In this episode of the Future Proof HR podcast, Thomas Kunjappu, CEO of Cleary, sits down with Fernando Garcia,  Chief People and Legal Officer at Opta Group (at the time of recording), to unpack what HR leaders are getting wrong about AI, and what they should be doing instead.

Fernando brings a rare blend of experience across HR, legal, compliance, and business strategy, and he makes a clear case for why AI will not replace HR, but it will reshape the job. The work that remains, he argues, is the work that matters most: human judgment, context, experience, and the ability to make value-based decisions when the answer is not obvious.

Together, they explore the practical reality of “shadow AI” already happening inside organizations, why guardrails matter more than hype, and how HR can work with legal as an early strategic partner rather than an emergency hotline once something has already escalated. The conversation also goes deep on the ethical tension of AI-driven hiring and fairness, including how bias can show up on both sides, whether decisions are made by people or machines, and why the “trolley problem” isn’t just a thought experiment anymore.

This episode is for HR leaders who want to adopt AI responsibly, without parking their judgment at the door, and who want to future-proof their work by leaning harder into the human side of leadership.

Topics Discussed:

1. AI will change HR jobs, not eliminate them
2. Why HR’s value is judgment, context, and the human element
3. The “three-legged stool” of business, legal, and people leadership
4. How HR and legal can partner early to prevent risk, not just react to it
5. The reality of “shadow AI” and why assuming no one uses AI is the biggest risk
6. Practical guardrails: data privacy, PII, sensitive employee info, and due diligence
7. Where AI helps today: drafting, surveys, training design, and early recruiting support
8. The ethics of AI in recruiting, fairness, and bias on both sides of the argument
9. Why culture, training, and peer learning matter more than expensive enablement programs
10. Skills that will matter most for future-proof HR: curiosity, EQ, relationship-building, and broad capability

If you’re an HR leader trying to balance innovation with compliance, curious about AI’s real use cases beyond automation, or navigating how to adopt these tools without losing the human element, this episode offers a grounded and thoughtful framework.

Additional Resources:

1. Cleary’s AI-powered HR Chatbot
2. Future Proof HR Community
3. Connect with Fernando Garcia on LinkedIn

Are you nervous about AI taking away work?

2

00:00:03,629 --> 00:00:06,049

I think no, I think it's going

to change our jobs and it's

3

00:00:06,049 --> 00:00:07,274

going to change what we do.

4

00:00:07,274 --> 00:00:10,874

But the value that we truly

add is that value judgment.

5

00:00:11,094 --> 00:00:12,024

That knowledge base.

6

00:00:12,024 --> 00:00:14,089

That experience that you're

bringing into your role.

7

00:00:14,334 --> 00:00:17,124

And I think that human touch,

that human element of it is

8

00:00:17,124 --> 00:00:18,084

really what adds our value.

9

00:00:18,084 --> 00:00:19,794

And I think that's no matter what,

that's never gonna be taken away.

10

00:00:20,514 --> 00:00:22,314

Thomas Kunjappu: They keep

telling us that it's all over.

11

00:00:22,314 --> 00:00:26,334

For HR, the age of AI is upon

us, and that means HR should

12

00:00:26,334 --> 00:00:27,984

be prepared to be decimated.

13

00:00:28,494 --> 00:00:29,844

We reject that message.

14

00:00:30,084 --> 00:00:32,183

The future of HR won't be handed to us.

15

00:00:32,334 --> 00:00:37,434

Instead, it'll be defined by those

ready to experiment, adopt, and adapt.

16

00:00:38,034 --> 00:00:42,894

Future Proof HR invites these builders to

share what they're trying, how it's going,

17

00:00:43,134 --> 00:00:44,994

what they've learned, and what's next.

18

00:00:45,474 --> 00:00:49,404

We are committed to arming HR

with the AI insights to not

19

00:00:49,404 --> 00:00:51,084

just survive, but to thrive.

20

00:00:56,134 --> 00:01:00,364

Hello and welcome to the Future Proof

HR podcast, where we explore how

21

00:01:00,364 --> 00:01:04,444

forward-thinking HR leaders are preparing

for disruption and redefining what it

22

00:01:04,444 --> 00:01:07,024

means to lead people in a changing world.

23

00:01:07,804 --> 00:01:10,714

I'm your host, Thomas

Kunjappu, CEO of Cleary.

24

00:01:11,134 --> 00:01:13,114

Today's guest is Fernando Garcia,

25

00:01:13,194 --> 00:01:15,884

Vice President Corporate

Services at Cargojet.

26

00:01:16,319 --> 00:01:21,329

With 15-plus years of experience across

general counsel, HR and labor relations

27

00:01:21,329 --> 00:01:25,329

compliance and corporate secretary

roles, plus a BA in labor studies and a

28

00:01:25,329 --> 00:01:27,769

master's in industrial relations and HR.

29

00:01:28,429 --> 00:01:32,239

Dual civil and common law degrees,

and an MBA in strategic management.

30

00:01:33,109 --> 00:01:36,589

I think it's fair to say, Fernando

brings a rare T-shaped blend of legal

31

00:01:36,589 --> 00:01:38,719

depth and business people breadth.

32

00:01:39,229 --> 00:01:43,069

So he leads at the intersection

of law, business, and innovation.

33

00:01:43,489 --> 00:01:48,979

And advocates for pragmatic adoption of

legal and HR tech to unlock efficiency

34

00:01:49,249 --> 00:01:51,378

while at the same time managing risk.

35

00:01:51,529 --> 00:01:53,329

Fernando, welcome to the podcast.

36

00:01:53,674 --> 00:01:54,454

Fernando Garcia: Thank you very much.

37

00:01:54,454 --> 00:01:56,884

It's a pleasure to be here and,

looking forward to our conversation.

38

00:01:57,869 --> 00:01:58,479

Thomas Kunjappu: Absolutely.

39

00:01:58,479 --> 00:02:03,349

So, I'd love to set up our

conversation with this concept of

40

00:02:03,349 --> 00:02:06,739

the three-legged stool I think we

were talking about a bit earlier.

41

00:02:07,189 --> 00:02:10,219

We've got business, legal,

risk management, and people.

42

00:02:10,848 --> 00:02:14,539

Now you've pursued degrees and have

worked in a combination of all three.

43

00:02:14,689 --> 00:02:16,039

How do you see that all coming together?

44

00:02:16,159 --> 00:02:18,499

Fernando Garcia: When I look at

what makes a business successful,

45

00:02:18,919 --> 00:02:22,728

I think, and looking back at it

when I first started my career, I

46

00:02:22,728 --> 00:02:23,959

thought there were three elements.

47

00:02:23,959 --> 00:02:25,629

One is obviously the people.

48

00:02:25,849 --> 00:02:27,489

The HR part is critical.

49

00:02:27,489 --> 00:02:30,439

It is what distinguishes

one company from another.

50

00:02:30,439 --> 00:02:35,389

That is what makes a place attractive

for someone to go work at and stay there.

51

00:02:35,849 --> 00:02:37,599

The other part is legal.

52

00:02:37,599 --> 00:02:43,879

We're all involved in an environment

of laws and regulations and compliance,

53

00:02:44,029 --> 00:02:45,409

and that affects how we do our job.

54

00:02:45,409 --> 00:02:49,099

That affects how companies go to business

and which companies are successful.

55

00:02:49,279 --> 00:02:53,299

And then obviously the business

component of it, the business strategy,

56

00:02:53,299 --> 00:02:57,019

the planning, and, making sure your

products are meeting the needs of

57

00:02:57,289 --> 00:03:01,099

society and they're adapting as

necessary to be able to be successful

58

00:03:01,099 --> 00:03:03,639

in your environment or your market.

59

00:03:04,039 --> 00:03:08,359

So when I was looking and thinking, how

do I wanna shape my career moving forward?

60

00:03:09,079 --> 00:03:12,228

I thought those are three elements

that I really wanted to touch on,

61

00:03:13,219 --> 00:03:15,169

become experienced in and really

62

00:03:17,249 --> 00:03:18,019

gather experience.

63

00:03:18,019 --> 00:03:23,478

So one was the educational background,

which you've read already, quite exciting.

64

00:03:23,779 --> 00:03:26,959

My wife always complains that if I

could, I would be a lifetime student.

65

00:03:26,959 --> 00:03:29,689

And I think in many ways,

we are all lifetime students

66

00:03:29,689 --> 00:03:31,039

because we never stop learning.

67

00:03:31,039 --> 00:03:32,239

We never stop developing.

68

00:03:32,239 --> 00:03:32,969

We never stop growing.

69

00:03:33,339 --> 00:03:35,349

And if we do, that's

when things get boring.

70

00:03:35,749 --> 00:03:37,069

So we're always looking at that.

71

00:03:37,069 --> 00:03:41,059

But those are three areas that

are of interest to me in terms of

72

00:03:41,059 --> 00:03:43,939

education, but also in terms of

what I do on a day-to-day basis.

73

00:03:44,089 --> 00:03:48,918

I don't want to be just a lawyer, and

I don't want to be just an HR person.

74

00:03:48,918 --> 00:03:53,329

I want to be a business person

with an HR and legal background.

75

00:03:53,394 --> 00:03:55,759

Thomas Kunjappu: I love that

concept of combining these two,

76

00:03:55,759 --> 00:03:57,114

so it's like one C two functions.

77

00:03:58,009 --> 00:04:02,919

but tell me where you see the

strongest synergies between

78

00:04:02,919 --> 00:04:06,079

specifically HR and legal.

79

00:04:06,079 --> 00:04:09,069

I mean, my mind and probably everyone's

mind goes to like employee relations

80

00:04:09,069 --> 00:04:16,009

and like, just making sure that

you're, you know, firing and doing

81

00:04:16,039 --> 00:04:17,629

things in a compliant way, right.

82

00:04:17,629 --> 00:04:18,889

And on the people side.

83

00:04:19,769 --> 00:04:24,249

But yeah, tell me about how you see the

two coming together in your day to day.

84

00:04:24,939 --> 00:04:26,889

Fernando Garcia: I think a lot

of it is also risk management.

85

00:04:27,549 --> 00:04:30,974

Some of those elements that you mentioned

are very important on a day-to-day.

86

00:04:31,644 --> 00:04:34,874

Making sure you avoid litigation,

making sure you're following proper

87

00:04:34,874 --> 00:04:38,274

processes and hiring, firing, and

all those day-to-day things of HR.

88

00:04:38,454 --> 00:04:42,933

But there's an element of HR that

looks longer-term, more strategic, more

89

00:04:42,933 --> 00:04:47,634

longer term value-based, and the legal

environment and the legal background also

90

00:04:47,634 --> 00:04:50,034

helps you mitigate risk in the long run.

