WEBVTT
00:00:00.160 --> 00:00:04.959
Welcome back to Intelligence Real and Imagined from the Work AI Institute at Glean.
00:00:05.120 --> 00:00:11.679
This is the show where we sort through what's real, what's hype, and what actually works with AI at work.
00:00:12.080 --> 00:00:16.399
I'm your host, Rebecca Hines, and I lead our Work AI Institute here at Glean.
00:00:16.800 --> 00:00:22.960
Today I'm joined by Phil Kirshner from PK Consulting and Mark Christensen from Glean.
00:00:23.199 --> 00:00:27.359
This episode is inspired by our AI Transformation 100 report.
00:00:27.519 --> 00:00:32.159
We're talking about how do you rewire your organization for AI?
00:00:32.479 --> 00:00:37.359
Because real adoption doesn't come from access to the technology alone.
00:00:37.520 --> 00:00:42.000
It comes from redesigning workflows, habits, and decision making.
00:00:42.079 --> 00:00:45.600
So AI becomes part of how work actually gets done.
00:00:45.759 --> 00:00:47.119
Let's dive in.
00:00:50.159 --> 00:00:53.920
I'm joined by two outstanding guests and also friends.
00:00:54.240 --> 00:00:57.359
Phil, why don't you kick us off and introduce yourself?
00:00:57.840 --> 00:00:58.560
Thank you, Rebecca.
00:00:58.640 --> 00:00:59.920
And thank you, everyone, for joining.
00:01:00.079 --> 00:01:01.280
My name is Phil Kirschner.
00:01:01.359 --> 00:01:03.679
I'm a New York City-based modern work consultant.
00:01:03.840 --> 00:01:10.959
Um, have fallen into the world of employee experience by way of workplace strategy in the built environment, which we'll get into uh today.
00:01:11.120 --> 00:01:19.599
Um, I have my own consultancy, work with clients across industries, but usually at the intersection of groups like HR IT and real estate.
00:01:19.840 --> 00:01:28.719
And I write a newsletter called The Work Line, which is designed to help people kind of build their cross-functional courage and lead across those lines for uh better day at work.
00:01:29.439 --> 00:01:30.000
Wonderful.
00:01:30.159 --> 00:01:37.120
And for anyone who hasn't read The Work Line, it's one of my go-to resources for all things work and evolution of work.
00:01:37.200 --> 00:01:39.040
So check it out if you haven't already.
00:01:39.200 --> 00:01:41.040
Uh Mark, please introduce yourself.
00:01:41.519 --> 00:01:42.959
Yeah, Mark Christensen.
00:01:43.040 --> 00:01:48.799
I am uh here in sunny central Florida and uh I've been in tech for 30 years.
00:01:49.200 --> 00:01:57.439
For got into digital workplace and started to figure out uh like how the evolution of things happened post-COVID.
00:01:57.599 --> 00:02:12.639
So I wrote a book called The Productivity Paradox, and uh have been in kind of where tech meets the human condition and uh most recently uh was doing digital workplace and and putting Glean into things, and now I'm working for Glean.
00:02:13.120 --> 00:02:13.680
Wonderful.
00:02:13.840 --> 00:02:18.639
And I think the productivity paradox summarizes so much of what we're seeing right now with with AI.
00:02:18.719 --> 00:02:21.039
So I'm excited to dive into that.
00:02:21.599 --> 00:02:29.039
So, Phil and Mark, both of you think very carefully about AI in general, but in particular in very human terms, right?
00:02:29.199 --> 00:02:32.719
Not just as a technical problem, but as a question of work design.
00:02:32.960 --> 00:02:46.960
And what we're starting to see across the board is that many of our org structures, when we think about tasks and roles and trust and space as well, physical and virtual, they're being called into question in the world of AI.
00:02:47.199 --> 00:02:49.360
I want to start with roles.
00:02:49.599 --> 00:02:54.800
So, new roles within our organizations, evolved roles within our organizations.
00:02:55.039 --> 00:02:58.960
Phil, you've described the need for a chief work officer.
00:02:59.199 --> 00:03:00.639
Can you unpack that for us?
00:03:00.800 --> 00:03:02.080
What does that mean?
00:03:02.319 --> 00:03:04.080
What are they responsible for?
00:03:04.319 --> 00:03:07.680
And why do you think this role is important right now?
00:03:08.159 --> 00:03:08.560
Sure.
00:03:08.800 --> 00:03:24.560
So a couple months ago, when Tracy Franklin was uh sort of nominated as the head of HR and IT at Life Sciences for Moderna, it was a lot of like news about kind of future forward roles and what that means for marrying a human and agentic workforce.
00:03:24.800 --> 00:03:28.960
And having just started my newsletter, I kind of leapt to the page and said, This is excellent.
00:03:29.039 --> 00:03:33.039
I love this progress, but I think there's an element that's missing.
00:03:33.120 --> 00:03:36.560
And I cited a report that's now 12 years old.
00:03:36.639 --> 00:03:44.400
So like I did not come up with the term actually chief of work that I now hear more commonly as chief work officer, but you know, potato potato.
00:03:44.879 --> 00:03:53.439
Um, where the original definition was more about someone who is responsible for the holistic experience of working for a company.
00:03:53.520 --> 00:04:08.240
And that included a kind of policy and talent and cultural principles that may come from HR, uh, digital workplaces as Mark hinted at from uh from IT, the tools that we use, how we use the tools that we use, uh, and then the physical workplace.
00:04:08.319 --> 00:04:17.360
So all of those things having to be looked after by some number of people or group of people uh without any real specification about where that sits.
