editor of the HR Gazette.
It's my pleasure to be your hostalong with David Krillman, CEO
of Krillman Research.
We're partnering with the HRChat Podcast on a series to help
HR professionals and leadersnavigate AI's impact on
(00:46):
organizations, jobs, and people.
SPEAKER_02 (00:48):
In this episode, we are speaking with Jessica
Kriegel.
Jessica is Chief StrategyOfficer at Culture Partners in
California.
She has a strong background inthe world of technology, having
spent a decade in changemanagement at Oracle.
Since then, Jessica has become arecognized expert in
organizational culture andstrategy.
Her work has been featured byCNBC, CNN, The New York Times,
(01:10):
the Wall Street Journal, andForbes, among others.
We connected with her as we wereeager to hear the use cases for
AI that she has seen amongst herclients and their impact, as
well as to draw on her changemanagement expertise.
We're also happy to share thatJessica has a book coming out in
January.
It's called Surrender to Lead.
So we look forward to that.
SPEAKER_04 (01:30):
Jessica, just to kick us off, can you tell me
about the advisory board thatyou were on where you met David?
SPEAKER_01 (01:37):
Yes, I would love to.
We're both on the WorkforceInstitute, which is a think tank
that's part of UKG that focuseson the future of work, what are
best practices to get results,what are the needs of employers
and employees, and where dothose needs meet and where did
they diverge.
So it's been an absolutepleasure meeting David there.
(01:59):
Thank you.
SPEAKER_04 (02:00):
I see you do a lot of podcasts, keynotes.
I'm interested to hear what HRaudiences are asking you to
speak to right now, what they'remost interested in.
SPEAKER_01 (02:12):
Um, right now, the biggest talk track I have to
talk about is change management.
This is the moment for changemanagement.
Interestingly, I'm not at HRTech, but at this year at HR
Tech.
It was standing room only in thechange management session put on
by Deloitte.
That is the big question markfor people because of the nature
(02:33):
of the speed of change and thenumber of changes happening in
any organization at one time.
And that's a verypeople-oriented challenge.
Thank you.
SPEAKER_04 (02:42):
With that, is there anything additionally you feel
that they should be focusing onthat you like to draw, pull
their attention to?
SPEAKER_01 (02:51):
Yeah, my opinion, my expertise is around how do you
get the hearts and minds of youremployees aligned with the
organization's objectives.
That's really the key in mymind.
And with change management, whatit typically looks like is a
bunch of training and a bunch ofPR and a bunch of trying to get
(03:15):
people on board with an ideawith communication and rousing
speeches in town halls, multipleemails and systems that'll uh
allow you to implement thatchange, whatever the change is,
right?
The change could be, oh, we'regoing to AI.
The change could be we need todo a digital implementation, or
the change could simply be weneed to grow by X percent.
And that's a change from theamount that we've been at up
(03:38):
until this point.
And a lot of leaders get stuckin the action trap of what do we
got to do?
What do we got to do?
We got to do this, we got to dothat, do this, do that.
And I think that they lose theunderlying motivators that
people have that will get themto offer the discretionary
effort to get on board withwhatever it is they're trying to
engage their workforce in.
(03:59):
And that's really what do theybelieve about the change?
And I think if more leadersthought about the beliefs that
people hold about their work andabout the company, they'd be way
more effective in drivingresults than just worrying about
what do we got to do.
SPEAKER_04 (04:15):
On that, I'd just like to pause to uh to underline
what you've said or to give youa chance to underline what the
practical ways organizations cantap into people's belief systems
to bring them alongpsychologically with change.
SPEAKER_01 (04:32):
I mean, when we work with our clients, the first
thing we do is ask them thequestion what beliefs do your
team hold right now that aregetting in the way of you
achieving your results?
And usually when you have thatconversation, the same themes
bubble up.
You don't need a consultant toask that question.
You get people in a room and youask that question and the truth
will arise.
The question is, how do youshift those beliefs?
(04:54):
So you identify the beliefs thatare not aligned with the results
you're trying to achieve.
And then the more powerfulquestion is what do you want
those beliefs to be?
And you shift beliefs withexperiences.
So leaders are really in theexperience management business
where they have to think aboutwhat experiences am I creating
and what beliefs are thosedriving, and how do I need to
change those two things.
SPEAKER_04 (05:14):
I like that because you're also shifting their
thinking around how you need toshow up to really model and
support change and meet peoplewhere they're at and address
their underlying concerns,thoughts, perspectives.
SPEAKER_01 (05:29):
Yeah, I mean, I think a lot of leaders are
deluded by the idea that becausethey have a job title that says
manager, they can control peoplethat report into them.
I mean, if it was as simple asI'm gonna tell you what to do
and you're gonna do it, wewouldn't have all these
leadership books being written,right?
It's not that simple.
(05:50):
You can't control people, youcan only control yourself.
And so the question is, you as aleader, how do you need to show
up?
What experiences do you need tocreate to get beliefs in line
with the actions you need peopleto take to get the results?
That's the voodoo, you know, thejujitsu of leadership.
I have a book coming out inJanuary called Surrender to
Lead.
And it's really about this idealike you have to surrender this
(06:12):
illusion that you have anycontrol over anything outside of
you and figure out how do youneed to show up best in order to
drive results.
And there's a way to get resultswithout waving the white flag of
surrender.
It's really surrendering to theillusion that you have power you
don't have.
SPEAKER_02 (06:28):
So, Jessica, one of the drivers of change requiring
all these change management andmindset management interventions
is, of course, AI.
And that's what we like to talkabout on this podcast.
So, to make it concrete, I'minterested in what you're
actually seeing in theorganizations you work with in
terms of what they're doing withAI.
So, so let's start with one ofthe sort of easiest use cases.
(06:51):
Have you seen anyone makingespecially good use of AI for
creating content?
SPEAKER_01 (06:56):
Oh, yeah, absolutely.
And by creating content, ifyou're talking about marketing
content, that is an output thatI've seen lots of marketing
teams across industries using.
And that I think is pretty, Imean, I use it, right?
I create content, I use Chat GPTto help me create that content.
And that's that's like awell-known, long-used use case
(07:18):
of AI.
But I also think sometimes thecontent is internal, right?
As a leader, I might be writingan email to someone.
That's content that I am usingAI to facilitate.
So that simple use of justlanguage creation making that
easier and faster, I think ispretty widespread at this point.
SPEAKER_02 (07:38):
Have you seen anybody do anything you think,
oh, well, that was particularlyclever?
SPEAKER_01 (07:43):
Particularly clever?
I mean, no.
I haven't seen anything thatisn't maybe certain kind of
prompt engineering is moresophisticated than other prompt
engineering, and that might bequite clever, and that depends
on how far have you leaned intoAI.
But that's I haven't seenorganizations embrace that
wholeheartedly versus others whodon't.
(08:03):
It's really at the individuallevel.
Are you embracing AI?
And if so, have you done thework to learn your skills
necessary to best leverage AI?
And there I've seen some peoplebe better than others.
SPEAKER_02 (08:15):
Yeah, and I I like that because it gives us a clear
focus that the low-hanging fruitis individual use.
And if you can get individualsdoing it well, there's a lot to
be gained there.
SPEAKER_04 (08:28):
Yeah.
And leveraging it as aproductivity tool and enabling
supporting the organization.
And we'll come back to thinkingthrough how we how we enable
organizations.
SPEAKER_00 (08:40):
Thanks for listening to this episode of the HR Chat
Podcast.
If you enjoy the audio contentwe produce, you'll love our
articles on the HR Gazette.
Learn more at hrgazette.com.
And now back to the show.
SPEAKER_04 (08:56):
I'd also be interested in hearing of use
cases related to leveraging datamore effectively.
And wondering what you've seen.
It doesn't have to be an HRexample, whether you've seen
particularly good use ofleveraging the capabilities of
AI to analyze data, to useinternal data more effectively.
SPEAKER_01 (09:16):
Yeah, that's where I think the most advances are
being made right now.
Content creation was the storyof last year.
This year, I think the story ischurning through data.