91

00:04:50,183 --> 00:04:54,084

So helping strengthen your brand

and making your brand stronger, both

92

00:04:54,084 --> 00:04:57,324

locally and internationally, and

understanding what are the things that

93

00:04:57,324 --> 00:04:58,854

are gonna impact it so that you can.

94

00:04:59,289 --> 00:05:01,149

Be successful in that environment.

95

00:05:01,149 --> 00:05:04,539

And part of that is a huge part of

it is having the right people in the

96

00:05:04,539 --> 00:05:09,069

right jobs and feeling like they wanna

stay and be developed, have their

97

00:05:09,069 --> 00:05:12,939

skill set acknowledged really fully

developed within the corporation.

98

00:05:13,119 --> 00:05:17,589

So it's in the short term and in the long

run that I think there's a ton, tons of

99

00:05:17,589 --> 00:05:19,659

synergies between HR and legal itself.

100

00:05:20,764 --> 00:05:23,304

Thomas Kunjappu: When you say 'short

term and long run,' what do you mean?

101

00:05:23,304 --> 00:05:26,004

Like in day-to-day

tactical things, but also.

102

00:05:26,694 --> 00:05:30,264

Fernando Garcia: If something happens

and I need support or I need your

103

00:05:30,264 --> 00:05:33,264

advice regarding this particular

issue that we're dealing with today.

104

00:05:33,324 --> 00:05:36,144

But also thinking about longer-term

in terms of where businesses are

105

00:05:36,144 --> 00:05:40,554

going and what are some of the

unique needs of risk mitigation,

106

00:05:41,304 --> 00:05:43,164

development and all the fun stuff.

107

00:05:43,164 --> 00:05:44,964

And AI walks right into that.

108

00:05:46,294 --> 00:05:50,984

Thomas Kunjappu: So okay, let

me ask for some advice because a

109

00:05:50,984 --> 00:05:56,974

lot of HR leaders don't have your

background in and training in law.

110

00:05:57,004 --> 00:06:02,844

But of course are very much in the weeds

or their teams are in compliance, right?

111

00:06:02,859 --> 00:06:03,069

Fernando Garcia: Yes.

112

00:06:03,284 --> 00:06:10,209

Thomas Kunjappu: So what have you seen

or what would be your advice for HR

113

00:06:10,209 --> 00:06:14,649

leaders in terms of like how they can

best work with the legal department?

114

00:06:15,564 --> 00:06:17,914

Fernando Garcia: Yeah, I think

it's also a function of the

115

00:06:17,914 --> 00:06:21,934

mindset of your legal department.

116

00:06:21,954 --> 00:06:25,224

There's legal departments that

are like I always say that the

117

00:06:25,224 --> 00:06:28,974

people who are in charge of forms

347C, and all they do is 347C.

118

00:06:28,974 --> 00:06:31,594

And if you have a legal question

about that form, you go to them.

119

00:06:31,984 --> 00:06:34,689

But then there's also the

legal departments And I've been

120

00:06:34,689 --> 00:06:36,099

advocating for a long time.

121

00:06:36,519 --> 00:06:39,499

you mentioned that the T-shaped

lawyer or I called it the plus-shaped

122

00:06:39,519 --> 00:06:42,099

lawyer in one of my articles

that I wrote many years ago now.

123

00:06:43,209 --> 00:06:46,719

Not just being a legal department

that handles legal questions, but

124

00:06:46,719 --> 00:06:49,668

that it's part of the business

integrated into the business.

125

00:06:49,668 --> 00:06:51,121

And that's the case.

126

00:06:51,121 --> 00:06:56,205

I think HR and legal working together

really help to fill the void to

127

00:06:56,205 --> 00:07:00,562

mitigate risk, which is the things

that are very important for us.

128

00:07:00,562 --> 00:07:03,831

And for anyone really operating

in an environment, right?

129

00:07:03,831 --> 00:07:09,954

Because you don't want to create

any kind of a situation or in the

130

00:07:09,954 --> 00:07:11,064

environment or to the company.

131

00:07:12,774 --> 00:07:17,444

Thomas Kunjappu: If you're talking to

like your friends, colleagues in HR

132

00:07:17,704 --> 00:07:22,704

who are not obviously lawyers as well.

133

00:07:24,844 --> 00:07:27,879

Like what are the kinds of things

that were you or situations,

134

00:07:28,029 --> 00:07:32,979

did any situations come to mind

where you feel like this was I

135

00:07:35,129 --> 00:07:38,259

wish in this case the HR

department had like checked in or

136

00:07:38,259 --> 00:07:40,319

worked more closely with legal.

137

00:07:40,909 --> 00:07:44,919

And are there any kinds of situations

that keep coming up in your mind where

138

00:07:46,479 --> 00:07:51,339

you know, the HR function could work

more tightly with the legal department?

139

00:07:52,404 --> 00:07:55,194

Fernando Garcia: I think if you look

at your legal team as a business

140

00:07:55,194 --> 00:07:58,464

strategic advisor and really

part of the business itself, then

141

00:07:58,464 --> 00:08:00,704

you get them involved early on.

142

00:08:00,704 --> 00:08:04,674

The biggest issue or what I look at one

of the greatest limitations is when you

143

00:08:04,674 --> 00:08:06,204

get involved with legal but too late.

144

00:08:06,564 --> 00:08:10,614

Once something has already blown up,

once something's already a conflict or

145

00:08:10,614 --> 00:08:12,144

there's a potential litigation issue.

146

00:08:12,384 --> 00:08:15,594

At that point, it's much harder

because your tools or the available

147

00:08:15,714 --> 00:08:17,214

outcomes are already much more limited.

148

00:08:17,454 --> 00:08:20,064

But if you get them involved early

on, if the relationship is one of

149

00:08:20,064 --> 00:08:23,694

constant communication, interaction

and working together on things, you

150

00:08:23,694 --> 00:08:26,774

can be proactive and address the

issues before they even become issues.

151

00:08:27,294 --> 00:08:30,054

And that's where when you

add the most value, because

152

00:08:30,054 --> 00:08:31,224

you're working hand in hand.

153

00:08:31,884 --> 00:08:34,073

And again, I've seen many

situations of people who are

154

00:08:34,073 --> 00:08:37,804

lawyers who go and in terms of HR.

155

00:08:38,063 --> 00:08:40,299

And I've seen the other way

around too, HR people who are very

156

00:08:40,299 --> 00:08:43,568

specialized, especially in areas like

labor relations, and then they go

157

00:08:43,709 --> 00:08:43,948

Thomas Kunjappu: Oh.

158

00:08:43,959 --> 00:08:46,449

Fernando Garcia: they take

on a quasi-legal function,

159

00:08:46,764 --> 00:08:47,094

Thomas Kunjappu: Sure.

160

00:08:47,229 --> 00:08:48,189

Fernando Garcia: their organization.

161

00:08:48,399 --> 00:08:49,419

So it goes both ways.

162

00:08:49,419 --> 00:08:49,759

But I,

163

00:08:49,794 --> 00:08:50,034

Thomas Kunjappu: Yeah.

164

00:08:50,049 --> 00:08:53,529

Fernando Garcia: find that there's a

lot of synergies there in terms of the

165

00:08:53,529 --> 00:08:55,509

functions and the roles that we play.

166

00:08:56,784 --> 00:08:59,874

Thomas Kunjappu: So, you talked a

little bit about that mindset then.

167

00:09:00,204 --> 00:09:04,464

The T-shaped mindset, but

also the ability to be like a

168

00:09:04,464 --> 00:09:05,814

focus on the business outcome.

169

00:09:05,844 --> 00:09:08,184

Like regardless of kind of function.

170

00:09:08,604 --> 00:09:15,189

So sounds like that might apply equally

to HR and legal on both sides, right?

171

00:09:16,029 --> 00:09:21,069

But you also mentioned how that

mindset may be actually even like

172

00:09:21,079 --> 00:09:23,209

a perfectly fit for the age of AI.

173

00:09:23,659 --> 00:09:27,069

So tell me about that.

174

00:09:27,139 --> 00:09:28,099

From both perspectives.

175

00:09:28,149 --> 00:09:32,079

And especially from a risk mitigation

and compliance perspective, which is what

176

00:09:32,079 --> 00:09:33,939

we've been talking about so far, right?

177

00:09:34,359 --> 00:09:38,129

How do you think about AI,

especially with that mindset of

178

00:09:38,129 --> 00:09:41,889

being flexible, agile, T-shaped?

179

00:09:42,819 --> 00:09:48,669

Fernando Garcia: Yeah, I think AI brings

another element to our toolkit and it's

180

00:09:48,669 --> 00:09:54,489

a critical tool in terms of helping take

away maybe some of the more administrative

181

00:09:54,489 --> 00:09:58,239

functions or responsibilities and

really focusing more on the value-added.

182

00:09:58,254 --> 00:10:03,294

Giving you another person that they, it's

a virtual person that you can run ideas by

183

00:10:03,534 --> 00:10:04,674

come up with alternatives.

184

00:10:04,854 --> 00:10:08,994

Now again, I wouldn't say it's a

professional partner because I think

185

00:10:08,994 --> 00:10:10,554

there's still some development to go.

186

00:10:11,124 --> 00:10:13,494

and you've got to be somewhat critical

and you've got to be careful about

187

00:10:13,584 --> 00:10:16,314

what information is given to you and

how you are using that information.

188

00:10:16,314 --> 00:10:19,344

So you've got to check in and you

still have to go do your due diligence.

189

00:10:19,404 --> 00:10:19,644

Thomas Kunjappu: Sure.

190

00:10:19,644 --> 00:10:22,564

Fernando Garcia: I think it's a critical

tool in terms of helping take away

191

00:10:22,714 --> 00:10:26,619

the more intensive administrative

functions and really helping you focus

192

00:10:26,619 --> 00:10:29,889

on where you can add the greatest

value in terms of your organization

193

00:10:29,889 --> 00:10:31,749

and even your function itself.

194

00:10:32,709 --> 00:10:35,059

With the caveat you've got to

be careful with it as well.

195

00:10:35,059 --> 00:10:35,619

Thomas Kunjappu: And I

196

00:10:35,619 --> 00:10:36,979

Yeah, let's get very specific then.

197

00:10:36,979 --> 00:10:40,894

Like what's your safe list or

the more exciting things to talk

198

00:10:40,894 --> 00:10:42,964

about is like, don't do this.

199

00:10:42,964 --> 00:10:46,204

These areas are just like,

you know, you've got to be

200

00:10:46,204 --> 00:10:47,914

really careful about with AI.

201

00:10:47,914 --> 00:10:50,614

But how do you or maybe you

can kind of answer both.

202

00:10:50,614 --> 00:10:51,214

Like, how do you think of it?

203

00:10:51,214 --> 00:10:52,634

What is safe?

204

00:10:52,844 --> 00:10:56,984

What are you gonna go into, what's

okay to experiment with versus where

205

00:10:56,984 --> 00:10:59,144

do you really need to be careful?