00:04:17.759 --> 00:04:27.120
But it sent me down a path of just realizing, like, especially as our day at work is getting gnarlier by the minute, less predictable by the minute.
00:04:27.199 --> 00:04:29.759
Um, we're working for multiple teams, we're working with different tools.
00:04:29.839 --> 00:04:36.079
You have people like me uh who has, I don't know, a half a dozen different email addresses for different client projects.
00:04:36.240 --> 00:04:46.639
Um, everything is getting more chaotic, and we haven't even really entered into the the next realm of future of work topics on the kind of gig and freelance point, right?
00:04:46.720 --> 00:04:50.560
We've we've covered remote and hybrid, not covered, but it's very much here.
00:04:50.720 --> 00:04:52.639
And AI is very much here.
00:04:52.800 --> 00:05:02.560
But as our team composition starts to include different ranges of uh employees or contractors or freelancers in addition to agents, the variability in our day will go up.
00:05:02.639 --> 00:05:31.920
And I believe that someone has to be sitting there off on the side, hoovering up all the signals about the journeys and the choices that we make, not just to make sure that it feels good in a you know, hugs and kittens kind of way, but are we making the best choices for us, for our teams, for the organization to balance kind of our objectives and their objectives, given all the signals that are available and the choices that other people may be making that I may not be aware of?
00:05:32.079 --> 00:05:35.519
A little bit of like a organizational Fitbit, so to speak.
00:05:36.240 --> 00:05:37.439
It's it's fascinating.
00:05:37.519 --> 00:05:48.639
And I want to drill into the physical environment piece, because as you mentioned, we have you know seen and heard the return to office debate for for years now.
00:05:48.879 --> 00:05:52.240
We're seeing AI infiltrate the workplace.
00:05:52.399 --> 00:05:59.279
Why is the physical environment so important when our world is becoming increasingly digital?
00:05:59.839 --> 00:06:00.560
That's a great idea.
00:06:00.720 --> 00:06:09.040
Well, uh one precondition maybe to keep in mind, um, most companies never asked people what they thought about the place where they spent half their day.
00:06:09.279 --> 00:06:14.639
And that largely comes from the fact that changes in the built environment are slow and expensive.
00:06:14.800 --> 00:06:18.639
If somebody doesn't like a wall, it's difficult to do something about that.
00:06:18.800 --> 00:06:24.639
Whereas if you say, oh gosh, this Google Doc is not very helpful now, you can just change it, right?
00:06:24.720 --> 00:06:28.959
Like you can start another Slack channel, you can try a different way to run a meeting.
00:06:29.120 --> 00:06:30.639
Um, that's free.
00:06:30.800 --> 00:06:34.319
Uh, but to move buildings or change things in the environment is not.
00:06:34.800 --> 00:06:42.879
And uh therefore a lot of facility teams even just come from a world of never asking for feedback, despite the fact that we were there all the time.
00:06:43.120 --> 00:06:44.879
But we didn't have a choice.
00:06:45.199 --> 00:06:56.000
Now, even for the people who are working for companies that have a very strong preference for their presence, no matter what, they have a choice.
00:06:56.160 --> 00:07:04.639
The bar is much lower now for when I'm gonna bail out because of you know delayed flights or traffic or weather or something, or my kid being sick.
00:07:04.959 --> 00:07:09.600
So um we don't we don't go to bad restaurants twice.
00:07:09.920 --> 00:07:13.439
And most offices are unfortunately like bad restaurants.
00:07:13.519 --> 00:07:16.959
Uh and yeah, I see in the the chat this is I'm talking about the private sector.
00:07:17.040 --> 00:07:23.199
It is definitely um more of a problem than the public sector where there's less capital spent on those environments.
00:07:23.439 --> 00:07:31.759
So it's maybe more important than now or than ever now because we're basically making retail-oriented choices about going to work.
00:07:32.000 --> 00:07:37.040
Um, and it's not that we don't travel and take time to do fun things.
00:07:37.199 --> 00:07:42.319
We we will go to great distances for theaters and to visit family and sporting events.
00:07:42.480 --> 00:07:45.120
It's not about that, it's like what is the value to us?
00:07:45.279 --> 00:08:04.079
And that puts a much higher priority on the office having to have a clearly defined and measurable purpose for why I should go there, um, not just because, and needing the infrastructure to allow facilities seems to measure all the choices that we're making there.
00:08:04.319 --> 00:08:28.560
But it just remains this bastion of the old way of thinking about work, uh, because even the most change positive, test and learn, curious, seeming leaders will maybe having having renovated a bathroom that one time at home, think they know what the built environment should be like for their company.
00:08:28.639 --> 00:08:39.200
So they'll be very open to test and learn for AI or for digital workplace, but then say, not only does the office have to look like this, but you have to sit there all the time.
00:08:39.360 --> 00:08:40.399
And that just doesn't clock.
00:08:40.480 --> 00:08:46.639
And maybe the last signal um I'm reading more and more books about future organizations and the future of work holistically.
00:08:46.879 --> 00:08:59.360
And I'd say nine out of ten don't even mention the existence of the built world, whether it's offices or warehouses or hospitals or homes, doesn't make it, it just doesn't come up.
00:08:59.840 --> 00:09:02.639
And I think that's crazy personally.
00:09:03.360 --> 00:09:04.559
It's it's really important.
00:09:04.720 --> 00:09:27.440
I was with a group of CHROs last week and uh talking with a brilliant CHRO who made the decision to relocate his physical office to be right next to the chief digital officer, recognizing that you know where we sit in the office influences, and there's so much research to suggest this is the case, influences how we communicate, how we collaborate.