So there's one manufacturingcompany that we've been working
with where in the supportsystems, let's call it finance.
Well, it was, I'll tell you, itwas the sales team and then it
(09:38):
was the finance team after that.
They have had particular leaderswithin the organization who have
embraced the idea that AI iscoming, it's going to help us be
more efficient.
And what they've leveraged AIfor is churning through data,
analyzing data.
And so it started with the salesteam, where they wanted to be
able to interpret the pipeline,where leads were coming from,
(10:02):
what leads were translating toclosed deals, and asking that
question the way that it waswhen they began this process was
let's go into Salesforce, let'spull a bunch of data based on
what has been entered into thesystem, and then let's send it
to our data analyst to thenanswer these particular
questions that they have, right?
(10:22):
What they did is they integratedan AI tool.
So now the lay person on theteam, which means lower in the
organization, frontlinemanagers, can go into this
chatbot and say, Can you explainto me where the most, uh the
highest close rate lead sourcesare?
And now they can ask questionsthat previously they probably
couldn't ask because it was theexecutives who were directing
(10:45):
the data analysis work.
And it's allowing morevisibility and more
understanding of data.
When this particularmanufacturing team implemented
that, well, then the financeteam saw that and said, Well,
wait a minute, let's hook themup to our systems and see what
kind of financial analysis wecan do.
Important to note here that nojobs got replaced in that
(11:06):
situation, right?
It wasn't like, oh, we need lesssales managers now or we need
less data analysts now becausethe data analysts were on the
finance team.
The finance team still has allthose people because those
people are now rather than doingthe work in spreadsheets,
they're checking the work thatthe AI tool did and then
interpreting that work for themanagers asking the question.
(11:29):
So it's still not replacing jobsin that situation in the clients
that I've seen, but it isallowing for more visibility and
ease of computation.
SPEAKER_04 (11:39):
Thank you.
I just find it so interesting asyou consider that when
organizations are able to havemore advanced use cases, the
expertise becomes more valuable.
Rather than displacing people,we have more opportunity to
leverage those skills.
Building on that example, wouldwelcome your thoughts on what
the lesson for HR is from theexamples that you're seeing
(12:02):
within operations.
SPEAKER_01 (12:04):
Well, I think HR has been trying to get a seat at the
table for the last 30 years, andthey've they've still not quite
gotten there because I don'tthink they can speak the
language of the CFO aseffectively as they should.
They're still focused onengagement scores and things
that don't make the boardmembers jump.
The board members jump when theCFO says, I got a metric here,
(12:27):
that's a problem.
They don't jump when HR says,I've got a metric here, that's a
problem.
And so this ability to look intothe analytics and the impact of
the analytics on the bottom linewill probably get them closer to
having that seat at the table.
But it also requires, onceagain, the mindset shift.
Um, that's how I would be usingit as HR.
(12:48):
What I've seen in HR is they'restill focused on the people side
of it, which trust me, there'sno person that cares more about
people than me.
When I'm a culture expert, thisis my thing, right?
But I also know thatpeople-focused initiatives are
short-term if they aren't tiedto results, because ultimately
(13:08):
capitalism will win.
And so you have to see whatworks for results and people.
And so what I've seen HR peopledo is um, for example, there's
one technology company thatimplemented a tool, you know,
these AI Zoom note takers thatlisten to your calls and then
(13:29):
they summarize.
That's fine.
This particular tool that theyimplemented actually does
psychoanalysis of the culture onyour call.
So it's listening to the call.
And then I read the report ofone call.
I watched a call and then I readthe report afterwards.
And um, the CEO was on the call,and it was a business review
(13:51):
call with one particulardepartment, and multiple
departments were represented onthe call.
And the CEO was cursing andsaying, Well, I don't know why
the F we would care about that,et cetera, and so forth, right?
And I knew this CEO, and I knewthat it's just a very casual
tech company, right?
I mean, these are the companiesthat they wear hoodies and it's
(14:12):
just like that.
No one on the call, based on myunderstanding of that particular
culture, was feeling like he wasbeing aggressive.
It was just vulgar and casual,right?
So the AI bot pumped out thisreport that said, well, this CEO
is cursing, and that can beinterpreted one of two ways.
Either he's being aggressive, orthis is a level of psychological
(14:34):
safety in which that is a normof behavior.
Based on the fact that thisother person cursed later in the
call, I believe it's because ofthe former.
I mean, it's like giving youthis opinion of the way that
culture is manifesting in theconversation on the Zoom call.
That's interesting.
Now, what are you gonna do withthat data, HR, right?
Are you gonna go give the CEOfeedback about cursing?
(14:56):
Are you gonna report to theboard that the CEO cursed and
this could mean one of twothings?
I don't know that that'snecessarily gonna move the
needle, but we're starting tosee ways in which AI is
psychoanalyzing people.
That's one interesting way.
I've seen other problematic waysat other companies that uh maybe
getting into legal hot water.
(15:17):
That's what I'll be watching.
SPEAKER_02 (15:20):
One of the things you mentioned in our
pre-discussion was you've seenorganizations using AI for
logic.
And I wasn't quite sure what youmeant by that.
SPEAKER_01 (15:27):
Yeah.
So, well, this actuallydovetails well from the previous
thing I was just describing, thepsychoanalysis.
So there's one organization thatbasically provides like
background checks as a service,and they implemented an AI tool
to look at the data that comesback from the background check
(15:49):
and then provide apsychoanalysis of the person
based on their background checkresults.
So typical background checksright now come back and they say
this person has a lien on theirhouse and they've been arrested,
or you know, they have whateverthe background check data is,
right?
It's just factual data presentedthat is given to an employer
(16:10):
when you're considering hiringsomeone.
Well, this exploration wasaround what does this mean for
the profile of someone likethis?
It's like, well, based on thefact that they had a DUI 15
years ago, but their credit wasreally bad and then it turned
really good.
We actually think this problemperson probably achieved
sobriety and therefore is on adifferent path of life.
(16:30):
And so it's a low-risk DUI ascompared to someone who had a
DUI three years ago and also isstill in bankruptcy, right?
I mean, that's what the AI toolwas doing.
It was analyzing someone, whichis, I mean, first of all,
illegal when it comes to whoyou're gonna hire and fire to
profile people in that way.
I was just talking to someonewho does a lot of consulting in
(16:52):
this space, and they have this.
I think you're not allowed touse AI to analyze candidates at
all, according to someregulation, but you'll have to
fact check me on that because Idon't actually know.
But what other use cases arethere for background checks?
I mean, it could be to whetherto hire or fire, but it could
also just be fornon-work-related things, private
(17:12):
investigators, for example,looking into people and trying
to piece together somethingthey've been hired to figure
out.
That's totally creepy in mymind, right?
And they pulled the plug onmoving forward with that because
they had the same creep factorcome up for them when they did
the analysis.
So, logic in that case is we'regonna take this data and we're
(17:35):
gonna make sense of it.
AI is really good at languagemodel, is creating language
around the input prompts.
But to logic, I don't know thatit's reliable enough
necessarily.
I have another example, is aprofessional services
organization that had a bunch ofdifferent um clients that they
serve.
And one of the things that theydid was they suggested that they
(17:58):
could match the clients'services with the clients' needs
that they had and using AI to dothat, asking AI to find, okay,
here's all the services thatthese clients have.
Here's also the needs that ourclients have.
Can we pair them with eachother?
That kind of logic is not veryeasy or effective or reliable
right now at the AI level.
(18:19):
And so that's what we're stillworking on figuring out how to
do well with AI.
SPEAKER_04 (18:25):
I appreciate the examples and also the
complexities that you'repointing to as well.
And we know in the Canadianmarketplace, if any part of a
decision is related toprotective grounds, um,
employers are on the wrong sideof employment law.
So if you had something predict,right, that someone was
suffering from an illness,addiction, then that would
(18:48):
absolutely create a ton of riskwithin the higher rank process.
SPEAKER_01 (18:53):
Well, I wonder how insurance companies are using AI
for this, right?
I mean, because they have topredict risk as part of their
business model.
I just had um the CHRO of AFLAC,or no, he's the chief strategy
officer, but he is the head ofHR, which is an interesting
title dynamic.