206

00:10:59,509 --> 00:11:01,589

Fernando Garcia: Yeah, I think

like anything, I want to give you

207

00:11:01,609 --> 00:11:03,299

lawyer answer, but it depends.

208

00:11:03,319 --> 00:11:06,379

really the fundamental here is to make

sure that you're understanding terms and

209

00:11:06,379 --> 00:11:07,969

conditions in terms of what you're using.

210

00:11:08,099 --> 00:11:10,919

If you're putting in private information,

make that you understand that it stays

211

00:11:10,919 --> 00:11:15,049

and contains and, working IT department

to make sure it's a closed system versus

212

00:11:15,099 --> 00:11:18,219

the information putting in all of a

sudden becomes part of the public domain.

213

00:11:18,269 --> 00:11:21,359

you're putting in sensitive employee

information in there, possible health

214

00:11:21,359 --> 00:11:24,959

information, which is obviously, HIPAA

and some of the other concerns that

215

00:11:24,959 --> 00:11:26,699

you have in that particular case.

216

00:11:26,759 --> 00:11:30,459

But if you have all that covered

up that issue is not a concern.

217

00:11:30,589 --> 00:11:34,339

It does open up the possibilities

in terms of how you can use it.

218

00:11:34,339 --> 00:11:38,024

Just mindful of whatever information

you put in there, you have to double

219

00:11:38,024 --> 00:11:40,184

check, you have to review it's safe.

220

00:11:40,404 --> 00:11:43,944

Not gonna have loss of privilege

issues or you're putting out

221

00:11:43,944 --> 00:11:45,504

personal information onto the domain.

222

00:11:45,834 --> 00:11:49,404

But in terms of that element of it, I

think there's a lot of opportunities

223

00:11:49,404 --> 00:11:54,544

too in terms of recruiting in terms of

drafting letters, in terms of just really

224

00:11:54,634 --> 00:11:56,284

sometimes even just getting you started.

225

00:11:56,284 --> 00:11:58,144

But the key is you

still have to review it.

226

00:11:58,174 --> 00:12:00,664

'Cause it does tend to

hallucinate at times.

227

00:12:01,039 --> 00:12:03,979

And there's also the concept of

garbage in is garbage out, right?

228

00:12:03,979 --> 00:12:07,129

So you have to make sure that whatever

you're, how you're prompting it or

229

00:12:07,129 --> 00:12:10,159

what information you're feeding it

is gonna affect what the outcome is.

230

00:12:10,219 --> 00:12:12,259

And you got too be mindful

of that part of it as well.

231

00:12:13,674 --> 00:12:15,014

Thomas Kunjappu: So I guess

there's a lot to consider.

232

00:12:15,259 --> 00:12:16,249

Are there, but.

233

00:12:17,929 --> 00:12:21,039

it sounds like, I think you're saying

there there's a clear set in your

234

00:12:21,039 --> 00:12:25,329

mind, of areas you have to like

make sure that you're bulletproof

235

00:12:25,329 --> 00:12:27,889

on in terms of risk mitigation.

236

00:12:27,959 --> 00:12:28,279

Right?

237

00:12:28,279 --> 00:12:29,379

And data privacy.

238

00:12:29,379 --> 00:12:35,729

And once that's in place, would you say

that's pretty universal is it really in

239

00:12:35,729 --> 00:12:41,744

your view, besides cultural expectations

where some companies are more AI forward

240

00:12:41,744 --> 00:12:44,054

versus, or innovative versus others,

241

00:12:44,734 --> 00:12:45,174

Fernando Garcia: Yeah.

242

00:12:45,284 --> 00:12:47,924

Thomas Kunjappu: like there's like

a kind of a generic sense of, you

243

00:12:47,924 --> 00:12:52,084

know set of guardrails across all

industries, across all companies?

244

00:12:52,414 --> 00:12:59,594

Like, hey, no training with this

data for the core LLM models, there's

245

00:12:59,744 --> 00:13:02,429

some kind of like onboarding and

training for every employee who's

246

00:13:02,429 --> 00:13:03,809

gonna have access to these things.

247

00:13:04,679 --> 00:13:08,639

PII or customer data should not be on

this or this is the specific guardrails.

248

00:13:08,639 --> 00:13:09,599

And there's trainings on that.

249

00:13:09,749 --> 00:13:11,159

You know, there's a set of

things you could come up with.

250

00:13:11,219 --> 00:13:14,009

And do you feel like that's like pretty

universal and really in the only gap

251

00:13:14,009 --> 00:13:17,609

is cultural or in your view like from

what you can tell, it's just like, it

252

00:13:17,609 --> 00:13:22,739

just, every industry is like different

and you really have like nuances about

253

00:13:22,954 --> 00:13:23,134

Fernando Garcia: Yeah.

254

00:13:23,369 --> 00:13:24,839

Thomas Kunjappu: is

what the guardrails are.

255

00:13:25,374 --> 00:13:27,934

Fernando Garcia: I think it's

beyond just every industry.

256

00:13:27,934 --> 00:13:33,184

I think every company, every

location is at a sort of a

257

00:13:33,184 --> 00:13:34,564

different stage in their journey.

258

00:13:34,939 --> 00:13:37,729

And some people are still

very much experimental.

259

00:13:38,119 --> 00:13:42,069

Some people are all in and they

want all their employees to have

260

00:13:42,069 --> 00:13:45,069

access, all their employees to be

properly trained and other employees.

261

00:13:45,459 --> 00:13:49,179

The reality though is that most

people, whether it's on a personal

262

00:13:49,179 --> 00:13:52,119

basis or at work, are using something.

263

00:13:52,769 --> 00:13:55,419

you've got to make sure that you train

them up in terms of saying, look, if

264

00:13:55,419 --> 00:13:59,079

you don't have the proper guardrails

in place, still use it, but be mindful.

265

00:13:59,079 --> 00:14:00,879

Don't put employee information in there.

266

00:14:00,969 --> 00:14:03,369

Don't put names, don't put

personal information, don't put

267

00:14:03,369 --> 00:14:06,009

stuff about products I haven't

released to the public yet.

268

00:14:06,099 --> 00:14:11,769

Don't put market share data sensitive

pricing information unless you are at

269

00:14:11,769 --> 00:14:15,059

that stage where you've now checked

the box and everything's okay,

270

00:14:15,239 --> 00:14:17,099

and there isn't that risk element.

271

00:14:17,704 --> 00:14:20,944

So every company, every location

will be at a different stage of that

272

00:14:20,944 --> 00:14:25,174

journey, but I think the biggest risk

of them all is assuming that nobody in

273

00:14:25,174 --> 00:14:27,454

your company or no employees using AI.

274

00:14:27,664 --> 00:14:31,444

because I think more and more, whether

it's just to write a greeting card or

275

00:14:31,444 --> 00:14:37,024

to start an email, or really to check

a report or review something, or even

276

00:14:37,024 --> 00:14:39,244

draft, initial stages of reports.

277

00:14:39,364 --> 00:14:40,114

People are using it.

278

00:14:40,444 --> 00:14:42,874

So you have to be mindful of that

and you've got to be mindful of

279

00:14:42,934 --> 00:14:45,964

how knowledgeable are they of

it and how can they protect the

280

00:14:45,964 --> 00:14:47,224

information as much as possible.

281

00:14:47,244 --> 00:14:47,534

Thomas Kunjappu: Yeah.

282

00:14:47,584 --> 00:14:49,954

Fernando Garcia: So some minor

training I think is critical.

283

00:14:49,954 --> 00:14:51,184

Regardless of where you

are in that journey.

284

00:14:52,594 --> 00:14:52,744

Thomas Kunjappu: Yeah.

285

00:14:52,744 --> 00:14:57,154

So regardless of whether you've

adopted wall to wall some kind of

286

00:14:57,154 --> 00:14:59,644

like LLM enterprise solution or not.

287

00:15:00,064 --> 00:15:02,524

In all likelihood, someone

out there is using it.

288

00:15:03,214 --> 00:15:06,754

In previous years, we'd had this

concept called like shadow IT and that

289

00:15:06,754 --> 00:15:11,474

when SaaS was first coming into the

fore and now there's like Shadow AI.

290

00:15:11,704 --> 00:15:15,714

That there's usage happening and

it's similar things, like your

291

00:15:15,714 --> 00:15:18,714

data could be leaking without

the right kind of guardrails.

292

00:15:19,914 --> 00:15:24,324

I'm very curious, given your two

halves here, where do you think there's

293

00:15:24,324 --> 00:15:31,374

more, there's been more adoption of

AI tools within the legal kind of

294

00:15:31,374 --> 00:15:35,074

community for all types of contract

law, employee law, there's all types

295

00:15:35,074 --> 00:15:39,244

of paralegal type of work or the HR

world which, you know, our audience

296

00:15:39,244 --> 00:15:40,234

obviously like, knows a lot about.

297

00:15:40,234 --> 00:15:42,874

But do you think there's a

I don't know, an emergent

298

00:15:45,844 --> 00:15:49,604

any differences in terms of adoption

or like, if you see like the different

299

00:15:49,604 --> 00:15:53,844

communities that you're traveling in and

the general stance towards these tools?

300

00:15:54,374 --> 00:15:57,244

Fernando Garcia: Yeah, For example, we

were talking about the legal industry.

301

00:15:57,244 --> 00:15:58,774

I think you can break

it up into two parts.

302

00:15:58,774 --> 00:16:02,284

Because you have the in-house counsel

and then you have the external

303

00:16:02,284 --> 00:16:04,024

lawyers who work at law firms.

304

00:16:04,714 --> 00:16:08,014

Big law firms tend to have more resources,

tend to have more training and tend to

305

00:16:08,014 --> 00:16:12,744

have more of that giving them the tools

and having the parameters properly set.

306

00:16:12,744 --> 00:16:18,229

In-house counsel, depending on the

size of the companies they work at they

307

00:16:18,229 --> 00:16:21,079

can have that available to employees

and training and everything else.

308

00:16:21,079 --> 00:16:23,959

Or you could be in a smaller

organization where there's less

309

00:16:23,959 --> 00:16:25,689

of that, less of that training.

310

00:16:26,219 --> 00:16:29,239

But I think in HR, same concept.

311

00:16:29,409 --> 00:16:33,959

I hear from people who are very much just

drafting emails with it or doing some

312

00:16:33,959 --> 00:16:39,904

preliminary to people who are getting

resumes for a particular role and they

313

00:16:39,904 --> 00:16:43,234

run a preliminary search through the a

hundred resumes, say, pick me the top 10

314

00:16:43,234 --> 00:16:44,734

based on the following job description.

315

00:16:45,664 --> 00:16:48,034

But again, you've got to be

very mindful of that as well.

316

00:16:49,064 --> 00:16:50,944

That you've got to be careful

that you're still doing that.