00:09:27.600 --> 00:09:30.399
And um, it's it's really important to think about.
00:09:30.559 --> 00:09:31.519
So thank you.
00:09:31.840 --> 00:09:33.600
Mark, I want to turn to you.
00:09:33.840 --> 00:09:36.159
You've had such an interesting career.
00:09:36.320 --> 00:09:41.200
You've hold held multiple roles with AI in the title.
00:09:41.679 --> 00:09:49.840
From your time at Easy Cater and now at Glean, what does it look like to have a role with AI in the title?
00:09:49.919 --> 00:09:51.840
And what does that mean right now?
00:09:53.440 --> 00:09:55.919
Um I wish I didn't.
00:09:56.080 --> 00:10:02.559
The the funny part is that AI, and the joke is, and I've been using this and it doesn't come from me, but it's amazing.
00:10:02.639 --> 00:10:04.240
It's uh almost implemented.
00:10:04.320 --> 00:10:06.399
Like so AI isn't like a cooked thing.
00:10:06.480 --> 00:10:07.360
Like we haven't figured out.
00:10:07.440 --> 00:10:12.720
Every time we branch something off, we say, okay, this is a skill is an agent, this is a OCR, whatever the heck it is.
00:10:12.799 --> 00:10:16.559
Like when we productionize it, we give it a name and uh we brand it.
00:10:16.799 --> 00:10:20.879
AI is this nebulous, like smarter than something thing.
00:10:21.039 --> 00:10:29.360
And so uh, I think uh where why it ends up in the title now is because there's such there's so much unknown about what's going on in the market.
00:10:29.679 --> 00:10:37.279
Um for me, it opens doors, like okay, cool, it takes you seriously, it's not a side hustle.
00:10:37.360 --> 00:10:39.360
Oh, you also do the AI thing.
00:10:39.440 --> 00:10:49.679
Uh, so it's important to kind of have that on the placard, but really the the whole thing is is it should kind of melt away and just be part of the work uh or part of your role.
00:10:49.840 --> 00:11:03.360
I don't see it, I think the where we're gonna see it later is that AI is going to be on somebody who's going to be looking at it and saying, what are they doing for the entire company holistically, you know, in that space, you generating LLMs, you're doing stuff uh for that.
00:11:03.440 --> 00:11:21.840
But I think as far as like me, digital workplace, you know, or um yeah, AI outcomes manager is is good now, but really it should just be work outcomes or engagement outcomes because that's what we're doing here, is just kind of going through and saying, how do we make I hate the word work because it work is toil.
00:11:21.919 --> 00:11:25.600
You know, it's like in your brain, it's like the effort on shovel and stuff.
00:11:25.759 --> 00:11:29.360
Um, this is the kind of thing where AI is gonna take it away from there.
00:11:29.440 --> 00:11:35.039
So it's like then we're saying, like, what's the creativity thing that you really want to do, that you love to do, the reason why?
00:11:35.200 --> 00:11:37.200
Hopefully you got into the job you are.
00:11:37.360 --> 00:11:41.759
Uh, that's that's kind of where the where I really think that we're gonna see that shift.
00:11:42.639 --> 00:11:43.440
Fascinating.
00:11:43.600 --> 00:12:00.799
And one of the things, Mark, I've been struck by by you in particular, is you describe yourself as a new, as a near futurist, having that near future lens, someone who thinks about where work is headed in the short-term future, but not necessarily the long-term future.
00:12:01.039 --> 00:12:12.320
Can you talk to us a little bit about that framing and whether you think this near future lens is something that we all ought to adopt or can benefit from adopting?
00:12:12.879 --> 00:12:13.200
Yeah.
00:12:13.519 --> 00:12:17.759
So I joke and tell people the futurists are the Isaac Asma, right?
00:12:17.840 --> 00:12:27.440
So they're long dead before they just start judging what you what you said, uh, which is fine for their relatives, but you don't have to stand up and and kind of face it.
00:12:27.519 --> 00:12:32.960
Uh, and I look at it and say, my role is to look three to four years out.
00:12:33.039 --> 00:12:38.559
Uh, I think it's obviously it's compressing, I think, with with the state of AI and the speed of things.
00:12:38.879 --> 00:12:48.159
But it's important to look at it and say, uh there's a lot of change, but you have to look critically and say, how is this going to impact my role?
00:12:48.320 --> 00:12:50.639
How is this going to impact what we do?
00:12:50.879 --> 00:12:59.360
To Phil's point about the physical office space, I'm looking at it and saying, What how what does that evolution look like in the next few years?
00:12:59.519 --> 00:13:02.080
We're trying to force people back to the office, but what are we doing?
00:13:02.320 --> 00:13:03.039
Why are we doing it?
00:13:03.200 --> 00:13:04.720
Is it because of lack of engagement?
00:13:04.879 --> 00:13:08.720
Or is it because someone has to go and renegotiate a 10-year contract for the rental?
00:13:08.879 --> 00:13:14.559
And so, how do we make the office do a thing that it hasn't done or it hasn't had to do?
00:13:14.720 --> 00:13:16.320
It has to hold its own water.
00:13:16.799 --> 00:13:23.919
So I tend to try to frame everything and say, okay, if I take this out, the logical progression of where this is going, we branch off.
00:13:24.080 --> 00:13:25.840
What does that look like in a couple of years?