Anyway, um, they're in thebusiness of risk.
I should have asked him ifthey're using AI to predict risk
(19:16):
with their consumers, you know.
I mean, is that illegal?
I don't even know.
SPEAKER_04 (19:20):
So really underlining there's there's the
advances and there's thecautions, and just it comes back
to the importance of testing, ofpiloting, of making sure that
when we're applying a use case,it's just something that we have
we have a solid foundation ofand we've uh assessed the risk.
Before we move to risk, though,we'd like to dig a bit deeper on
the culture side and whetheryou've seen use cases around to
(19:45):
measure culture, to supportculture.
SPEAKER_01 (19:49):
Yeah.
So um I had a call with a vendorwho said that they one of the
big four consulting firms wastheir client, then they were
reselling it.
And this vendor has a tool inwhich it integrates with
Outlook, with Slack, with theinternal communication systems
(20:10):
to help facilitate bettercommunications.
So you're writing an email,basically, it gets a profile of
everyone in the company based onsome psychological assessment
that they've taken, right?
So let's say uh ACME Companyhires this vendor, right?
Every one of the employees atACME Company fills out a
(20:33):
psychological survey and itunderstands now this is how you
like to communicate, this is howlike you like to be communicated
with, this is your priority inworkplace dynamics.
Well, then I'm gonna write anemail to Joe at that company.
And I write the email, but thenbefore I send the email, this AI
bot is reading the email,understanding Joe's profile, and
(20:56):
making suggestions to me abouthow I could make my email
resonate better with Joe basedon his profile and the way he
likes to be communicated with.
So it'll say, well, Joe isreally likes to prioritize
relationship over tasks.
And so maybe you want to startwith asking how his day was,
right?
And it'll rewrite the email foryou before you send it to Joe,
(21:17):
which sounds great.
Then Joe receives the email.
It's a wonderful email in Joe'smind because it was carefully
crafted for his personalitytype.
And then Joe wants to writeback, blah, blah, blah.
My email, my weekend was great.
Thank you so much.
Let me send it back to you.
And then the AI bot sees Joe'snote to you and says, Oh, you
know what, Joe?
Jessica actually really preferswhen people just get down to
(21:39):
business.
Why don't you take out theweekend stuff and just make a
bullet point of tasks becausethat's how she likes to
communicate?
Joe's like, that sounds good.
He sends it off.
I think Joe is so great.
I love the way Joe communicates.
What's really happening thoughis that we're not communicating
with each other.
We've filtered how wecommunicate with each other
through this AI tool that'schanging it to be the way I like
(22:02):
to receive communication.
And I actually don't know muchabout Joe because Joe is not
talking to me the way Joe talksto me.
Joe is talking to me the way Ilike to be talked to through an
AI filter.
I haven't had any clientsimplement it, but I know clients
that they've listed that didimplement it.
And I have serious concernsabout what that.
I mean, their argument is it'sgonna make culture better
(22:24):
because everyone's gonna behappier with the way that it's
gonna reduce friction andcommunication.
But come on, I mean, is itreally communication if it's
just robots talking to robots,you know?
Thank you.
SPEAKER_04 (22:37):
That's such a just a great example of how we can
endeavor to improve somethingand remove friction and and
we're actually uh damaging itand without thinking through
just the consequences of peoplenot really getting to know each
other or understanding theirpreferences or getting to know
them in the way they want to beknown as as well.
(22:59):
Are there use cases that even ifyou haven't seen it, that excite
you that could actually moveculture forward in a way that
would be more helpful or waysthat you would like HR to
consider how to leverage it inan effective way rather than a
way that uh is more gettingalong through smoke and mares.
SPEAKER_01 (23:15):
Yeah.
So the instill AI example Igave, which is they listen to
the calls and then they do apsychoanalysis of those calls.
I've been excited about that.
And the reason I'm excited aboutthat is because the culture
partners, what we're alwaysdoing is trying to share with
people that if you really wantto get people's hearts and minds
aligned with the results you'retrying to create, you got to get
(23:37):
at the belief level, right?
And what instillAI wouldtheoretically be able to do is
track, measure, and analyzewhether those beliefs are in
place or not.
And if you see people not havingthose beliefs in place, that's a
predictive tool to show you thatyour change effort is not going
(23:58):
to be successful.
And it can flag to the leader,here's where your beliefs need
to shift.
It's like a mirror, right?
It also is a self-awareness toolbecause it can send a note to
the leader who cursed on thatcall and say, Hey, maybe you
want to think about not cursingon that call, which no one on
the call is gonna say to thatperson because they all report
(24:20):
into him, right?
So the self-awareness thing isthe big unlock that we need to
create better leaders.
We need leaders to be willing tobe self-aware.
And it's hard to get leadersfeedback because they're the
leaders.
They're in charge of your job,they're in charge of your
salary, they're in charge ofyour security at this company.
(24:40):
And so I'm not gonna tell myboss something he doesn't want
to hear at my own peril.
I'm not that principled.
I don't have that muchintegrity.
I want to make sure that I havea job so I can send my daughter
to school before I care aboutsaying the truth to power, you
know.
Sorry, that's just my perceptionof, you know, my priorities.
So this AI bot can speak totruth to power, though.
(25:02):
And I think that's interesting.
I'm excited about that.
SPEAKER_04 (25:06):
It's interesting because you think of the
examples that we're workingthrough, how many of them keep
coming back to supportingindividual productivity,
individual growth,self-awareness, and how that can
support the broaderorganization?
Yeah.
SPEAKER_02 (25:21):
Now, uh, Jessica, we've been focusing on some of
the interesting thingsorganizations who are leaning
into AI are doing.
What about those organizationsyou've seen who are failing to
make use of AI?
Do you see any commonalities inthose organizations that are
just very slow on adopting AI?
SPEAKER_01 (25:42):
I think more organizations are in that
category than not because it'sstill so new to transform an
organization around this brandnew technology requires a lot of
work.
And so you're seeing it inpockets based on individual
buy-in to the idea.
You're I I haven't seen, I mean,Fiverr just made this
(26:04):
announcement this week thatthey're gonna go back to being a
startup, they're laying off 250people and they're gonna be AI
first.
That's interesting, right?
The CEO of Fiverr, when he madethat announcement, got
absolutely slaughtered online.
I mean, he was just inundatedwith comments and messages about
how dare you! I'm never usingyour service again.
(26:25):
You know the worst thing aboutcorporate greed and blah, blah,
blah, right?
Because the CEO said, we'regonna be AI first and it
requires us to lay people off.
He is the 6,000th CEO to makethat decision.
And yet he's still getting thepushback.
Other CEOs, when they've doneit, what are they doing?
Are they replacing people withAI yet?
(26:46):
No, no one is.
I haven't seen any clients.
I have zero clients that havereplaced large groups of people
with AI efficiency.
I think that's going to comewith agentic AI, which is just
around the corner, but it hasn'thappened yet.
What you're seeing is peoplelaying off their populations of
employees to make room in thebudget by reducing labor costs
(27:08):
so that they can invest inexploring the possibility of AI
efficiencies.
It's it's technically replacingpeople, but not because the
efficiencies already exist.
It's because ideologically wewant it to exist.
So let's get rid of thesepeople, which is really, I think
AI has become a scapegoat foroverhiring and laying people
(27:30):
off, even though the LinkedIncomments look bad.
The market responds positivelyto layoffs now, which it didn't
10 years ago.
So you make an announcement thatyou're laying a large group of
people off, and the the marketresponds with great, you're a
forward-thinking,efficiency-based leader that's
results oriented.
We love this for you.
We we have we're positive aboutyour future.
(27:52):
10 years ago, or let's really go20 years ago, if you announced
layoffs, it meant you were atyour last straw before the
bankruptcy and breaking point ofyour company, and so you must be
about to fail.
That's not the perception of itanymore.
SPEAKER_04 (28:08):
Do you have any thoughts on how organizations
are governing AI, how they'remanaging risks, and would love
your thoughts on what you see asineffective, what you see as
effective.