317

00:16:50,974 --> 00:16:54,484

Your proper check and your proper

due diligence because I hate to

318

00:16:54,484 --> 00:17:00,249

be in a situation where people are

using AI to create their resumes.

319

00:17:00,519 --> 00:17:03,549

Then they're using AI to check amongst

the resumes to see which is the best

320

00:17:03,549 --> 00:17:07,358

one to then check it against the job

description that was developed by AI.

321

00:17:07,628 --> 00:17:10,839

And at the end of the day, AI's just

doing your recruiting and it takes

322

00:17:10,839 --> 00:17:14,049

away that human element, which is I

think, at the end of the day, critical.

323

00:17:14,348 --> 00:17:18,189

What a resume shows or what a resume

provides doesn't necessarily mean

324

00:17:18,314 --> 00:17:20,519

that person's gonna be a good fit

within the environment, within the

325

00:17:20,519 --> 00:17:22,919

industry that you work in and that

they're going to be successful.

326

00:17:23,503 --> 00:17:24,914

Thomas Kunjappu: Well, the part

that you're missing in that

327

00:17:24,914 --> 00:17:28,243

equation is that the candidate

pool is also not standing still.

328

00:17:28,364 --> 00:17:33,554

And that resume that these AI tools are

reviewing are half generated customized

329

00:17:33,554 --> 00:17:39,154

in response to the AI generator job

description to give it a great likelihood

330

00:17:39,154 --> 00:17:41,349

of advancing to the next stage.

331

00:17:41,789 --> 00:17:42,539

Fernando Garcia: Yeah.

332

00:17:43,149 --> 00:17:45,639

Thomas Kunjappu: Just AI

tools fighting AI tools.

333

00:17:45,639 --> 00:17:48,819

And there's just like volume

of applications jumping up like

334

00:17:48,819 --> 00:17:50,409

dramatically in all types of roles

335

00:17:50,474 --> 00:17:50,594

Fernando Garcia: Oh,

336

00:17:51,279 --> 00:17:51,879

Thomas Kunjappu: because it's,

337

00:17:51,914 --> 00:17:52,124

Fernando Garcia: That's

338

00:17:52,144 --> 00:17:52,434

Thomas Kunjappu: yeah.

339

00:17:52,454 --> 00:17:54,074

Fernando Garcia: trend that

you're seeing that for any role

340

00:17:54,074 --> 00:17:57,614

where you might have had 20 to 25

applicants, now you're getting 200.

341

00:17:58,694 --> 00:18:01,094

Which makes it even more important

to have the tool to go through and

342

00:18:01,094 --> 00:18:04,504

least the initial process to try to

narrow it down and to focus on it.

343

00:18:04,504 --> 00:18:08,644

But again, you've got to be very careful

in terms of what it is doing and that

344

00:18:08,644 --> 00:18:12,394

you're still having an element of

human judgment and discretion in it.

345

00:18:12,544 --> 00:18:14,404

Because I think that's

where you add the value.

346

00:18:14,639 --> 00:18:17,259

Thomas Kunjappu: I think you're

getting to a key, almost ethical

347

00:18:17,259 --> 00:18:18,849

concept, arguably, right?

348

00:18:18,849 --> 00:18:24,169

Which is sitting at the

intersection here about fairness.

349

00:18:24,169 --> 00:18:27,139

And really about bias avoidance.

350

00:18:28,654 --> 00:18:31,984

It's very obvious as a concept

in the recruiting process.

351

00:18:32,284 --> 00:18:35,974

But you know, also in performance

management all the way to

352

00:18:35,974 --> 00:18:41,934

performance improvement processes

and firing situations as well.

353

00:18:42,734 --> 00:18:48,144

But it's across the board like important

as AI tools get into the workflows here.

354

00:18:48,914 --> 00:18:53,784

So yeah, how do you think

about the ethical side of it.

355

00:18:53,784 --> 00:19:00,894

Besides the efficiency and besides

the legal ramifications, how do you

356

00:19:01,844 --> 00:19:05,399

even think through and we can take any

one of those like use cases and like

357

00:19:05,399 --> 00:19:11,219

think through a little bit together the

ethical consequences of leveraging AI.

358

00:19:11,269 --> 00:19:14,419

Whether there's bias or judgment

that you're now outsourcing

359

00:19:14,419 --> 00:19:16,429

to Artificial Intelligence.

360

00:19:16,479 --> 00:19:18,489

Fernando Garcia: Yeah, and it's

interesting because you get people

361

00:19:18,489 --> 00:19:20,449

who are on both sides of the equation.

362

00:19:20,789 --> 00:19:24,924

There are some people that say that

the more you use AI, the more you

363

00:19:24,924 --> 00:19:27,714

take away the potential bias of

the person who's doing the review.

364

00:19:28,254 --> 00:19:28,674

Thomas Kunjappu: Right.

365

00:19:29,004 --> 00:19:31,404

Fernando Garcia: Other people say, hold

on a second, somebody has programmed

366

00:19:31,404 --> 00:19:33,024

it or it's learned from something.

367

00:19:33,254 --> 00:19:35,334

It just doesn't come

up with these concepts.

368

00:19:35,994 --> 00:19:36,594

The training.

369

00:19:36,774 --> 00:19:40,854

So somebody at some trained it, and

then the biases or the experiences

370

00:19:40,854 --> 00:19:44,904

or whatever else of that training

element, wherever it's coming from,

371

00:19:45,294 --> 00:19:47,544

will creep into the decision making.

372

00:19:47,784 --> 00:19:50,644

So it's almost like one of those dilemmas.

373

00:19:50,644 --> 00:19:59,289

Is it taking away the bias by using

more AI or is it just perpetuating the

374

00:19:59,289 --> 00:20:02,439

bias by using the tool that was created

by somebody who obviously had bias?

375

00:20:02,559 --> 00:20:04,719

What at what point do you minimize that.

376

00:20:04,719 --> 00:20:06,209

And do you get the best result?

377

00:20:07,029 --> 00:20:07,479

Thomas Kunjappu: Right.

378

00:20:08,689 --> 00:20:10,869

Well, that's a great question.

379

00:20:10,869 --> 00:20:15,844

I mean, in the world of self-driving

cars, you have a specific outcome

380

00:20:16,144 --> 00:20:20,194

where you're trying to reduce accidents

and make sure that streets are safer.

381

00:20:20,554 --> 00:20:24,524

And you have an outcome metric

that can be relatively objective.

382

00:20:25,274 --> 00:20:28,074

Maybe the issue in hiring is

that the outcome metric is

383

00:20:28,074 --> 00:20:29,984

like who judges the outcome.

384

00:20:30,624 --> 00:20:31,194

Fernando Garcia: Yeah.

385

00:20:31,334 --> 00:20:35,384

Thomas Kunjappu: Like fatalities is an

objective measure and there's like data

386

00:20:35,384 --> 00:20:36,794

that you can get that you can trust.

387

00:20:37,424 --> 00:20:40,214

In this case, it's like, well, you

can make hires, but is it a good hire?

388

00:20:40,214 --> 00:20:40,994

Was it the right hire?

389

00:20:40,994 --> 00:20:42,014

Do we miss other people?

390

00:20:42,014 --> 00:20:46,124

There's so many other aspects to it

and at least this one facet, arguably.

391

00:20:46,524 --> 00:20:50,564

You know, hiring humans is more of a to

392

00:20:53,684 --> 00:20:56,254

really get to clarity in

393

00:20:58,774 --> 00:21:01,974

what is like the best outcome.

394

00:21:02,214 --> 00:21:04,714

But you mentioned these both sides

of the argument are you convinced

395

00:21:04,714 --> 00:21:05,944

or swayed by either side or...

396

00:21:06,089 --> 00:21:07,109

specifically with recruiting?

397

00:21:07,124 --> 00:21:07,634

Fernando Garcia: sides.

398

00:21:07,904 --> 00:21:11,444

I wanna remain cautious and

knowledgeable that there are risks

399

00:21:11,474 --> 00:21:13,304

and there are limitations both ends.

400

00:21:13,724 --> 00:21:15,764

And you mentioned the autonomous

vehicles, for example.

401

00:21:16,484 --> 00:21:17,594

Even that has a bias.

402

00:21:17,594 --> 00:21:19,904

And there's a great website, I'm

trying to remember the name of it.

403

00:21:20,054 --> 00:21:22,274

It's by MIT and I think

it's called Moral Machines.

404

00:21:22,649 --> 00:21:24,959

And it basically says this: you're

driving your autonomous vehicle.

405

00:21:25,439 --> 00:21:27,359

And who's driving it?

406

00:21:27,419 --> 00:21:29,389

The autonomous vehicle is.

407

00:21:29,419 --> 00:21:34,469

And it comes to a situation where

it must decide, is it going to

408

00:21:34,499 --> 00:21:37,619

get into an accident and kill the

occupant or is it going to kill one

409

00:21:37,619 --> 00:21:38,699

person who's crossing the street?

410

00:21:39,149 --> 00:21:41,339

What if it's two people crossing

the street versus one occupant?

411

00:21:41,669 --> 00:21:45,389

What if it's other people crossing the

street and it's two young children in

412

00:21:45,389 --> 00:21:49,289

the car and it starts creating all these

moral dilemmas in terms of saying at

413

00:21:49,289 --> 00:21:50,879

some point something's got to decide.

414

00:21:51,089 --> 00:21:55,169

And depending on what your value is,

if you're a car company that comes

415

00:21:55,169 --> 00:21:59,519

out and says, as of 2030, I'm going to

have zero occupancy deaths in my car.

416

00:22:00,209 --> 00:22:02,659

Then that means that when it has to

make that moral decision, it's going

417

00:22:02,849 --> 00:22:06,659

to obviously prioritize the safety and

wellbeing of the people in the car versus

418

00:22:06,659 --> 00:22:09,539

the ones who are crossing the street

and getting to an awkward situation.

419

00:22:09,779 --> 00:22:12,749

So I think even when you're

looking at something as autonomous

420

00:22:12,749 --> 00:22:15,629

vehicles, there's still a value

judgment that will come into play.

421

00:22:16,059 --> 00:22:20,009

And I can totally see it, that at some

point it'll say, whoa, I don't know.

422

00:22:20,009 --> 00:22:21,509

I can't make that moral decision for you.

423

00:22:21,599 --> 00:22:22,829

Here's a steering wheel back.

424

00:22:22,859 --> 00:22:25,259

You decide, are you going to put

yourself in a ditch or are you

425

00:22:25,259 --> 00:22:27,509

going to run over the two people

who are crossing the street.

426

00:22:27,869 --> 00:22:29,379

Maybe jaywalking or something else.

427

00:22:30,089 --> 00:22:34,869

Even at that level, even autonomous

vehicles, that technology still

428

00:22:34,869 --> 00:22:38,039

will have a number of bias to it.