00:13:26.000 --> 00:13:37.279
It doesn't have to be perfect, it's just the constant thing I see with the C level executives just trying to work this out because they don't get the luxury of saying, I just want to make a decision that lasts for three, six, nine, twelve months.
00:13:37.519 --> 00:13:39.759
They're looking out two, three, four years.
00:13:39.840 --> 00:13:42.879
And that's a very it's getting harder and harder.
00:13:44.000 --> 00:13:46.320
That reminds me, or just as a riff on that.
00:13:46.480 --> 00:13:54.080
Um, I think it's important to hold a version of the future that is, you know, I judge like the minority report version, right?
00:13:54.159 --> 00:13:56.000
Like, will cars fly one day?
00:13:56.240 --> 00:13:56.960
Probably.
00:13:57.200 --> 00:14:01.759
Soon, no, but we could all kind of get our heads around, like, yeah, probably.
00:14:01.919 --> 00:14:16.159
And then just in making sure that the version that you're holding, whether it's physical or digital or AI, the plausible version in the three to five years is at least inspired by that 10 to 20 year version and not looking backwards.
00:14:16.240 --> 00:14:25.759
And and again on the play side, hybrid as a construct is almost by definition trying to hold on to something that we had before.
00:14:26.080 --> 00:14:27.519
It was like, oh, we did that then.
00:14:27.600 --> 00:14:30.159
And I'm like, oh, I kind of wish we still were, but we're not.
00:14:30.240 --> 00:14:31.600
So I'm gonna call it something new.
00:14:31.759 --> 00:14:36.960
Instead of saying mobility and moving around is almost inevitable.
00:14:37.039 --> 00:14:37.120
Yeah.
00:14:37.279 --> 00:14:45.279
Uh, whether that's sharing in our own space or moving between home and co-working or being driven around by driverless cars to meetings that we didn't know we were gonna have until that morning, right?
00:14:45.440 --> 00:14:47.200
That's probably going to happen.
00:14:47.360 --> 00:14:52.960
So it's ridiculous that we're fighting about, you know, how early in advance should I be able to book a desk, right?
00:14:53.039 --> 00:14:54.480
It just feels very restrained.
00:14:54.639 --> 00:14:56.639
And the same can be applied to AI, right?
00:14:56.720 --> 00:15:03.519
You can either look backwards and say, Oh, I'm gonna do the thing I used to do differently instead of reimagining the way you're gonna do it for a different.
00:15:04.240 --> 00:15:12.399
I think it's for me, it's the effectiveness, digital effectiveness, officer, because it's really then it's then it's whatever the technology or whatever the the thing is, really.
00:15:12.480 --> 00:15:14.399
It's how do you work effectively?
00:15:14.639 --> 00:15:23.360
Um but that's that that's the the nuance in this, is just to go through and say, hey, we have, but you have to have something of a North Star.
00:15:23.600 --> 00:15:26.240
You have to say, look, in three years, we're kind of heading in this direction.
00:15:26.399 --> 00:15:37.279
Otherwise, you just it's the anxiety, it's ADD theater where you know the new thing comes and you shift and shift and shift, and change management is is getting horribly overlooked.
00:15:37.440 --> 00:15:41.759
Um, AI is is more about the human condition than it is the technical.
00:15:41.840 --> 00:15:43.200
The technical is relatively easy.
00:15:43.279 --> 00:15:44.240
How do we get people to use it?
00:15:44.320 --> 00:15:46.159
How do we get the people to use it too effectively?
00:15:46.320 --> 00:15:48.960
And then what happens when they use it effectively?
00:15:49.039 --> 00:15:50.320
What do we do with the time?
00:15:50.480 --> 00:15:52.960
So it's like there's a whole bunch of things we have to do here.
00:15:53.120 --> 00:16:01.600
So effectiveness is really what I would say is if I was looking for a title, that'd be you know, like digital effectiveness or chief effectiveness officer.
00:16:01.759 --> 00:16:05.840
Because as weird as that sounds, it's about where do you need to go to be effective?
00:16:05.919 --> 00:16:07.200
What do you need to do to be effective?
00:16:07.279 --> 00:16:09.519
And what what what do you need to be supported?
00:16:10.480 --> 00:16:19.840
And I would argue that this you know near future lens is especially important in a world of a lot of uncertainty, too, because employees are looking for that clarity.
00:16:20.000 --> 00:16:31.279
We see it all the time in AI policies and AI principles, having a policy and a principle and acknowledging that it's a working, living, breathing document that is going to evolve.
00:16:31.440 --> 00:16:42.480
Employees, you know, in almost every case would much rather have that than nothing because it gives them, you know, the clarity in terms of when can I experiment versus, you know, when can I not?
00:16:42.639 --> 00:16:45.840
How do I use this technology and push the boundaries?
00:16:47.039 --> 00:16:55.039
So I want to turn to one of the big themes of our AI Transformation 100 report, which is around structure and ownership.
00:16:55.200 --> 00:16:56.480
Where does AI live?
00:16:56.639 --> 00:17:01.919
And in particular, when do you centralize versus when do you decentralize?
00:17:02.159 --> 00:17:03.440
Mark, I'll start with you.
00:17:03.600 --> 00:17:08.559
How do you think about this tension between centralization and decentralization?
00:17:09.119 --> 00:17:12.960
Do you have examples of organizations that have gotten this right?
00:17:13.119 --> 00:17:15.599
And when does it tend to break down?
00:17:17.200 --> 00:17:19.359
I think everyone's kind of working it out.