SPEAKER_01 (28:23):
I think AI, the AI equivalent of cybersecurity, is
really in its nation stages, butwill be very, very popular as
more high-profile AI failuresbecome well known.
And there have been many ofthem, right?
I'm thinking of there was onereal estate organization that
(28:46):
started using AI to analyze whatproperties they should invest
in.
And the model was built based onhistorical economic data.
And so the model assumed thatthe economic situation was
static as it did its analysis.
Well, what happened is themarket shifted and the housing
(29:07):
market shifted, but the AI modelwas still based on assumptions
that we were in a different kindof economy.
And so it was popping out abunch of recommendations that
were totally badrecommendations.
And the company investedmillions of dollars in those
recommendations withoutrealizing that they were bad
recommendations because ithadn't adjusted for a new
(29:30):
economic reality.
That's a very straightforwardcase of AI taking a business
down a bad rabbit hole becauseit didn't have securities in
place.
And there's other things that AIcan do, the hallucination, for
example, right?
There's um healthcareinstitutions that have
implemented AI for, for example,analyzing radiology, right?
(29:53):
You analyze radiology using AI,and what you saw in one
particular case was it startedhallucinating.
Um, things and basically it wasdetecting problems, you know,
80% more of the time than itactually existed.
It's better than if it haddetected problems 80% less of
the time than it actuallyexisted, but it was just as
(30:14):
likely to go in one direction orthe other because it didn't have
this is the logic thing.
It didn't have the kind of logicthat can understand nuance and
context that the people readingthe radiology exam before did,
right?
That the people analyzing theinvestment portfolio before did.
And so that's where work stillneeds to be done.
And that's what AI quote unquotecybersecurity looks like.
(30:38):
It's different thancybersecurity where you're
looking, you know, you gotscammers coming in and trying to
get your data, right?
It's like, well, what is AI'serror point, failure point going
to look like?
And how can we protect ourselvesfrom that?
That will be, I think, uhsomething that gets popular next
year and the year after.
SPEAKER_04 (30:56):
We've seen practical ways organizations manage the
risks and opportunities.
SPEAKER_01 (31:02):
Yeah, people checking the work.
I mean, that's why I don't thinkyou've seen that many layoffs.
It's great, you've popped thisout.
Now we're gonna look at it andsee if it's remotely accurate or
not.
I mean, that is the peoplesaying that AI is not gonna
replace people because of that.
I think that's maybe true forabout four more months, you
know, in my opinion.
Uh I anyone who says they knowwhat the future of AI is going
(31:25):
to be, they're trying to sellyou something.
This is just my opinion.
It will get better.
It will replace people.
Agentic AI has so muchopportunity.
There are these big high-profilelike Klerna.
They laid off a bunch of people,said they were going to replace
the organization with agenticAI, and then they went back
because it failed miserably.
And now they want to be known asthe person who will guarantee
(31:46):
that you'll get a person on thephone.
I have one organization that's autility company that's exploring
the idea of replacing their callagents with AI.
They bought the tool, theyimplemented it.
It is uncanny.
I listened to a recording of acall agent helping a customer,
and the customer had no ideathey were talking to AI.
(32:08):
If I listened to a recording anddidn't know it is AI, I would
not have thought it was AI.
And yet, they're still notpulling the trigger because
they're worried about, well, arewe sure?
You know, it's just not, we'renot quite there yet, but it's
just around the corner.
Thank you.
SPEAKER_02 (32:24):
And just I would just wrap up my thoughts on
this, that one just has to moveforward with caution.
Explore, uh not necessarilybelieve what the vendor is
telling you.
Um, but if you explore, youmight find it works extremely
well, better than expected, oryou might find it's doing things
badly worse than you hadthought, and you just need to do
(32:48):
the pilots and keep an eye onhow it's going.
SPEAKER_01 (32:51):
Well, this is the this is the challenge of
capitalism and growth, right?
Innovation requires us to failfast and take risks.
And so what risks are youwilling to take?
This is these are the decisionsthat leaders have been making
for a long time.
Now they're pointing it at AI,right?
I mean, personally, and this isirrelevant to your listener
(33:11):
because who am I?
But personally, I'm a little bitscared of AI, right?
And I would love if everyonejust said, you know what, forget
it.
It's not worth it.
But I feel the same way aboutsocial media.
I feel the same way abouttechnology.
I mean, if it if I had mydrudhers, we'd go back to being
hunters and gatherers, you know?
I'd go back to pre-agriculturalage because I think that's
(33:33):
probably better for our mentalhealth.
But that's not the nature of theworld.
So here we are.
People are trying to make money,they're going to continue to
innovate, and that innovation isgoing to have consequences.
And so I'm here to try and helppeople navigate the reality of
the world the way that it is.
So the reality is no one's goingto slow down.
As long as they don't have to,they're not going to slow down.
(33:53):
And therefore, as an employee,as a person in this world, how
can you adapt to deal with thatreality?
It's again going back to thisidea of surrender.
You can resist that the futureis AI because you think it's
unsafe, or you can accept thatit's happening.
And so, therefore, what am Igoing to do in this new reality?
SPEAKER_04 (34:14):
Any practical tips in that regard?
Like how leaders can accept thepractical reality, how they
support their teens acceptingthe practical reality, how they
support enablement,democratization of capabilities
in this new realm.
SPEAKER_01 (34:30):
Number one, so I have a model in the book.
It's called the shift model thatis around this.
We don't have to go through thewhole shift model, but first
step is stop fighting reality.
And reality is everythinghappening outside of you.
We've spend so much time in ourheads wishing our employees were
different.
If only they were more engaged,if only the leaders created
(34:52):
lower objectives for me to hit.
If only the administrationdidn't blank, if only the
competitors weren't X, I mean,there's so much, it should be
this way, if only it were thatway.
That think of how much energyand time you spend spinning your
wheels on that.
That's a lack of surrender,right?
We call that below the linethinking.
(35:12):
So to go above the line thinkingis to say, okay, what's going
on?
And what about that can Icontrol?
Let me just put my energy there.
And if I just focus on makingthe personal choice to focus on
what I can control to driveresults, well, now I've at least
focused my energy in somethinguseful instead of spinning my
(35:34):
wheels on the way I wish it was.
Then once you've identifiedwhere you actually can make an
impact, well, then you can makedecisions about how you can show
up to create the rightexperiences that will shape the
right beliefs, that will get theright actions, that will get the
results.
I mean, I think the number onething people need to do is get
(35:56):
out of the action trap ofconstantly trying to do
something, that endless cycle ofactivity that feels like
progress, but is really notmoving the needle.
And get out from below the linethinking where you're pointing
fingers, you're like, well,that's not my job, and it should
be this way.
And it, those folks need to bedifferent.
All of that is wasted energy.
(36:16):
There's something in betweenthat I think is where we need to
spend our time.
SPEAKER_04 (36:20):
Thank you, Jessica.
This has been a short andinteresting conversation for us.
For those who would like to getin contact with you, what's the
best way to do so?
SPEAKER_01 (36:31):
You can go to my website, jessica.com.
Uh Kriegel is calledK-R-I-E-G-E-L, or you can go to
our company's website, which isculturepartners.com.
And I have a weekly newsletteron LinkedIn if you want to join
that.
I pop out insights everyWednesday called This Week in
Culture, and it just tells youwhat's going on in the world of
culture.
(36:51):
Thank you.
SPEAKER_03 (36:56):
Thanks for listening to the HR Chat Show.
If you enjoyed this episode, whynot subscribe and listen to some
of the hundreds of episodespublished by HR Gazette?
And remember, for what's new inthe world of work, subscribe to
the show, follow us on socialmedia, and visit hrgazette.com.
Welcome to the HR Chat Show, one of the world's
most downloaded and sharedpodcasts designed for HR pros,
talented defects, techenthusiasts, and business
leaders.
For hundreds more episodes andwhat's new in the world of work,
subscribe to the show, follow uson social media, and visit
hrgazette.com.
SPEAKER_04 (00:25):
Hello and welcome to the HR Chat Podcast.
I'm Pauline James, founder andCEO of Anchor HR and associate
editor of the HR Gazette.