429

00:22:38,234 --> 00:22:38,894

Thomas Kunjappu: Fair point.

430

00:22:38,894 --> 00:22:40,964

I mean, you're talking about the

trolley problem, but instead of

431

00:22:40,964 --> 00:22:46,364

like deciding between two different

external parties and a human deciding,

432

00:22:46,364 --> 00:22:49,754

it's an AI deciding between, in this

case, like a customer who bought

433

00:22:49,754 --> 00:22:51,644

their product and like someone else.

434

00:22:51,644 --> 00:22:56,224

But that's a tricky dilemma to get into.

435

00:22:56,224 --> 00:23:01,029

And the outcome or like the escape valve

let's call it is that well, okay, we don't

436

00:23:01,029 --> 00:23:03,399

know, let's let the human needs to decide.

437

00:23:03,399 --> 00:23:07,309

So in the recruiting world, if we come

back to what we're talking about, it's

438

00:23:07,309 --> 00:23:13,249

like we need to let the recruiter or the

manager with all the human flaws and all.

439

00:23:13,789 --> 00:23:16,519

We're gonna attribute it to

this person so that we can

440

00:23:18,859 --> 00:23:21,199

move it out of the hands of a machine.

441

00:23:22,434 --> 00:23:24,144

Fernando Garcia: You can do all

the behavioral testing you want.

442

00:23:24,144 --> 00:23:27,564

And you can do all the

testing and the AI reviews.

443

00:23:27,744 --> 00:23:31,614

But at the end of the day, having

that person meet, whether virtually

444

00:23:31,614 --> 00:23:32,934

or in person with the team.

445

00:23:33,174 --> 00:23:35,224

and seeing that interaction,

seeing how they're going to fit

446

00:23:35,224 --> 00:23:38,364

in, seeing how well they're going

to thrive within that environment.

447

00:23:38,694 --> 00:23:40,454

Nothing will replicate that.

448

00:23:40,694 --> 00:23:45,604

Why people always say, are you

nervous about AI taking away work?

449

00:23:45,934 --> 00:23:48,359

Or I think no.

450

00:23:48,359 --> 00:23:51,034

I think it's going to change our jobs

and it's going to change what we do.

451

00:23:51,034 --> 00:23:54,634

But the value that we truly

add is that value judgment.

452

00:23:54,854 --> 00:23:55,784

That knowledge base.

453

00:23:55,784 --> 00:23:57,849

That experience that you're

bringing into your role.

454

00:23:58,094 --> 00:24:00,884

And I think that human touch,

that human element of it is

455

00:24:00,884 --> 00:24:01,844

really what adds our value.

456

00:24:01,844 --> 00:24:03,554

And I think that's no matter what,

that's never gonna be taken away.

457

00:24:03,794 --> 00:24:07,874

There might be less labor intensive

elements of it and there's gonna be

458

00:24:07,874 --> 00:24:11,054

questions about how do you get people

trained and make sure that when you're

459

00:24:11,054 --> 00:24:14,294

using AI you're not just parking your

brain at the door that you're still being

460

00:24:14,594 --> 00:24:17,174

critical and you're using it properly

and you're learning how to prompt.

461

00:24:17,174 --> 00:24:18,884

But you're also learning

how to apply the knowledge.

462

00:24:19,094 --> 00:24:22,244

But at the end of the day, I think

that is truly where we add value.

463

00:24:22,244 --> 00:24:23,504

Is the human piece.

464

00:24:23,504 --> 00:24:24,499

And I think that's never going to go away.

465

00:24:25,660 --> 00:24:27,790

This has been a fantastic

conversation so far.

466

00:24:28,030 --> 00:24:30,820

If you haven't already done so,

make sure to join our community.

467

00:24:31,060 --> 00:24:33,730

We are building a network of the

most forward-thinking, HR and

468

00:24:33,730 --> 00:24:36,910

people, operational professionals

who are defining the future.

469

00:24:37,390 --> 00:24:40,000

I will personally be sharing

news and ideas around how we

470

00:24:40,000 --> 00:24:42,010

can all thrive in the age of ai.

471

00:24:42,370 --> 00:24:45,355

You can find it at go cleary.com/cleary

472

00:24:45,785 --> 00:24:46,395

community.

473

00:24:47,020 --> 00:24:47,860

Now back to the show.

474

00:24:49,075 --> 00:24:52,435

Thomas Kunjappu: So how do you imagine

that transition happening then?

475

00:24:52,465 --> 00:24:57,495

So that folks and let's focus on

like the HR department as as a whole.

476

00:24:58,515 --> 00:25:01,365

You said maybe we're not gonna be spending

as much time on the administrative

477

00:25:01,365 --> 00:25:03,575

elements of it and the repetitive tasks.

478

00:25:03,575 --> 00:25:07,565

So how do you imagine going from

like today to like, what does

479

00:25:07,565 --> 00:25:08,885

that future state, look like?

480

00:25:10,175 --> 00:25:10,415

Fernando Garcia: Yeah.

481

00:25:10,745 --> 00:25:13,505

I think, if you gain the 200

resumes for that single job,

482

00:25:13,505 --> 00:25:14,495

if you can cut it down to 15.

483

00:25:15,395 --> 00:25:18,455

Using AI and then understanding that

within those 15 you still have to go

484

00:25:18,455 --> 00:25:23,105

in there and do the review, meet them,

do the screening, to see how they're

485

00:25:23,105 --> 00:25:25,865

gonna apply in your, workplace and

how they're going to thrive within it

486

00:25:25,865 --> 00:25:28,145

and what that we're gonna add to it.

487

00:25:28,145 --> 00:25:29,495

I think that's part of it.

488

00:25:29,615 --> 00:25:32,285

So it'll help you get partially

there, but you still have

489

00:25:32,285 --> 00:25:33,305

to drive it home at the end.

490

00:25:33,635 --> 00:25:35,465

And you still have to use

your judgment on that part.

491

00:25:36,790 --> 00:25:43,625

Thomas Kunjappu: So are we okay with

the moral hazard of AI eliminating

492

00:25:43,655 --> 00:25:48,755

90 out of a hundred resumes and

is is that gonna be unbiased?

493

00:25:48,805 --> 00:25:51,895

'Cause what we're saying is like

at least for that one phase of

494

00:25:51,895 --> 00:25:55,675

the application process, then we

are okay with the trolley problem.

495

00:25:55,675 --> 00:25:59,545

Whatever value judgment that we're

maybe helping program into the AI.

496

00:25:59,545 --> 00:26:01,085

But we're gonna.

497

00:26:01,515 --> 00:26:03,220

Use that to make some judgment, right?

498

00:26:03,220 --> 00:26:04,840

And then continue onwards.

499

00:26:05,440 --> 00:26:09,240

Ultimately it's a practicality I feel

like for a lot of recruiting teams.

500

00:26:09,240 --> 00:26:11,580

Because if you have a thousand

resumes like what do you,

501

00:26:12,360 --> 00:26:12,630

right.

502

00:26:13,080 --> 00:26:16,750

So do you kind of escape the ethical

problem because of the necessity?

503

00:26:17,590 --> 00:26:20,110

Fernando Garcia: I think you have

to be aware of the ethical problem

504

00:26:20,110 --> 00:26:23,500

and you have to be aware of the

potential risks associated with it.

505

00:26:23,500 --> 00:26:26,470

And then at that point, I think

it depends on the role, right?

506

00:26:26,470 --> 00:26:27,430

If it's a critical role.

507

00:26:27,435 --> 00:26:30,550

Then maybe you say look, this

is such an important role that I

508

00:26:30,550 --> 00:26:32,590

will go through the 200 resumes.

509

00:26:33,370 --> 00:26:38,905

Some other one that maybe isn't as

critical or one that maybe isn't fit

510

00:26:38,905 --> 00:26:40,615

is not as important or who knows.

511

00:26:40,615 --> 00:26:41,905

Whatever the parameters are.

512

00:26:42,055 --> 00:26:46,135

Maybe that one requires a

little bit more or allows you a

513

00:26:46,135 --> 00:26:47,335

little bit more space in that.

514

00:26:47,335 --> 00:26:49,855

But again, I think there are elements

of you always have to be aware of.

515

00:26:50,535 --> 00:26:51,655

And the biases are there.

516

00:26:51,655 --> 00:26:54,385

And but again, I think

there are risk both sides.

517

00:26:54,415 --> 00:26:57,555

'Cause the bias of a person we're

reviewing the resumes could also

518

00:26:57,555 --> 00:27:01,305

be as important as a bias in the

technology or how we're separating

519

00:27:01,305 --> 00:27:03,585

it or doing the first cut through AI.

520

00:27:03,645 --> 00:27:06,105

You are never gonna completely eliminate.

521

00:27:06,105 --> 00:27:07,305

I think regardless of the two.

522

00:27:08,055 --> 00:27:09,615

Gotta be aware of it and

you gotta be mindful of it.

523

00:27:10,410 --> 00:27:10,530

Thomas Kunjappu: Yeah.

524

00:27:10,530 --> 00:27:15,460

One idea I had as you're saying that is

that the roles which the organization

525

00:27:15,460 --> 00:27:18,910

has hired like hundreds of in the past.

526

00:27:19,800 --> 00:27:24,405

You've had like over many years and it's

like the same kind of background that's

527

00:27:24,405 --> 00:27:28,965

successful and you've established that

qualitatively and quantitatively, right?

528

00:27:28,965 --> 00:27:34,685

So these is all input into like whatever

you're using for screening as an AI tool.

529

00:27:34,685 --> 00:27:40,230

So the more I guess validated

specific input data you have on

530

00:27:40,230 --> 00:27:46,230

a specific decision, the more

maybe you've controlled bias.

531

00:27:46,290 --> 00:27:50,340

And you're maybe more comfortable

ethically, you know, moving efficiency

532

00:27:50,340 --> 00:27:54,450

around versus hiring for the first time

for a role ever in in a new context.

533

00:27:54,500 --> 00:27:55,820

You just literally don't know.

534

00:27:55,820 --> 00:27:56,000

right?

535

00:27:56,050 --> 00:27:59,980

What is the background or like what

are the set of characters that a

536

00:27:59,980 --> 00:28:04,720

resume can say that actually represent

success if it's never been done.

537

00:28:04,750 --> 00:28:06,220

Especially at this organization before.

538

00:28:06,630 --> 00:28:09,900

Fernando Garcia: But again, there's no

single answer and no single right answer.

539

00:28:09,900 --> 00:28:13,260

Because again, you always hear

that the most dangerous words are

540

00:28:13,260 --> 00:28:14,430

we've always done it that way.

541

00:28:14,830 --> 00:28:19,530

So just because always hired this way

and obtained this level of performance,

542

00:28:19,590 --> 00:28:24,120

you don't know if taking a different

approach or hiring people maybe who

543

00:28:24,120 --> 00:28:27,030

are not normally the people you would

hire for that particular role is not

544

00:28:27,030 --> 00:28:29,280

gonna elevate level of performance.