00:17:19.519 --> 00:17:29.119
Some of them I would say it's difficult to say what's right and what's wrong because of the fact that, you know, how long do you give them before you figure out, you know, did it work or did it not?
00:17:29.440 --> 00:17:34.640
I see things that are definitely better than others, and I like the patterning.
00:17:34.799 --> 00:17:46.319
Um, I did a responsibility pledge to go in and look at it and say, how do we deal with the fear and the resistance by the uh the company employees?
00:17:46.400 --> 00:17:47.440
It's something that you don't do.
00:17:47.519 --> 00:17:50.480
You normally you come in with the software and say, hey, we're putting a Microsoft Office.
00:17:50.720 --> 00:17:53.039
Nobody argues it's just like train them and go.
00:17:53.200 --> 00:17:54.799
Well, AI is a scary thing.
00:17:54.880 --> 00:17:58.880
Like some people might have a myth in their head, or they, this is gonna take my job.
00:17:59.039 --> 00:18:01.039
And so you have to deal with some of those things.
00:18:01.119 --> 00:18:07.759
And so you have to have centralization to manage kind of the overarching discussion.
00:18:07.920 --> 00:18:08.960
Like, what is this gonna do?
00:18:09.039 --> 00:18:10.480
And how are you going to apply it?
00:18:10.559 --> 00:18:13.519
And how do you communicate that to the employees?
00:18:13.680 --> 00:18:14.559
It's not as scary.
00:18:14.799 --> 00:18:24.400
Are you are you going to like some of these big companies say, okay, well, we're gonna do this thing and we're not gonna we're not gonna raise headcount, and hey, maybe we might do layoffs uh if they take ownership of that.
00:18:24.559 --> 00:18:28.240
If they don't, then you leave it up to the employees and they get nervous.
00:18:28.400 --> 00:18:40.319
And so I look at it and say, it's where I think it's best and when I'm seeing better activity is ones that are getting out in front of that and saying, what is AI doing for us and what it what is the expectation?
00:18:40.480 --> 00:18:45.599
I want it to augment ours our people so that we don't have to raise headcount to grow.
00:18:45.839 --> 00:18:50.640
Um, so it's not about not headcount, it's just smart headcount.
00:18:50.720 --> 00:18:54.799
Uh, get out in front of it, talk about it, and then give TASS a permission to drive that.
00:18:54.880 --> 00:19:00.799
And then individually, and this is where the decentralization has come in, give them the tools and then get out of their way.
00:19:00.960 --> 00:19:03.519
Because there's these are people that have been in these roles for a long time.
00:19:03.599 --> 00:19:14.400
And if they had the time to step back and look at what they do and what other people do, I guarantee you they already have 10 ideas that they just haven't had the time or the wherewithal to expand on.
00:19:14.559 --> 00:19:22.480
Now, if you say, hey, you can create a skill, you can create a workflow, you could you could think differently about how you do work, uh, do this job better.
00:19:22.799 --> 00:19:26.160
You've basically created RD at every level.
00:19:27.359 --> 00:19:34.240
And Phil, before I turn to you, Mark, you alluded to this idea or this concept of an AI responsibility pledge.
00:19:34.400 --> 00:19:35.920
I think it's very novel.
00:19:36.079 --> 00:19:40.559
Can you describe a little bit about what that was and what the motivation behind it was?
00:19:41.039 --> 00:19:47.599
Yeah, when I first joined Easy Kidder, I I weirdly went to the wait, I had to start to implement AI.
00:19:47.680 --> 00:19:54.160
They were behind it, and uh I saw the early signs of that tacit resistance.
00:19:54.240 --> 00:20:03.519
Okay, because the unlike anything else, if like they could say they use it, like AI in in the workplace can be used or can be ignored, they can continue to do what they do.
00:20:03.680 --> 00:20:09.519
Um, and so they don't have to be vocally resistant, they could just be not using it as much or not effectively.
00:20:09.599 --> 00:20:17.359
And and you could see it as like, okay, they're doing something, they're summarizing a document that's not really effective uses, or just kind of uh checking the box.
00:20:17.519 --> 00:20:19.519
So, Willie, what is the effective use?
00:20:19.759 --> 00:20:21.039
They can be resistant.
00:20:21.119 --> 00:20:27.039
So, what we wanted to do is say uh at EasyKitter was uh very customer service heavy, and that was the gold standard, right?
00:20:27.200 --> 00:20:28.319
Customer service.
00:20:28.559 --> 00:20:44.400
Um, so it was important to go out and say, look, the idea here is AI is is here to make our people better, to spend more time with the customer because I took away a bunch of the things that they didn't have to do, all those note-taking or all this digging for information or whatever.
00:20:44.559 --> 00:20:52.000
How do we support the team so that they can take the extra time and make sure that that white glove, the human experience is better?
00:20:52.160 --> 00:20:56.720
It wasn't about how do we take AI and shove it in front of the customer and back away from that.
00:20:56.799 --> 00:20:58.079
So make it more human.
00:20:58.240 --> 00:21:05.920
And so the pledge was just a way for the executives to kind of go and say, look, guys, we're not doing this to cut people.
00:21:06.000 --> 00:21:07.039
We want to make you better.
00:21:07.119 --> 00:21:11.279
We want you to have time, we want our customers to feel extra supported.
00:21:11.440 --> 00:21:14.799
And it just outlined kind of that agreement.
00:21:14.880 --> 00:21:21.200
It was just like going out and saying the quiet part out loud, but being very intentional about it.
00:21:21.359 --> 00:21:25.680
I think it gives a lot of people like, okay, cool, they're not coming for my job.