It's my pleasure to be your hostalong with David Krillman, CEO
of Krillman Research.
We're partnering with the HRChat Podcast on a series to help
HR professionals and leadersnavigate AI's impact on
(00:46):
organizations, jobs, and people.
SPEAKER_02 (00:48):
In this episode, we are speaking with Jessica
Kriegel.
Jessica is Chief StrategyOfficer at Culture Partners in
California.
She has a strong background inthe world of technology, having
spent a decade in changemanagement at Oracle.
Since then, Jessica has become arecognized expert in
organizational culture andstrategy.
Her work has been featured byCNBC, CNN, The New York Times,
(01:10):
the Wall Street Journal, andForbes, among others.
We connected with her as we wereeager to hear the use cases for
AI that she has seen amongst herclients and their impact, as
well as to draw on her changemanagement expertise.
We're also happy to share thatJessica has a book coming out in
January.
It's called Surrender to Lead.
So we look forward to that.
SPEAKER_04 (01:30):
Jessica, just to kick us off, can you tell me
about the advisory board thatyou were on where you met David?
SPEAKER_01 (01:37):
Yes, I would love to.
We're both on the WorkforceInstitute, which is a think tank
that's part of UKG that focuseson the future of work, what are
best practices to get results,what are the needs of employers
and employees, and where dothose needs meet and where did
they diverge.
So it's been an absolutepleasure meeting David there.
(01:59):
Thank you.
SPEAKER_04 (02:00):
I see you do a lot of podcasts, keynotes.
I'm interested to hear what HRaudiences are asking you to
speak to right now, what they'remost interested in.
SPEAKER_01 (02:12):
Um, right now, the biggest talk track I have to
talk about is change management.
This is the moment for changemanagement.
Interestingly, I'm not at HRTech, but at this year at HR
Tech.
It was standing room only in thechange management session put on
by Deloitte.
That is the big question markfor people because of the nature
(02:33):
of the speed of change and thenumber of changes happening in
any organization at one time.
And that's a verypeople-oriented challenge.
Thank you.
SPEAKER_04 (02:42):
With that, is there anything additionally you feel
that they should be focusing onthat you like to draw, pull
their attention to?
SPEAKER_01 (02:51):
Yeah, my opinion, my expertise is around how do you
get the hearts and minds of youremployees aligned with the
organization's objectives.
That's really the key in mymind.
And with change management, whatit typically looks like is a
bunch of training and a bunch ofPR and a bunch of trying to get
(03:15):
people on board with an ideawith communication and rousing
speeches in town halls, multipleemails and systems that'll uh
allow you to implement thatchange, whatever the change is,
right?
The change could be, oh, we'regoing to AI.
The change could be we need todo a digital implementation, or
the change could simply be weneed to grow by X percent.
And that's a change from theamount that we've been at up
(03:38):
until this point.
And a lot of leaders get stuckin the action trap of what do we
got to do?
What do we got to do?
We got to do this, we got to dothat, do this, do that.
And I think that they lose theunderlying motivators that
people have that will get themto offer the discretionary
effort to get on board withwhatever it is they're trying to
engage their workforce in.
(03:59):
And that's really what do theybelieve about the change?
And I think if more leadersthought about the beliefs that
people hold about their work andabout the company, they'd be way
more effective in drivingresults than just worrying about
what do we got to do.
SPEAKER_04 (04:15):
On that, I'd just like to pause to uh to underline
what you've said or to give youa chance to underline what the
practical ways organizations cantap into people's belief systems
to bring them alongpsychologically with change.
SPEAKER_01 (04:32):
I mean, when we work with our clients, the first
thing we do is ask them thequestion what beliefs do your
team hold right now that aregetting in the way of you
achieving your results?
And usually when you have thatconversation, the same themes
bubble up.
You don't need a consultant toask that question.
You get people in a room and youask that question and the truth
will arise.
The question is, how do youshift those beliefs?
(04:54):
So you identify the beliefs thatare not aligned with the results
you're trying to achieve.
And then the more powerfulquestion is what do you want
those beliefs to be?
And you shift beliefs withexperiences.
So leaders are really in theexperience management business
where they have to think aboutwhat experiences am I creating
and what beliefs are thosedriving, and how do I need to
change those two things.
SPEAKER_04 (05:14):
I like that because you're also shifting their
thinking around how you need toshow up to really model and
support change and meet peoplewhere they're at and address
their underlying concerns,thoughts, perspectives.
SPEAKER_01 (05:29):
Yeah, I mean, I think a lot of leaders are
deluded by the idea that becausethey have a job title that says
manager, they can control peoplethat report into them.
I mean, if it was as simple asI'm gonna tell you what to do
and you're gonna do it, wewouldn't have all these
leadership books being written,right?
It's not that simple.
(05:50):
You can't control people, youcan only control yourself.
And so the question is, you as aleader, how do you need to show
up?
What experiences do you need tocreate to get beliefs in line
with the actions you need peopleto take to get the results?
That's the voodoo, you know, thejujitsu of leadership.
I have a book coming out inJanuary called Surrender to
Lead.
And it's really about this idealike you have to surrender this
(06:12):
illusion that you have anycontrol over anything outside of
you and figure out how do youneed to show up best in order to
drive results.
And there's a way to get resultswithout waving the white flag of
surrender.
It's really surrendering to theillusion that you have power you
don't have.
SPEAKER_02 (06:28):
So, Jessica, one of the drivers of change requiring
all these change management andmindset management interventions
is, of course, AI.
And that's what we like to talkabout on this podcast.
So, to make it concrete, I'minterested in what you're
actually seeing in theorganizations you work with in
terms of what they're doing withAI.
So, so let's start with one ofthe sort of easiest use cases.
(06:51):
Have you seen anyone makingespecially good use of AI for
creating content?
SPEAKER_01 (06:56):
Oh, yeah, absolutely.
And by creating content, ifyou're talking about marketing
content, that is an output thatI've seen lots of marketing
teams across industries using.
And that I think is pretty, Imean, I use it, right?
I create content, I use Chat GPTto help me create that content.
And that's that's like awell-known, long-used use case
(07:18):
of AI.
But I also think sometimes thecontent is internal, right?
As a leader, I might be writingan email to someone.
That's content that I am usingAI to facilitate.
So that simple use of justlanguage creation making that
easier and faster, I think ispretty widespread at this point.
SPEAKER_02 (07:38):
Have you seen anybody do anything you think,
oh, well, that was particularlyclever?
SPEAKER_01 (07:43):
Particularly clever?
I mean, no.
I haven't seen anything thatisn't maybe certain kind of
prompt engineering is moresophisticated than other prompt
engineering, and that might bequite clever, and that depends
on how far have you leaned intoAI.
But that's I haven't seenorganizations embrace that
wholeheartedly versus others whodon't.
(08:03):
It's really at the individuallevel.
Are you embracing AI?
And if so, have you done thework to learn your skills
necessary to best leverage AI?
And there I've seen some peoplebe better than others.
SPEAKER_02 (08:15):
Yeah, and I I like that because it gives us a clear
focus that the low-hanging fruitis individual use.
And if you can get individualsdoing it well, there's a lot to
be gained there.
SPEAKER_04 (08:28):
Yeah.
And leveraging it as aproductivity tool and enabling
supporting the organization.
And we'll come back to thinkingthrough how we how we enable
organizations.
SPEAKER_00 (08:40):
Thanks for listening to this episode of the HR Chat
Podcast.
If you enjoy the audio contentwe produce, you'll love our
articles on the HR Gazette.
Learn more at hrgazette.com.
And now back to the show.
SPEAKER_04 (08:56):
I'd also be interested in hearing of use
cases related to leveraging datamore effectively.
And wondering what you've seen.
It doesn't have to be an HRexample, whether you've seen
particularly good use ofleveraging the capabilities of
AI to analyze data, to useinternal data more effectively.
SPEAKER_01 (09:16):
Yeah, that's where I think the most advances are
being made right now.
Content creation was the storyof last year.
This year, I think the story ischurning through data.