545

00:28:30,000 --> 00:28:32,160

So you also have to every once in

a while just break up from that

546

00:28:35,030 --> 00:28:39,260

trend and be able to think outside

the box and maybe go outside of

547

00:28:39,260 --> 00:28:40,790

the areas you normally hire from.

548

00:28:40,850 --> 00:28:44,090

Maybe go outside of the colleges you

normally hire from or whatever, because

549

00:28:44,150 --> 00:28:47,690

that is how you identify potential value

that might have otherwise been lost.

550

00:28:47,930 --> 00:28:52,520

And I think that's where AI can

maybe fail because it does look at

551

00:28:52,520 --> 00:28:56,510

trending, it does look at what have

you done and it establishes a basement.

552

00:28:56,510 --> 00:28:58,780

But sometimes you wanna go

above and beyond that basement.

553

00:28:59,460 --> 00:29:02,945

I think that's when you don't

even be a lawyer or anything else.

554

00:29:03,695 --> 00:29:07,925

The creativity of a novel argument

or some additional way of looking at

555

00:29:07,925 --> 00:29:12,665

something which has not been done before

is when you're truly adding value and

556

00:29:12,665 --> 00:29:15,665

you're elevating that performance from

what is normally you're accustomed to.

557

00:29:16,685 --> 00:29:20,345

And the question is, will AI ever

get to the point where it can do that

558

00:29:20,345 --> 00:29:24,785

additional critical thinking or jump to

say, this is what you've always done,

559

00:29:24,785 --> 00:29:26,885

this is what you've always achieved,

but if you do something different,

560

00:29:26,885 --> 00:29:28,235

you might be able to get here now.

561

00:29:29,645 --> 00:29:31,965

Because we tend to learn from the past.

562

00:29:32,195 --> 00:29:33,915

We tend to replicate the results.

563

00:29:34,305 --> 00:29:36,715

But the question is, it starts

getting to the point where it's now

564

00:29:36,715 --> 00:29:43,015

cognitive and starting to make it's

learn and go beyond just the confines

565

00:29:43,015 --> 00:29:44,875

of the past or previous decisions.

566

00:29:44,935 --> 00:29:46,795

That's when it starts getting interesting.

567

00:29:47,725 --> 00:29:48,085

Thomas Kunjappu: Right.

568

00:29:48,085 --> 00:29:52,375

Like incorporating new data about

like what the business needs

569

00:29:52,375 --> 00:29:56,485

now, for example, or like what's

shifted now and yeah, absolutely.

570

00:29:56,545 --> 00:29:58,645

Fernando Garcia: much risk is your

company willing to take in this role.

571

00:29:58,645 --> 00:30:02,365

To say look, maybe we go outside of

the normal beaten path and maybe look

572

00:30:02,365 --> 00:30:05,250

for that purple unicorn or it was a

purple squirrel that they call it.

573

00:30:05,985 --> 00:30:08,955

That additional person and sometimes

you gotta go outside of that to find it.

574

00:30:10,230 --> 00:30:12,060

Thomas Kunjappu: Let's get

a little bit practical then.

575

00:30:12,480 --> 00:30:16,320

Can you tell me about the key workflows,

use cases that you've personally seen.

576

00:30:16,800 --> 00:30:21,720

Like some success in and maybe others

that you're excited about where AI can

577

00:30:21,720 --> 00:30:24,670

start changing the way work has done.

578

00:30:25,165 --> 00:30:28,255

Fernando Garcia: Yeah, some of

the interesting functions that

579

00:30:28,255 --> 00:30:32,125

I've seen or applications are

things like employee surveys.

580

00:30:32,180 --> 00:30:37,100

Finding trends and identifying things

that maybe you need to focus on developing

581

00:30:37,220 --> 00:30:39,320

training programs for individuals.

582

00:30:40,380 --> 00:30:42,570

Go in there and say, look,

this is where I'm having a gap.

583

00:30:42,570 --> 00:30:43,500

How do you recommend that?

584

00:30:43,980 --> 00:30:46,260

training, it can help you

develop training programs.

585

00:30:46,710 --> 00:30:48,990

We talked about the recruiting

obviously that's a critical

586

00:30:48,990 --> 00:30:50,550

one and job descriptions.

587

00:30:52,050 --> 00:30:55,350

Helping you take that initial

draft and the things that you do.

588

00:30:55,830 --> 00:30:59,010

The letters, you have a transformation

coming up and all of a sudden

589

00:30:59,040 --> 00:31:00,330

you're last minute rushing it.

590

00:31:00,330 --> 00:31:03,570

It can help you get the first draft

going and then you can perfect it.

591

00:31:04,610 --> 00:31:08,930

So I think, if people are starting to get

curious, people are starting to use it,

592

00:31:09,920 --> 00:31:13,280

companies are developing pilot projects

and getting individuals who are trained

593

00:31:13,280 --> 00:31:16,905

to start applying it and then once they

see an element of success and comfort and

594

00:31:16,905 --> 00:31:19,515

they start adopting it to greater numbers.

595

00:31:20,205 --> 00:31:21,465

Everybody's handling it differently.

596

00:31:21,675 --> 00:31:23,265

But the beauty is people

are looking at it.

597

00:31:23,325 --> 00:31:23,745

right?

598

00:31:24,075 --> 00:31:29,535

And again, challenging the status quo

in many ways and just trying to be

599

00:31:29,585 --> 00:31:31,475

more value and doing more with less.

600

00:31:31,505 --> 00:31:34,950

Which is I think the reality of all

of our jobs and of our functions.

601

00:31:35,810 --> 00:31:36,800

Thomas Kunjappu: So

tell me more about that.

602

00:31:36,830 --> 00:31:40,790

What are you seeing within the world

of like HR specifically around this

603

00:31:40,790 --> 00:31:42,470

concept of doing more with less?

604

00:31:42,470 --> 00:31:44,390

Is that seeming like universal.

605

00:31:44,390 --> 00:31:45,170

Like with your peers?

606

00:31:45,170 --> 00:31:46,610

What's the solution here?

607

00:31:46,700 --> 00:31:50,750

Like how do we get to the point of

like being productive and getting

608

00:31:50,750 --> 00:31:52,600

resource enough to be able to do that?

609

00:31:53,205 --> 00:31:54,705

Fernando Garcia: Yeah, I think

there are those who are the

610

00:31:54,705 --> 00:31:56,415

trendsetters who are using it more.

611

00:31:56,445 --> 00:31:59,175

And then you go to a conference

or you go to an event and you

612

00:31:59,175 --> 00:32:00,105

start talking to other people.

613

00:32:00,155 --> 00:32:01,355

This is what I've done with it.

614

00:32:01,355 --> 00:32:02,535

This is how I used it.

615

00:32:02,895 --> 00:32:04,805

Somebody else goes, Ooh, that's

actually a pretty good idea.

616

00:32:05,070 --> 00:32:05,220

Yeah.

617

00:32:05,225 --> 00:32:05,705

Try that.

618

00:32:05,880 --> 00:32:06,120

Right.

619

00:32:06,645 --> 00:32:11,175

The critical element of this is the people

who are gonna be the trendsetters and

620

00:32:11,175 --> 00:32:15,495

the people who are gonna be able to adopt

the best practices or learn from what

621

00:32:15,495 --> 00:32:19,425

others are doing and start implementing

in their own play in their own way.

622

00:32:19,695 --> 00:32:24,525

But the key is just being

curious using experimentation.

623

00:32:24,585 --> 00:32:27,705

Bringing things in and doing

things differently the AI and

624

00:32:27,795 --> 00:32:29,055

where it's going and technology.

625

00:32:29,055 --> 00:32:32,165

But again, AI is one element but

it's not the only element, right?

626

00:32:32,565 --> 00:32:35,355

Those information that you're

taking from HRS systems, from

627

00:32:35,535 --> 00:32:36,915

contract management systems.

628

00:32:37,545 --> 00:32:42,015

Data is power and the information

in the technologies is critical

629

00:32:42,015 --> 00:32:43,455

in terms of us and our functions.

630

00:32:44,280 --> 00:32:47,040

Thomas Kunjappu: So if that's where

things are headed, like what do you

631

00:32:47,040 --> 00:32:51,520

think are the kinds of people or

skillsets that are coming into the fore?

632

00:32:51,590 --> 00:32:54,500

Who are gonna be more effective

going into the future?

633

00:32:54,500 --> 00:32:57,290

A lot of what we try to talk

about is future proofing.

634

00:32:57,350 --> 00:32:57,680

right?

635

00:32:58,235 --> 00:32:58,745

Fernando Garcia: Yeah.

636

00:32:58,985 --> 00:33:00,245

Thomas Kunjappu: In

like whatever function.

637

00:33:00,295 --> 00:33:02,395

And actually maybe let me

ask a specific question.

638

00:33:02,625 --> 00:33:05,805

Is it getting to the point where

job descriptions are starting to

639

00:33:05,805 --> 00:33:10,725

more heavily feature, you know,

either the ability to learn...

640

00:33:11,165 --> 00:33:15,665

Like rethink how you're doing

workflows and or all the way up to

641

00:33:15,905 --> 00:33:22,265

demonstrated experience in leveraging

AI to bring out efficiencies.

642

00:33:22,445 --> 00:33:22,655

Like...

643

00:33:22,775 --> 00:33:23,015

Fernando Garcia: Yeah, I

644

00:33:23,075 --> 00:33:23,285

Thomas Kunjappu: ...where,

645

00:33:23,285 --> 00:33:23,295

where.

646

00:33:23,465 --> 00:33:23,525

Fernando Garcia: ...a

647

00:33:23,525 --> 00:33:24,515

lot of that yet.

648

00:33:24,935 --> 00:33:28,175

I think I've seen that in terms

of jobs that are heavy with AI or

649

00:33:28,235 --> 00:33:31,265

that are involved in the FinTech

or some of those things like that.

650

00:33:31,325 --> 00:33:33,785

Where that's critical component right now.

651

00:33:34,085 --> 00:33:39,515

But I think at some point I wouldn't

be surprised if we start seeing

652

00:33:39,515 --> 00:33:42,935

things like experience in AI.

653

00:33:43,970 --> 00:33:44,150

Yeah.

654

00:33:45,965 --> 00:33:49,655

Or even then making the assumption that

maybe people are not coming in with that.

655

00:33:49,655 --> 00:33:51,775

But really focusing on

training that skill set.

656

00:33:52,165 --> 00:33:55,445

Because I think companies can do a

really good job in terms of helping

657

00:33:55,445 --> 00:33:58,925

people develop and get comfortable

and see how that can apply.