00:21:25.839 --> 00:21:27.039
I feel supported.
00:21:27.279 --> 00:21:28.559
This is what they're saying.
00:21:28.640 --> 00:21:32.000
And and then you can go in it and it's a much healthier environment.
00:21:32.880 --> 00:21:38.400
I think it's it's really important and something we can all take pieces from, I think, as inspiration.
00:21:38.799 --> 00:21:50.480
So, Phil, one of the things that both you and Mark have already alluded to today is that, you know, the more AI is embedded in the flow of work, the more that's a sign things are working.
00:21:50.559 --> 00:21:56.240
That, you know, you in particular have said we shouldn't think about doing AI just as we don't think about doing the internet.
00:21:56.559 --> 00:21:59.759
Instead, organizations should clearly state goals.
00:22:00.079 --> 00:22:04.079
And eliminate the micro behaviors that slow work down.
00:22:04.559 --> 00:22:06.559
What changes does that require?
00:22:06.720 --> 00:22:08.960
How do you do this in practice?
00:22:09.359 --> 00:22:09.599
Yeah.
00:22:09.839 --> 00:22:21.839
So first, uh all credit due to Anish Raman, who's an economist at LinkedIn, who I think is the first person I heard say that on a stage, the version of like, we don't say we do internet, and like just like we don't uh we say, like, oh, I'm gonna book tickets now.
00:22:22.000 --> 00:22:23.119
I'm gonna book my vacation.
00:22:23.200 --> 00:22:24.559
We don't say I'm like going to the internet.
00:22:24.799 --> 00:22:26.960
And we'll the same thing will be true for uh for AI.
00:22:27.119 --> 00:22:36.000
And I think a lot of the answer and what we have to change, uh Rebecca, comes back to what what all of us have been flirting with a little bit is both change management and clarity.
00:22:36.400 --> 00:22:45.119
And change management today, like I think employees are so sensitive to like low trust issues with organizations.
00:22:45.279 --> 00:22:50.960
We were all treated genuinely like humans in 2020 and 2021.
00:22:51.039 --> 00:22:54.880
It was the first time when all the, you know, the politics came down.
00:22:54.960 --> 00:23:00.000
Like I'm a human who feels unsafe, you are a human who feels unsafe, and we saw our leaders in a different way.
00:23:00.079 --> 00:23:08.880
And I think we're very sensitive now to how far the pendulum has swung the other way back to kind of corporate speak and uh wanted to be really careful what you say for shareholders.
00:23:09.039 --> 00:23:19.440
But a well-managed change, uh, or one that is really set up for sustainable success starts with kind of brutal honesty of the definition of what's wrong.
00:23:20.079 --> 00:23:24.640
So when employees are told, Oh, like Chat GPT is here, you have to do things differently.
00:23:25.440 --> 00:23:32.880
Uh, they're kind of asking, before you heard about this today, what yesterday did you think was actually broken?
00:23:33.359 --> 00:23:35.680
Like, did you think our sales numbers were not high enough?
00:23:35.759 --> 00:23:37.599
Did you think we were spending too much time in meetings?
00:23:37.759 --> 00:23:41.119
Did you think we had too much space in the office or whatever it was?
00:23:41.279 --> 00:23:51.839
If you had no preconceived notion that anything was wrong yesterday, and now you're telling me, you're trying to sell me on a future that I'm not aligned with, I'm going to, as Mark said, resist.
00:23:52.240 --> 00:23:58.960
And that that second part is uh maybe setting more of a North Star vision is very similar to a responsibility pledge.
00:23:59.039 --> 00:24:03.200
Like I'm going to articulate for you first what's broken.
00:24:03.359 --> 00:24:09.920
And that should be clear enough with uh backed by data and who made the decision that something's actually broken.
00:24:10.079 --> 00:24:42.400
So that when we step into the phase of, I'm now going to tell you where we're going over the next two or three years and the different thematic ways we're going to do it, how I'm going to know we're making that progress, and pledge to you that if little changes we make along the way, which could be we try a new tool or we combine two teams or we move you around, uh, if something doesn't work in service of that broader vision, like we'll back off and say, all right, the micro stream didn't work, but we're we've got lots of, we have a whole portfolio of changes going on.
00:24:42.480 --> 00:24:44.880
Um, but you know why we're doing it, I know why we're doing it.
00:24:44.960 --> 00:24:51.519
It has a long enough vision that the day in and day out bumps and bruises are not gonna kind of uh shift the whole game.
00:24:51.839 --> 00:24:59.359
And that gives people a feeling of psychological safety, a feeling of being able to participate, know that their little actions ladder up to something bigger.
00:24:59.599 --> 00:25:01.200
Now we're just missing that.
00:25:01.359 --> 00:25:09.599
Um, every every even change management role that I've seen kind of come across my feed is 90% communications.
00:25:09.759 --> 00:25:11.839
It feels like internal communications role.
00:25:11.920 --> 00:25:15.200
I'm going to tell you what we have decided behind closed doors.
00:25:15.359 --> 00:25:23.039
And that just does not sit well with employees today, I think, especially in a moment of sort of economic uncertainty where we are.
00:25:23.759 --> 00:25:26.799
And if I may, piggyback on that just for a moment.
00:25:26.880 --> 00:25:37.359
I try to put everything in a lens, and I'm not certified, but I would encourage a lot of people, and it's not not an ad for ProSi, but like ad car is uh awareness, desire, like knowledge, right?
00:25:37.519 --> 00:25:44.720
The change management is probably going to be bigger than we've ever seen it because there the pace of change is is there.