So there's one manufacturingcompany that we've been working
with where in the supportsystems, let's call it finance.
Well, it was, I'll tell you, itwas the sales team and then it
(09:38):
was the finance team after that.
They have had particular leaderswithin the organization who have
embraced the idea that AI iscoming, it's going to help us be
more efficient.
And what they've leveraged AIfor is churning through data,
analyzing data.
And so it started with the salesteam, where they wanted to be
able to interpret the pipeline,where leads were coming from,
(10:02):
what leads were translating toclosed deals, and asking that
question the way that it waswhen they began this process was
let's go into Salesforce, let'spull a bunch of data based on
what has been entered into thesystem, and then let's send it
to our data analyst to thenanswer these particular
questions that they have, right?
(10:22):
What they did is they integratedan AI tool.
So now the lay person on theteam, which means lower in the
organization, frontlinemanagers, can go into this
chatbot and say, Can you explainto me where the most, uh the
highest close rate lead sourcesare?
And now they can ask questionsthat previously they probably
couldn't ask because it was theexecutives who were directing
(10:45):
the data analysis work.
And it's allowing morevisibility and more
understanding of data.
When this particularmanufacturing team implemented
that, well, then the financeteam saw that and said, Well,
wait a minute, let's hook themup to our systems and see what
kind of financial analysis wecan do.
Important to note here that nojobs got replaced in that
(11:06):
situation, right?
It wasn't like, oh, we need lesssales managers now or we need
less data analysts now becausethe data analysts were on the
finance team.
The finance team still has allthose people because those
people are now rather than doingthe work in spreadsheets,
they're checking the work thatthe AI tool did and then
interpreting that work for themanagers asking the question.
(11:29):
So it's still not replacing jobsin that situation in the clients
that I've seen, but it isallowing for more visibility and
ease of computation.
SPEAKER_04 (11:39):
Thank you.
I just find it so interesting asyou consider that when
organizations are able to havemore advanced use cases, the
expertise becomes more valuable.
Rather than displacing people,we have more opportunity to
leverage those skills.
Building on that example, wouldwelcome your thoughts on what
the lesson for HR is from theexamples that you're seeing
(12:02):
within operations.
SPEAKER_01 (12:04):
Well, I think HR has been trying to get a seat at the
table for the last 30 years, andthey've they've still not quite
gotten there because I don'tthink they can speak the
language of the CFO aseffectively as they should.
They're still focused onengagement scores and things
that don't make the boardmembers jump.
The board members jump when theCFO says, I got a metric here,
(12:27):
that's a problem.
They don't jump when HR says,I've got a metric here, that's a
problem.
And so this ability to look intothe analytics and the impact of
the analytics on the bottom linewill probably get them closer to
having that seat at the table.
But it also requires, onceagain, the mindset shift.
Um, that's how I would be usingit as HR.
(12:48):
What I've seen in HR is they'restill focused on the people side
of it, which trust me, there'sno person that cares more about
people than me.
When I'm a culture expert, thisis my thing, right?
But I also know thatpeople-focused initiatives are
short-term if they aren't tiedto results, because ultimately
(13:08):
capitalism will win.
And so you have to see whatworks for results and people.
And so what I've seen HR peopledo is um, for example, there's
one technology company thatimplemented a tool, you know,
these AI Zoom note takers thatlisten to your calls and then
(13:29):
they summarize.
That's fine.
This particular tool that theyimplemented actually does
psychoanalysis of the culture onyour call.
So it's listening to the call.
And then I read the report ofone call.
I watched a call and then I readthe report afterwards.
And um, the CEO was on the call,and it was a business review
(13:51):
call with one particulardepartment, and multiple
departments were represented onthe call.
And the CEO was cursing andsaying, Well, I don't know why
the F we would care about that,et cetera, and so forth, right?
And I knew this CEO, and I knewthat it's just a very casual
tech company, right?
I mean, these are the companiesthat they wear hoodies and it's
(14:12):
just like that.
No one on the call, based on myunderstanding of that particular
culture, was feeling like he wasbeing aggressive.
It was just vulgar and casual,right?
So the AI bot pumped out thisreport that said, well, this CEO
is cursing, and that can beinterpreted one of two ways.
Either he's being aggressive, orthis is a level of psychological
(14:34):
safety in which that is a normof behavior.
Based on the fact that thisother person cursed later in the
call, I believe it's because ofthe former.
I mean, it's like giving youthis opinion of the way that
culture is manifesting in theconversation on the Zoom call.
That's interesting.
Now, what are you gonna do withthat data, HR, right?
Are you gonna go give the CEOfeedback about cursing?
(14:56):
Are you gonna report to theboard that the CEO cursed and
this could mean one of twothings?
I don't know that that'snecessarily gonna move the
needle, but we're starting tosee ways in which AI is
psychoanalyzing people.
That's one interesting way.
I've seen other problematic waysat other companies that uh maybe
getting into legal hot water.
(15:17):
That's what I'll be watching.
SPEAKER_02 (15:20):
One of the things you mentioned in our
pre-discussion was you've seenorganizations using AI for
logic.
And I wasn't quite sure what youmeant by that.
SPEAKER_01 (15:27):
Yeah.
So, well, this actuallydovetails well from the previous
thing I was just describing, thepsychoanalysis.
So there's one organization thatbasically provides like
background checks as a service,and they implemented an AI tool
to look at the data that comesback from the background check
(15:49):
and then provide apsychoanalysis of the person
based on their background checkresults.
So typical background checksright now come back and they say
this person has a lien on theirhouse and they've been arrested,
or you know, they have whateverthe background check data is,
right?
It's just factual data presentedthat is given to an employer
(16:10):
when you're considering hiringsomeone.
Well, this exploration wasaround what does this mean for
the profile of someone likethis?
It's like, well, based on thefact that they had a DUI 15
years ago, but their credit wasreally bad and then it turned
really good.
We actually think this problemperson probably achieved
sobriety and therefore is on adifferent path of life.
(16:30):
And so it's a low-risk DUI ascompared to someone who had a
DUI three years ago and also isstill in bankruptcy, right?
I mean, that's what the AI toolwas doing.
It was analyzing someone, whichis, I mean, first of all,
illegal when it comes to whoyou're gonna hire and fire to
profile people in that way.
I was just talking to someonewho does a lot of consulting in
(16:52):
this space, and they have this.
I think you're not allowed touse AI to analyze candidates at
all, according to someregulation, but you'll have to
fact check me on that because Idon't actually know.
But what other use cases arethere for background checks?
I mean, it could be to whetherto hire or fire, but it could
also just be fornon-work-related things, private
(17:12):
investigators, for example,looking into people and trying
to piece together somethingthey've been hired to figure
out.
That's totally creepy in mymind, right?
And they pulled the plug onmoving forward with that because
they had the same creep factorcome up for them when they did
the analysis.
So, logic in that case is we'regonna take this data and we're
(17:35):
gonna make sense of it.
AI is really good at languagemodel, is creating language
around the input prompts.
But to logic, I don't know thatit's reliable enough
necessarily.
I have another example, is aprofessional services
organization that had a bunch ofdifferent um clients that they
serve.
And one of the things that theydid was they suggested that they
(17:58):
could match the clients'services with the clients' needs
that they had and using AI to dothat, asking AI to find, okay,
here's all the services thatthese clients have.
Here's also the needs that ourclients have.
Can we pair them with eachother?
That kind of logic is not veryeasy or effective or reliable
right now at the AI level.
(18:19):
And so that's what we're stillworking on figuring out how to
do well with AI.
SPEAKER_04 (18:25):
I appreciate the examples and also the
complexities that you'repointing to as well.
And we know in the Canadianmarketplace, if any part of a
decision is related toprotective grounds, um,
employers are on the wrong sideof employment law.
So if you had something predict,right, that someone was
suffering from an illness,addiction, then that would
(18:48):
absolutely create a ton of riskwithin the higher rank process.
SPEAKER_01 (18:53):
Well, I wonder how insurance companies are using AI
for this, right?
I mean, because they have topredict risk as part of their
business model.