658

00:33:59,255 --> 00:34:01,865

Whether it's through mentorship

with people who are in the industry

659

00:34:01,865 --> 00:34:04,505

or in your department when you

come in and they can help you.

660

00:34:04,505 --> 00:34:06,385

Look, this is how I'm using.

661

00:34:06,815 --> 00:34:07,625

How are you?

662

00:34:07,625 --> 00:34:13,245

Things like lunchtime training sessions

and sharing your best practices and tips.

663

00:34:13,405 --> 00:34:15,205

Prompting, things like that.

664

00:34:15,205 --> 00:34:16,745

I think that they're critical.

665

00:34:16,864 --> 00:34:19,175

So it's a question of do

you buy it or you make it.

666

00:34:19,295 --> 00:34:24,005

And some I think it's gonna

become more and more that

667

00:34:24,395 --> 00:34:25,505

they put that as a preference.

668

00:34:25,505 --> 00:34:28,715

But I think at the same time, people

are saying look, let's train up.

669

00:34:29,074 --> 00:34:31,594

Let's get this as a skill

set that we're developing.

670

00:34:31,969 --> 00:34:34,940

No different than leadership training

or any other country that we do on a

671

00:34:34,940 --> 00:34:42,650

day-to-day that we can do AI efficiency

training or adaptive technology training.

672

00:34:43,639 --> 00:34:46,330

Again, it's gonna be

critical advantage one day.

673

00:34:47,469 --> 00:34:50,620

Thomas Kunjappu: So if that's

true, how do you square that?

674

00:34:50,620 --> 00:34:54,460

And I would imagine we'd agree

that, the enablement of that for

675

00:34:54,460 --> 00:34:57,820

the organization is primarily gonna

be done through the HR department.

676

00:34:57,820 --> 00:35:01,760

There's some like, individual on every

functional level, people comparing notes,

677

00:35:01,760 --> 00:35:06,350

but you're trying to enable the entire org

strategically up level it for some ROI.

678

00:35:07,520 --> 00:35:12,215

How do you square that with less,

fewer resources to make it so.

679

00:35:13,080 --> 00:35:16,410

Fernando Garcia: It's one of those where

you invest early to get the benefit after.

680

00:35:17,190 --> 00:35:18,840

But again, it doesn't

have to be expensive.

681

00:35:18,960 --> 00:35:22,450

You can get somebody who's very

knowledgeable in AI or very

682

00:35:22,450 --> 00:35:25,660

comfortable with it that can

just do a lunch hour session.

683

00:35:26,410 --> 00:35:29,140

Hey, here's some peaks on the

side and come and sit down.

684

00:35:29,350 --> 00:35:31,990

Let's, go through some

examples of how I'm using this.

685

00:35:33,250 --> 00:35:33,460

Doesn't...

686

00:35:33,540 --> 00:35:34,820

Thomas Kunjappu: Have you

seen the food inflation?

687

00:35:39,175 --> 00:35:39,685

Fernando Garcia: Granted.

688

00:35:40,595 --> 00:35:42,545

There's formal training, but

then there's training that can

689

00:35:42,545 --> 00:35:44,125

happen through peer-to-peer.

690

00:35:44,255 --> 00:35:46,295

There's the train-the-trainer concepts.

691

00:35:46,445 --> 00:35:51,185

And just associations,

industries, conferences, sharing

692

00:35:51,185 --> 00:35:52,835

best practices, sharing tips.

693

00:35:53,915 --> 00:35:55,745

I tend to find that's really, important.

694

00:35:56,020 --> 00:36:00,635

When I, first became how can I

say this comfortable with the

695

00:36:00,635 --> 00:36:02,315

AI, that's how it happened.

696

00:36:02,315 --> 00:36:04,265

I went to a conference

and we had a session on.

697

00:36:05,150 --> 00:36:07,400

Is what I'm using it for,

these are my tips to you.

698

00:36:07,400 --> 00:36:10,610

And now all of a sudden it's

wow, I can do so much with that.

699

00:36:10,850 --> 00:36:15,140

So it's really having those people who

are early champions, early adopters,

700

00:36:15,290 --> 00:36:18,170

and then having them share their

knowledge and experiences with others

701

00:36:18,230 --> 00:36:21,560

within their industry, within the

company and just amongst their peers.

702

00:36:22,690 --> 00:36:26,270

Thomas Kunjappu: As people are sharing

and like learning and upskilling.

703

00:36:26,480 --> 00:36:29,360

Do you imagine new roles and

titles might even appear?

704

00:36:29,690 --> 00:36:31,430

Especially within like HR and legal?

705

00:36:31,430 --> 00:36:31,790

I don't know.

706

00:36:31,790 --> 00:36:34,280

Within the compliance or

employee relations world or

707

00:36:34,750 --> 00:36:35,350

Fernando Garcia: Yeah,

708

00:36:35,510 --> 00:36:39,080

Thomas Kunjappu: within recruiting

or you know, just HR support...

709

00:36:39,580 --> 00:36:44,710

Fernando Garcia: HRIS-type individuals

who are in charge of data and putting

710

00:36:44,710 --> 00:36:47,140

in and analyzing totally see...

711

00:36:47,360 --> 00:36:48,680

Thomas Kunjappu: like

system configuration.

712

00:36:49,570 --> 00:36:50,980

Fernando Garcia:

Configuration and development.

713

00:36:50,980 --> 00:36:56,410

And I took a stat here, it was from

the Conference Board of Canada.

714

00:36:56,410 --> 00:37:00,940

Said the 4 in 10 HR teams are using

some sort of AI for talent management.

715

00:37:01,870 --> 00:37:05,410

So just looking at that number

again, 4 out of 10, it's not

716

00:37:05,410 --> 00:37:08,560

a lot, but it's a number.

717

00:37:08,560 --> 00:37:12,940

4 is getting to that critical mass

point where, you know, as long as

718

00:37:12,940 --> 00:37:15,760

they can start sharing their past

practice and their learnings with

719

00:37:15,760 --> 00:37:19,240

the other six, you'll slowly start

seeing that shift the other way.

720

00:37:19,580 --> 00:37:19,870

Thomas Kunjappu: Right.

721

00:37:19,960 --> 00:37:21,820

Fernando Garcia: I wouldn't be

surprised if 10 years from now we're

722

00:37:21,820 --> 00:37:25,330

looking at the number where eight

in 10 or nine and 10 are using it.

723

00:37:25,330 --> 00:37:26,490

Because I think it's a tool.

724

00:37:26,880 --> 00:37:27,960

It's no different than

725

00:37:30,090 --> 00:37:30,960

computers came up.

726

00:37:30,990 --> 00:37:34,200

And I remember working at a law

firm and for the most part, most

727

00:37:34,200 --> 00:37:36,060

lawyers were using computers.

728

00:37:36,455 --> 00:37:40,440

But there were still a few who

decided to have dictaphone and they

729

00:37:40,440 --> 00:37:42,950

were still dictating their notes

730

00:37:43,040 --> 00:37:43,880

Oh, interesting.

731

00:37:44,030 --> 00:37:44,900

type it up for them.

732

00:37:45,200 --> 00:37:46,970

So it took a while to get there.

733

00:37:46,970 --> 00:37:49,700

I don't think you'll see that anywhere

else in the firms anymore, anywhere.

734

00:37:50,030 --> 00:37:54,170

But there's that transition

period that has to happen.

735

00:37:55,250 --> 00:37:56,990

There are people who are gonna

be extremely comfortable with

736

00:37:56,990 --> 00:37:58,880

saying, look, hey, I'm curious.

737

00:37:58,940 --> 00:37:59,600

I wanna learn.

738

00:37:59,600 --> 00:38:01,340

I wanna develop myself and I wanna do it.

739

00:38:01,640 --> 00:38:03,140

And there's others who are

saying, hold on a second.

740

00:38:03,620 --> 00:38:05,120

I'm fairly confident in what I'm doing.

741

00:38:05,120 --> 00:38:06,890

I don't need that tool.

742

00:38:06,950 --> 00:38:08,960

So they're gonna be more

resistant to change.

743

00:38:09,020 --> 00:38:10,610

But like anything, like any change effort.

744

00:38:10,610 --> 00:38:10,820

right?

745

00:38:10,820 --> 00:38:13,580

There's gonna be those who

take the change head on.

746

00:38:13,580 --> 00:38:15,590

And those who are gonna be

more resistant to the change.

747

00:38:15,620 --> 00:38:18,950

You just have to start persuading them

of why it's important, how they can

748

00:38:19,370 --> 00:38:21,740

improve their performance by using it.

749

00:38:21,740 --> 00:38:26,690

And how maybe they're more worried

about the repercussions of the

750

00:38:26,690 --> 00:38:30,055

risks, how you're mitigating those

risks and training them to do that.

751

00:38:31,010 --> 00:38:32,810

Thomas Kunjappu: It's funny, maybe

the Dictaphone people were actually

752

00:38:32,810 --> 00:38:36,080

ahead of their time because if

you look at what's happening now.

753

00:38:36,825 --> 00:38:37,175

Fernando Garcia: Right.

754

00:38:39,090 --> 00:38:40,560

Thomas Kunjappu: No one's typing up notes.

755

00:38:40,560 --> 00:38:45,610

So if like future focus, you're actually

having the, like an AI transcription,

756

00:38:45,740 --> 00:38:48,720

happening off of the call to then

summarize and kind of go further.

757

00:38:48,720 --> 00:38:50,100

Which then you might like review.

758

00:38:50,790 --> 00:38:52,290

Which is closer in some ways, right?

759

00:38:52,290 --> 00:38:56,190

To the voice dictation, which

then is manually written up.

760

00:38:56,850 --> 00:39:00,400

If we kind of imagine that 4

to 10 gets to be like more like

761

00:39:01,060 --> 00:39:06,280

10 out 10 like eventually over time,

what do you think and to your point,

762

00:39:06,280 --> 00:39:07,930

yes, like AI it's a tool, right?

763

00:39:07,930 --> 00:39:12,010

It's a tool set that you can leverage

humans can and will leverage.

764

00:39:12,190 --> 00:39:14,590

And it's kind of the

next phase of technology.

765

00:39:14,890 --> 00:39:17,200

If you just focus on the

humans for a moment, right?

766

00:39:17,250 --> 00:39:21,880

What is the advice that you have for

someone who's young and just coming outta

767

00:39:21,880 --> 00:39:25,150

school, maybe looking to go into HR.

768

00:39:25,820 --> 00:39:27,380

Maybe even considering law school, right.

769

00:39:27,380 --> 00:39:31,250

And is just trying to think about

like what they should be, like what

770

00:39:31,250 --> 00:39:34,630

skill sets they should really be like

trying to focus on to make sure that

771

00:39:34,630 --> 00:39:37,060

they're employable in the future.

772

00:39:37,900 --> 00:39:38,755

What advice would you have for them?