00:25:44.880 --> 00:25:47.039
We have to create that awareness and desire.
00:25:47.200 --> 00:25:47.519
Why?
00:25:47.680 --> 00:25:48.880
What value does this have?
00:25:48.960 --> 00:25:51.599
You can't just come in and like push AI into every role.
00:25:51.680 --> 00:25:54.319
I see someone talking about that in the comments.
00:25:54.480 --> 00:25:56.480
It it's it may not be appropriate.
00:25:56.640 --> 00:25:58.319
Like what you do might be effective.
00:25:58.400 --> 00:26:02.000
And so what you have to do is kind of say, hey, this is a tool to get there.
00:26:02.160 --> 00:26:07.440
So it's like cool, you give it to them and let them do it, but don't say, hey, you have to do it.
00:26:07.599 --> 00:26:14.799
And then and then, but you have to give them the tools, the understanding, and the desire to want to address things and say, like, is there an opportunity?
00:26:14.960 --> 00:26:16.480
If it's an opportunity, go for it.
00:26:16.720 --> 00:26:20.240
We need a new discipline, I think, for change definition before we get to change management.
00:26:20.319 --> 00:26:21.680
And maybe that's the problem.
00:26:22.079 --> 00:26:24.240
I I think it's it's incredibly important.
00:26:24.319 --> 00:26:27.920
And that clarity, the why is something we consistently see.
00:26:28.079 --> 00:26:29.119
And this isn't new.
00:26:29.279 --> 00:26:32.559
We've known, you know, we saw it all the time with hybrid and remote work.
00:26:32.880 --> 00:26:39.519
When people understand the why behind the policy, they're much more likely to agree with the policy or at least support it.
00:26:39.599 --> 00:26:41.440
And I think that's true with AI.
00:26:41.599 --> 00:26:50.160
Change management is hard, it's multifaceted, but you know, a bare minimum is giving people clarity in terms of what is the why behind this and ownership.
00:26:50.319 --> 00:26:50.559
Yeah.
00:26:50.720 --> 00:26:50.880
Yeah.
00:26:51.519 --> 00:26:53.599
And um, okay, but carry on.
00:26:53.680 --> 00:26:54.799
I know we're gonna be quite moving.
00:26:55.440 --> 00:27:05.200
Okay, well, I'll move on to a related topic, which is you know, this tension that we're seeing between top-down change and bottom-up change.
00:27:05.279 --> 00:27:12.559
And Phil, you just alluded um so eloquently to how we can't rely too heavily on top-down change.
00:27:12.799 --> 00:27:14.559
Bottom-up change is also very important.
00:27:14.720 --> 00:27:23.759
And one of the ways that organizations have done this, and Mark, you've done it in your roles, is activating AI champions.
00:27:23.920 --> 00:27:27.680
AI champions, sometimes they're called AI influencers within the organization.
00:27:28.240 --> 00:27:34.319
Mark, how in your past role and current role do you identify those champions?
00:27:34.400 --> 00:27:39.680
And what is their actual mandate day to day in terms of AI transformation?
00:27:40.640 --> 00:27:58.559
Um It's it's the top 18, 20%, the people who are they're they're the people who want to be in on the beta, they're the people who want like intrigued, they're they'll eat a sandwich over a keyboard, not because they have to, but because there's something interesting.
00:27:58.799 --> 00:28:05.920
Um they genuinely want to help themselves, and then they're they're the kind of people that vote are vocal about when they win.
00:28:06.079 --> 00:28:11.119
And so I'm always looking for the people who want I I who I can make the hero of the story.
00:28:11.359 --> 00:28:14.000
Um, I can go in and I can say this is easy.
00:28:14.160 --> 00:28:15.839
And everybody goes, Yeah, it's easy.
00:28:15.920 --> 00:28:18.000
Yeah, you're you know, PhD and nerd them.
00:28:18.160 --> 00:28:26.240
And so, like I go in and say that I really look at it and go try to find for me, a champion is someone I can give them the information.
00:28:26.400 --> 00:28:33.039
They try it, they use it on themselves, they they validate the situation, and then they go in and they they talk to their team.
00:28:33.119 --> 00:28:35.119
And I call it the rule of the sixes, right?
00:28:35.200 --> 00:28:37.680
So that person probably touches six people.
00:28:37.839 --> 00:28:42.559
Those six people, if you look at fall off, those they touch three and those three one.
00:28:42.880 --> 00:28:48.799
And so you look at it and just say, hey, how many champions do I need to move an organization at some level?
00:28:48.960 --> 00:28:54.079
Uh, so I'm looking for that percentage of people that I can just go in there and say, Look, you can come back with your questions.
00:28:54.160 --> 00:28:55.279
I feed the hungry.
00:28:55.440 --> 00:29:00.079
Go in there and tell them, don't bother with the main body who are their head down, they're busy.
00:29:00.160 --> 00:29:06.799
Uh, but when someone shows up in their team and says, Look what I just created, and I did it in three and a half minutes, and we don't have to do that anymore.
00:29:06.880 --> 00:29:07.759
And this is awesome.
00:29:08.000 --> 00:29:09.440
People know them, they like them.
00:29:09.599 --> 00:29:13.680
This is again that the that whole thing about the desire and awareness.
00:29:13.839 --> 00:29:20.079
Um, and they're more likely to adopt change because that person is probably at a similar technology level.
00:29:21.279 --> 00:29:31.839
And either Mark or Phil, I'm curious because I am seeing this strategy in more and more organizations where previously I think there was a narrative and a strategy around let's get everyone adopting AI.