I just had um the CHRO of AFLAC,or no, he's the chief strategy
officer, but he is the head ofHR, which is an interesting
title dynamic.
Anyway, um, they're in thebusiness of risk.
I should have asked him ifthey're using AI to predict risk
(19:16):
with their consumers, you know.
I mean, is that illegal?
I don't even know.
SPEAKER_04 (19:20):
So really underlining there's there's the
advances and there's thecautions, and just it comes back
to the importance of testing, ofpiloting, of making sure that
when we're applying a use case,it's just something that we have
we have a solid foundation ofand we've uh assessed the risk.
Before we move to risk, though,we'd like to dig a bit deeper on
the culture side and whetheryou've seen use cases around to
(19:45):
measure culture, to supportculture.
SPEAKER_01 (19:49):
Yeah.
So um I had a call with a vendorwho said that they one of the
big four consulting firms wastheir client, then they were
reselling it.
And this vendor has a tool inwhich it integrates with
Outlook, with Slack, with theinternal communication systems
(20:10):
to help facilitate bettercommunications.
So you're writing an email,basically, it gets a profile of
everyone in the company based onsome psychological assessment
that they've taken, right?
So let's say uh ACME Companyhires this vendor, right?
Every one of the employees atACME Company fills out a
(20:33):
psychological survey and itunderstands now this is how you
like to communicate, this is howlike you like to be communicated
with, this is your priority inworkplace dynamics.
Well, then I'm gonna write anemail to Joe at that company.
And I write the email, but thenbefore I send the email, this AI
bot is reading the email,understanding Joe's profile, and
(20:56):
making suggestions to me abouthow I could make my email
resonate better with Joe basedon his profile and the way he
likes to be communicated with.
So it'll say, well, Joe isreally likes to prioritize
relationship over tasks.
And so maybe you want to startwith asking how his day was,
right?
And it'll rewrite the email foryou before you send it to Joe,
(21:17):
which sounds great.
Then Joe receives the email.
It's a wonderful email in Joe'smind because it was carefully
crafted for his personalitytype.
And then Joe wants to writeback, blah, blah, blah.
My email, my weekend was great.
Thank you so much.
Let me send it back to you.
And then the AI bot sees Joe'snote to you and says, Oh, you
know what, Joe?
Jessica actually really preferswhen people just get down to
(21:39):
business.
Why don't you take out theweekend stuff and just make a
bullet point of tasks becausethat's how she likes to
communicate?
Joe's like, that sounds good.
He sends it off.
I think Joe is so great.
I love the way Joe communicates.
What's really happening thoughis that we're not communicating
with each other.
We've filtered how wecommunicate with each other
through this AI tool that'schanging it to be the way I like
(22:02):
to receive communication.
And I actually don't know muchabout Joe because Joe is not
talking to me the way Joe talksto me.
Joe is talking to me the way Ilike to be talked to through an
AI filter.
I haven't had any clientsimplement it, but I know clients
that they've listed that didimplement it.
And I have serious concernsabout what that.
I mean, their argument is it'sgonna make culture better
(22:24):
because everyone's gonna behappier with the way that it's
gonna reduce friction andcommunication.
But come on, I mean, is itreally communication if it's
just robots talking to robots,you know?
Thank you.
SPEAKER_04 (22:37):
That's such a just a great example of how we can
endeavor to improve somethingand remove friction and and
we're actually uh damaging itand without thinking through
just the consequences of peoplenot really getting to know each
other or understanding theirpreferences or getting to know
them in the way they want to beknown as as well.
(22:59):
Are there use cases that even ifyou haven't seen it, that excite
you that could actually moveculture forward in a way that
would be more helpful or waysthat you would like HR to
consider how to leverage it inan effective way rather than a
way that uh is more gettingalong through smoke and mares.
SPEAKER_01 (23:15):
Yeah.
So the instill AI example Igave, which is they listen to
the calls and then they do apsychoanalysis of those calls.
I've been excited about that.
And the reason I'm excited aboutthat is because the culture
partners, what we're alwaysdoing is trying to share with
people that if you really wantto get people's hearts and minds
aligned with the results you'retrying to create, you got to get
(23:37):
at the belief level, right?
And what instillAI wouldtheoretically be able to do is
track, measure, and analyzewhether those beliefs are in
place or not.
And if you see people not havingthose beliefs in place, that's a
predictive tool to show you thatyour change effort is not going
(23:58):
to be successful.
And it can flag to the leader,here's where your beliefs need
to shift.
It's like a mirror, right?
It also is a self-awareness toolbecause it can send a note to
the leader who cursed on thatcall and say, Hey, maybe you
want to think about not cursingon that call, which no one on
the call is gonna say to thatperson because they all report
(24:20):
into him, right?
So the self-awareness thing isthe big unlock that we need to
create better leaders.
We need leaders to be willing tobe self-aware.
And it's hard to get leadersfeedback because they're the
leaders.
They're in charge of your job,they're in charge of your
salary, they're in charge ofyour security at this company.
(24:40):
And so I'm not gonna tell myboss something he doesn't want
to hear at my own peril.
I'm not that principled.
I don't have that muchintegrity.
I want to make sure that I havea job so I can send my daughter
to school before I care aboutsaying the truth to power, you
know.
Sorry, that's just my perceptionof, you know, my priorities.
So this AI bot can speak totruth to power, though.
(25:02):
And I think that's interesting.
I'm excited about that.
SPEAKER_04 (25:06):
It's interesting because you think of the
examples that we're workingthrough, how many of them keep
coming back to supportingindividual productivity,
individual growth,self-awareness, and how that can
support the broaderorganization?
Yeah.
SPEAKER_02 (25:21):
Now, uh, Jessica, we've been focusing on some of
the interesting thingsorganizations who are leaning
into AI are doing.
What about those organizationsyou've seen who are failing to
make use of AI?
Do you see any commonalities inthose organizations that are
just very slow on adopting AI?
SPEAKER_01 (25:42):
I think more organizations are in that
category than not because it'sstill so new to transform an
organization around this brandnew technology requires a lot of
work.
And so you're seeing it inpockets based on individual
buy-in to the idea.
You're I I haven't seen, I mean,Fiverr just made this
(26:04):
announcement this week thatthey're gonna go back to being a
startup, they're laying off 250people and they're gonna be AI
first.
That's interesting, right?
The CEO of Fiverr, when he madethat announcement, got
absolutely slaughtered online.
I mean, he was just inundatedwith comments and messages about
how dare you! I'm never usingyour service again.
(26:25):
You know the worst thing aboutcorporate greed and blah, blah,
blah, right?
Because the CEO said, we'regonna be AI first and it
requires us to lay people off.
He is the 6,000th CEO to makethat decision.
And yet he's still getting thepushback.
Other CEOs, when they've doneit, what are they doing?
Are they replacing people withAI yet?
(26:46):
No, no one is.
I haven't seen any clients.
I have zero clients that havereplaced large groups of people
with AI efficiency.
I think that's going to comewith agentic AI, which is just
around the corner, but it hasn'thappened yet.
What you're seeing is peoplelaying off their populations of
employees to make room in thebudget by reducing labor costs
(27:08):
so that they can invest inexploring the possibility of AI
efficiencies.
It's it's technically replacingpeople, but not because the
efficiencies already exist.
It's because ideologically wewant it to exist.
So let's get rid of thesepeople, which is really, I think
AI has become a scapegoat foroverhiring and laying people
(27:30):
off, even though the LinkedIncomments look bad.
The market responds positivelyto layoffs now, which it didn't
10 years ago.
So you make an announcement thatyou're laying a large group of
people off, and the the marketresponds with great, you're a
forward-thinking,efficiency-based leader that's
results oriented.
We love this for you.
We we have we're positive aboutyour future.
(27:52):
10 years ago, or let's really go20 years ago, if you announced
layoffs, it meant you were atyour last straw before the
bankruptcy and breaking point ofyour company, and so you must be
about to fail.
That's not the perception of itanymore.
SPEAKER_04 (28:08):
Do you have any thoughts on how organizations
are governing AI, how they'remanaging risks, and would love
your thoughts on what you see asineffective, what you see as
effective.