773

00:39:39,220 --> 00:39:39,490

Fernando Garcia: Yeah.

774

00:39:39,540 --> 00:39:43,660

I always say try to focus on

getting the broader or as broad

775

00:39:43,660 --> 00:39:44,890

of a skill set as possible.

776

00:39:44,890 --> 00:39:48,430

And the curiosity and the use

of technology is one of'em.

777

00:39:49,300 --> 00:39:51,550

Don't lose track of

the ability to network.

778

00:39:51,670 --> 00:39:54,430

The ability to build human

connections and interactions.

779

00:39:54,490 --> 00:39:56,470

To be able to work with people.

780

00:39:57,220 --> 00:40:01,480

The emotional intelligence is critical

in terms of everything that we do in

781

00:40:01,740 --> 00:40:03,490

HR, legal or anything else that we do.

782

00:40:03,910 --> 00:40:05,620

And it goes back to that T-shaped concept.

783

00:40:05,620 --> 00:40:06,010

right?

784

00:40:06,370 --> 00:40:11,350

Where your legal or your HR knowledge

could be this, but then there's

785

00:40:11,500 --> 00:40:14,530

the whole horizontal piece where

there's all these other skill sets

786

00:40:14,530 --> 00:40:16,090

that's so critical for you to have.

787

00:40:16,180 --> 00:40:22,140

Whether it's project management skill

sets the relationship building or with

788

00:40:22,140 --> 00:40:24,180

diverse and international workforces.

789

00:40:24,180 --> 00:40:27,510

That there's all these things that

add value to you as a person that

790

00:40:27,510 --> 00:40:29,460

are not gonna be just technology.

791

00:40:30,150 --> 00:40:31,600

So you've got to be careful not to be the.

792

00:40:32,475 --> 00:40:35,295

the person focuses on a

technology's helping you do A.

793

00:40:35,295 --> 00:40:36,435

But what else are you doing?

794

00:40:36,435 --> 00:40:39,135

How else are you doing in your

skill set and how else are you

795

00:40:39,135 --> 00:40:40,605

contributing and adding value to it?

796

00:40:41,025 --> 00:40:44,075

The technology's gonna help us

get to a certain point, but the

797

00:40:44,075 --> 00:40:47,135

human piece is always what's gonna

take you above and beyond that and

798

00:40:47,135 --> 00:40:49,355

make you successful and adaptable.

799

00:40:49,355 --> 00:40:53,435

And adding value to whatever organization

you're in or whatever function

800

00:40:53,435 --> 00:40:54,795

or task or role you're taking on.

801

00:40:56,525 --> 00:41:00,115

Thomas Kunjappu: That's great advice

and seems timeless in some ways.

802

00:41:00,805 --> 00:41:04,945

But then while I have you,

I have to ask, Fernando.

803

00:41:04,995 --> 00:41:08,570

Before I let you go, as you're looking

in the horizon of the next couple of

804

00:41:08,570 --> 00:41:15,720

years, is there anything in general that

you're working on or a concept that you

805

00:41:15,720 --> 00:41:19,150

feel like is going to come to fruition

that you're particularly passionate

806

00:41:19,150 --> 00:41:20,360

about that you'd be willing to share?

807

00:41:22,015 --> 00:41:22,675

Fernando Garcia: That's interesting.

808

00:41:25,015 --> 00:41:28,875

Personally, I've been working a lot in

terms of the legal industry, in terms of

809

00:41:28,875 --> 00:41:33,495

making sure that we think of ourselves

as especially as in-house counsel, more

810

00:41:33,495 --> 00:41:38,655

as business executives with a legal

background or in an HR background.

811

00:41:38,655 --> 00:41:41,415

And really thinking about how we're

adding value to our organization.

812

00:41:41,415 --> 00:41:45,165

How we're helping people grow within

the industry or within our companies.

813

00:41:45,225 --> 00:41:48,260

How we're identifying talent, how

we're helping those talents stay.

814

00:41:49,035 --> 00:41:52,535

I think there isn't one particular

area, but I think it's more about how

815

00:41:52,685 --> 00:41:55,415

holistically we're all becoming a little

bit more knowledgeable, a little bit

816

00:41:55,415 --> 00:41:59,035

more skilled, and how we're incorporating

all those skill sets to do better work.

817

00:41:59,035 --> 00:42:03,805

And to make sure that we truly say

that people value, people matter.

818

00:42:04,135 --> 00:42:07,625

And that's one of our strengths

and we're adopting all these tools.

819

00:42:07,895 --> 00:42:12,565

Because the one risk of technology

is if you stick to it too

820

00:42:12,565 --> 00:42:14,815

much, it might dehumanize you.

821

00:42:14,990 --> 00:42:18,290

I think at the end of the day,

that human element of it will

822

00:42:18,290 --> 00:42:21,530

never go outta style, will never be

something that doesn't add value.

823

00:42:21,740 --> 00:42:23,555

And that's not something we

should ever stop developing.

824

00:42:24,710 --> 00:42:26,450

Thomas Kunjappu: I love that thought.

825

00:42:26,450 --> 00:42:30,470

And it's something I can personally relate

to and having wonderful conversations

826

00:42:30,470 --> 00:42:35,040

like this on this podcast is I feel

it humanizes me more personally.

827

00:42:35,040 --> 00:42:37,560

I'm in terms of having conversations.

828

00:42:37,560 --> 00:42:42,780

Because you know, working in the

software and AI world constantly,

829

00:42:42,780 --> 00:42:44,860

you're just kind of honed in there.

830

00:42:44,860 --> 00:42:48,630

But this helps you keep that

human connection no matter

831

00:42:48,630 --> 00:42:50,635

what it is that you're doing.

832

00:42:51,235 --> 00:42:53,935

So thank you for this

conversation, Fernando.

833

00:42:53,935 --> 00:42:58,395

Because we covered a lot of ground

from some I didn't expect to go

834

00:42:58,395 --> 00:43:00,175

into the trolley problem where

835

00:43:00,275 --> 00:43:04,255

the remix is that AI is making

the decision and you are one

836

00:43:04,255 --> 00:43:06,415

of the lives not these others.

837

00:43:07,495 --> 00:43:13,115

And relating that to talent

acquisition funnel problems.

838

00:43:13,955 --> 00:43:18,265

But it's really interesting how

you're kind of looking at things

839

00:43:18,265 --> 00:43:24,210

from a legal, customer contract and

compliance perspective while also

840

00:43:24,420 --> 00:43:27,510

very much keeping the HR hat on.

841

00:43:27,960 --> 00:43:31,940

And I would think that maybe you

could put finance in there as well.

842

00:43:31,940 --> 00:43:35,810

But you know, these are the functions

that typically are the most risk-averse I

843

00:43:35,810 --> 00:43:38,150

think it's fair to say in an organization.

844

00:43:38,150 --> 00:43:40,900

And maybe by design, sales and

marketing and the CEO is supposed

845

00:43:40,900 --> 00:43:42,660

to push for new boundaries.

846

00:43:42,660 --> 00:43:46,250

And you're trying to say hey, you

don't want to do that because we'll

847

00:43:46,250 --> 00:43:47,970

get sued and we've got to hold back.

848

00:43:48,390 --> 00:43:53,140

But it's really refreshing to

hear about your nuanced position.

849

00:43:53,140 --> 00:43:58,480

Because even if that's true, you're

very much like an early adopter of

850

00:43:58,960 --> 00:44:04,090

AI tools, which have ethical and

efficiency issues and trade-offs.

851

00:44:04,810 --> 00:44:06,460

But simply cannot be ignored.

852

00:44:06,910 --> 00:44:10,840

And you know, I like the

nuance in our conversation.

853

00:44:10,840 --> 00:44:17,180

Where it's every company and every

role might have a different nuance for

854

00:44:17,180 --> 00:44:19,250

how you might use AI versus people.

855

00:44:19,970 --> 00:44:21,410

And that probably goes more broadly.

856

00:44:21,410 --> 00:44:23,570

And it's really important

to weigh those things.

857

00:44:23,570 --> 00:44:26,810

Even though you have some fixed

guardrails to get started with.

858

00:44:26,810 --> 00:44:28,820

And thanks for going through

some of those, because

859

00:44:29,015 --> 00:44:29,315

Fernando Garcia: Yeah.

860

00:44:29,330 --> 00:44:31,970

Thomas Kunjappu: A lot of companies

are just getting in the early stages

861

00:44:31,970 --> 00:44:35,960

of that journey of figuring out how

to go wall to wall or enable people.

862

00:44:36,120 --> 00:44:39,740

So I think that would be helpful for

folks who are out there listening,

863

00:44:40,280 --> 00:44:43,160

who are looking to future proof

their own functions in HR but

864

00:44:43,160 --> 00:44:45,440

also their organizations overall.

865

00:44:45,710 --> 00:44:48,500

So thank you once again,

Fernando, for the conversation.

866

00:44:49,295 --> 00:44:49,955

Fernando Garcia: It'll be interesting.

867

00:44:49,985 --> 00:44:53,975

Five years from now, we'll look back

on this and say wow, we either got it

868

00:44:53,975 --> 00:44:56,105

right or we completely missed the boat.

869

00:44:56,105 --> 00:44:59,735

Or technology and AI started

developing completely different

870

00:44:59,735 --> 00:45:03,425

ways and the applications were

that we never thought about.

871

00:45:03,425 --> 00:45:06,935

But it's an interesting time and

if I can say anything it's just

872

00:45:06,935 --> 00:45:11,300

stay curious and do it safely.

873

00:45:11,300 --> 00:45:13,670

And do it within the guardrails.

874

00:45:14,150 --> 00:45:16,220

But do it because it's an exciting time.

875

00:45:16,320 --> 00:45:18,540

We have an incredible tool that we're

seeing developed in front of our

876

00:45:18,540 --> 00:45:20,160

eyes that's growing exponentially.

877

00:45:20,430 --> 00:45:23,040

And you don't wanna miss out on that.

878

00:45:24,005 --> 00:45:24,815

Thomas Kunjappu: Absolutely.

879

00:45:24,815 --> 00:45:26,555

So let's leave it there.

880

00:45:26,555 --> 00:45:29,435

So thank you and everyone out there.

881

00:45:29,440 --> 00:45:30,400

We'll see you on the next one.

882

00:45:30,460 --> 00:45:30,880

Bye now.

883

00:45:31,840 --> 00:45:34,900

Thanks for joining us on this

episode of Future Proof HR.

884

00:45:35,140 --> 00:45:37,750

If you like the discussion, make

sure you leave us a five star

885

00:45:37,750 --> 00:45:40,630

review on the platform you're

listening to or watching us on.

886

00:45:41,050 --> 00:45:43,990

Or share this with a friend or colleague

who may find value in the message.

887

00:45:44,290 --> 00:45:49,240

See you next time as we keep our pulse on

how we can all thrive in the age on AI.