SPEAKER_01 (28:23):
I think AI, the AI equivalent of cybersecurity, is
really in its nation stages, butwill be very, very popular as
more high-profile AI failuresbecome well known.
And there have been many ofthem, right?
I'm thinking of there was onereal estate organization that
(28:46):
started using AI to analyze whatproperties they should invest
in.
And the model was built based onhistorical economic data.
And so the model assumed thatthe economic situation was
static as it did its analysis.
Well, what happened is themarket shifted and the housing
(29:07):
market shifted, but the AI modelwas still based on assumptions
that we were in a different kindof economy.
And so it was popping out abunch of recommendations that
were totally badrecommendations.
And the company investedmillions of dollars in those
recommendations withoutrealizing that they were bad
recommendations because ithadn't adjusted for a new
(29:30):
economic reality.
That's a very straightforwardcase of AI taking a business
down a bad rabbit hole becauseit didn't have securities in
place.
And there's other things that AIcan do, the hallucination, for
example, right?
There's um healthcareinstitutions that have
implemented AI for, for example,analyzing radiology, right?
(29:53):
You analyze radiology using AI,and what you saw in one
particular case was it startedhallucinating.
Um, things and basically it wasdetecting problems, you know,
80% more of the time than itactually existed.
It's better than if it haddetected problems 80% less of
the time than it actuallyexisted, but it was just as
(30:14):
likely to go in one direction orthe other because it didn't have
this is the logic thing.
It didn't have the kind of logicthat can understand nuance and
context that the people readingthe radiology exam before did,
right?
That the people analyzing theinvestment portfolio before did.
And so that's where work stillneeds to be done.
And that's what AI quote unquotecybersecurity looks like.
(30:38):
It's different thancybersecurity where you're
looking, you know, you gotscammers coming in and trying to
get your data, right?
It's like, well, what is AI'serror point, failure point going
to look like?
And how can we protect ourselvesfrom that?
That will be, I think, uhsomething that gets popular next
year and the year after.
SPEAKER_04 (30:56):
We've seen practical ways organizations manage the
risks and opportunities.
SPEAKER_01 (31:02):
Yeah, people checking the work.
I mean, that's why I don't thinkyou've seen that many layoffs.
It's great, you've popped thisout.
Now we're gonna look at it andsee if it's remotely accurate or
not.
I mean, that is the peoplesaying that AI is not gonna
replace people because of that.
I think that's maybe true forabout four more months, you
know, in my opinion.
Uh I anyone who says they knowwhat the future of AI is going
(31:25):
to be, they're trying to sellyou something.
This is just my opinion.
It will get better.
It will replace people.
Agentic AI has so muchopportunity.
There are these big high-profilelike Klerna.
They laid off a bunch of people,said they were going to replace
the organization with agenticAI, and then they went back
because it failed miserably.
And now they want to be known asthe person who will guarantee
(31:46):
that you'll get a person on thephone.
I have one organization that's autility company that's exploring
the idea of replacing their callagents with AI.
They bought the tool, theyimplemented it.
It is uncanny.
I listened to a recording of acall agent helping a customer,
and the customer had no ideathey were talking to AI.
(32:08):
If I listened to a recording anddidn't know it is AI, I would
not have thought it was AI.
And yet, they're still notpulling the trigger because
they're worried about, well, arewe sure?
You know, it's just not, we'renot quite there yet, but it's
just around the corner.
Thank you.
SPEAKER_02 (32:24):
And just I would just wrap up my thoughts on
this, that one just has to moveforward with caution.
Explore, uh not necessarilybelieve what the vendor is
telling you.
Um, but if you explore, youmight find it works extremely
well, better than expected, oryou might find it's doing things
badly worse than you hadthought, and you just need to do
(32:48):
the pilots and keep an eye onhow it's going.
SPEAKER_01 (32:51):
Well, this is the this is the challenge of
capitalism and growth, right?
Innovation requires us to failfast and take risks.
And so what risks are youwilling to take?
This is these are the decisionsthat leaders have been making
for a long time.
Now they're pointing it at AI,right?
I mean, personally, and this isirrelevant to your listener
(33:11):
because who am I?
But personally, I'm a little bitscared of AI, right?
And I would love if everyonejust said, you know what, forget
it.
It's not worth it.
But I feel the same way aboutsocial media.
I feel the same way abouttechnology.
I mean, if it if I had mydrudhers, we'd go back to being
hunters and gatherers, you know?
I'd go back to pre-agriculturalage because I think that's
(33:33):
probably better for our mentalhealth.
But that's not the nature of theworld.
So here we are.
People are trying to make money,they're going to continue to
innovate, and that innovation isgoing to have consequences.
And so I'm here to try and helppeople navigate the reality of
the world the way that it is.
So the reality is no one's goingto slow down.
As long as they don't have to,they're not going to slow down.
(33:53):
And therefore, as an employee,as a person in this world, how
can you adapt to deal with thatreality?
It's again going back to thisidea of surrender.
You can resist that the futureis AI because you think it's
unsafe, or you can accept thatit's happening.
And so, therefore, what am Igoing to do in this new reality?
SPEAKER_04 (34:14):
Any practical tips in that regard?
Like how leaders can accept thepractical reality, how they
support their teens acceptingthe practical reality, how they
support enablement,democratization of capabilities
in this new realm.
SPEAKER_01 (34:30):
Number one, so I have a model in the book.
It's called the shift model thatis around this.
We don't have to go through thewhole shift model, but first
step is stop fighting reality.
And reality is everythinghappening outside of you.
We've spend so much time in ourheads wishing our employees were
different.
If only they were more engaged,if only the leaders created
(34:52):
lower objectives for me to hit.
If only the administrationdidn't blank, if only the
competitors weren't X, I mean,there's so much, it should be
this way, if only it were thatway.
That think of how much energyand time you spend spinning your
wheels on that.
That's a lack of surrender,right?
We call that below the linethinking.
(35:12):
So to go above the line thinkingis to say, okay, what's going
on?
And what about that can Icontrol?
Let me just put my energy there.
And if I just focus on makingthe personal choice to focus on
what I can control to driveresults, well, now I've at least
focused my energy in somethinguseful instead of spinning my
(35:34):
wheels on the way I wish it was.
Then once you've identifiedwhere you actually can make an
impact, well, then you can makedecisions about how you can show
up to create the rightexperiences that will shape the
right beliefs, that will get theright actions, that will get the
results.
I mean, I think the number onething people need to do is get
(35:56):
out of the action trap ofconstantly trying to do
something, that endless cycle ofactivity that feels like
progress, but is really notmoving the needle.
And get out from below the linethinking where you're pointing
fingers, you're like, well,that's not my job, and it should
be this way.
And it, those folks need to bedifferent.
All of that is wasted energy.
(36:16):
There's something in betweenthat I think is where we need to
spend our time.
SPEAKER_04 (36:20):
Thank you, Jessica.
This has been a short andinteresting conversation for us.
For those who would like to getin contact with you, what's the
best way to do so?
SPEAKER_01 (36:31):
You can go to my website, jessica.com.
Uh Kriegel is calledK-R-I-E-G-E-L, or you can go to
our company's website, which isculturepartners.com.
And I have a weekly newsletteron LinkedIn if you want to join
that.
I pop out insights everyWednesday called This Week in
Culture, and it just tells youwhat's going on in the world of
culture.
(36:51):
Thank you.
SPEAKER_03 (36:56):
Thanks for listening to the HR Chat Show.
If you enjoyed this episode, whynot subscribe and listen to some
of the hundreds of episodespublished by HR Gazette?
And remember, for what's new inthe world of work, subscribe to
the show, follow us on socialmedia, and visit hrgazette.com.
Popular Podcasts
Las Culturistas with Matt Rogers and Bowen Yang
Ding dong! Join your culture consultants, Matt Rogers and Bowen Yang, on an unforgettable journey into the beating heart of CULTURE. Alongside sizzling special guests, they GET INTO the hottest pop-culture moments of the day and the formative cultural experiences that turned them into Culturistas. Produced by the Big Money Players Network and iHeartRadio.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.