00:29:31.920 --> 00:29:42.160
We need all our employees adopting versus let's focus on this 20% and make sure they're champions so that they can, you know, spread the word, spread these actions to others.
00:29:42.559 --> 00:29:45.039
How do you get then that 80%?
00:29:45.599 --> 00:29:49.680
Do you rely on the good intentions, the goodwill of the 20%?
00:29:50.079 --> 00:30:04.960
Or are there specific strategies that you've seen either of you effective in once you've identified your champions, then spreading that, spreading adoption, excitement, enthusiasm around the technology to the, you know, the other folks within the organization?
00:30:06.160 --> 00:30:14.160
I think the, you know, the the carrot is the other part of successful changes of any kind that is often overlooked.
00:30:14.240 --> 00:30:19.920
And that's the sustaining measures, which could include things like performance management, right?
00:30:20.160 --> 00:30:28.640
Um, you tell people I want you to behave in a different way, but you don't actually put, you know, give it like real teeth.
00:30:28.799 --> 00:30:34.240
Um, in the end, the you know, the the organ, uh, the like the body will reject the organ.
00:30:34.480 --> 00:30:45.359
There's lots of programs that just kind of go back when that one person who was holding together the change agent network who was the most passionate and had the greatest activism and influence, they leave for whatever reason.
00:30:45.519 --> 00:30:47.279
And everyone sort of forgets.
00:30:47.440 --> 00:30:48.559
You go back to the old way.
00:30:48.720 --> 00:31:03.839
So it's back to you know, define the change originally, have a clear vision that people can align on, getting leaders' role modeling the behavior, getting a change agent network that you've got, and then those sort of sustaining change mechanisms and metrics to know that it's all working.
00:31:04.000 --> 00:31:06.400
That's true for any change.
00:31:06.640 --> 00:31:25.839
Um, and failure to go all the way through that chain is very similar to failure in identifying the right sponsors for whatever you're doing, telling uh whole cohorts usually of second and third tier managers that whether they know it or not, you are a sponsor of this change.
00:31:25.920 --> 00:31:26.960
And that's a job.
00:31:27.039 --> 00:31:36.880
And we have to train you on what that means, just like identifying the right change agents and just like uh really onboarding everyone who is influenced by the change, who has a job too.
00:31:37.119 --> 00:31:41.839
You may not have to like it, but you can't totally bury your head in the sand.
00:31:42.000 --> 00:31:52.480
And some of the other change management methodologies, uh, in addition to ProSi, also spend more or less time on really like knighting everybody into those roles.
00:31:52.799 --> 00:31:57.839
And instead of just going, you're supposed to cascade this message, uh, and kind of back to something Mark said.
00:31:57.920 --> 00:32:07.759
In particular, with changes like AI, those second and third level managers may very much be in the um like their own panic and and haven't gotten over the change themselves.
00:32:08.000 --> 00:32:12.400
So they can't uh what is it like put your own oxygen mask on first before helping others?
00:32:12.480 --> 00:32:20.079
Like they're told help others tomorrow and have not had any amount of time to sit with it, and they've got jobs and they've got fears and concerns.
00:32:20.160 --> 00:32:24.480
And this is one of those uh resistible changes.
00:32:24.720 --> 00:32:26.480
So it's it's just all that scaffolding.
00:32:26.559 --> 00:32:32.079
And if you have that and you can align people, and this is how we're going to compensate you, this is how the organization is going to shift.
00:32:32.160 --> 00:32:35.519
Uh, if you don't like it, it will come back to bite you at some point.
00:32:35.599 --> 00:32:42.640
And everyone else hopefully can swing in the more positive, uh, clarity-oriented direction about why we are doing this.
00:32:43.920 --> 00:32:45.200
It's it's so important.
00:32:45.440 --> 00:32:46.079
Thank you, Mark.
00:32:46.160 --> 00:32:46.720
Thank you, Phil.
00:32:46.880 --> 00:32:48.160
Thank you, everyone, for joining us.
00:32:48.240 --> 00:32:52.559
Uh, we'll send along the recording and join us uh next time for the last episode.
00:32:52.720 --> 00:32:53.359
Thanks so much.
00:32:53.599 --> 00:32:58.079
If you enjoyed that conversation, please subscribe so you don't miss another episode.
00:32:58.240 --> 00:33:03.440
And download the AI Transformation 100 report using the link in the show notes.
00:33:03.599 --> 00:33:06.319
Thanks for listening, and we'll see you next time.
00:00:00.160 --> 00:00:04.959
Welcome back to Intelligence Real and Imagined from the Work AI Institute at Glean.
00:00:05.120 --> 00:00:11.679
This is the show where we sort through what's real, what's hype, and what actually works with AI at work.
00:00:12.080 --> 00:00:16.399
I'm your host, Rebecca Hines, and I lead our Work AI Institute here at Glean.
00:00:16.800 --> 00:00:22.960
Today I'm joined by Phil Kirshner from PK Consulting and Mark Christensen from Glean.
00:00:23.199 --> 00:00:27.359
This episode is inspired by our AI Transformation 100 report.
00:00:27.519 --> 00:00:32.159
We're talking about how do you rewire your organization for AI?
00:00:32.479 --> 00:00:37.359
Because real adoption doesn't come from access to the technology alone.
00:00:37.520 --> 00:00:42.000
It comes from redesigning workflows, habits, and decision making.
00:00:42.079 --> 00:00:45.600
So AI becomes part of how work actually gets done.
00:00:45.759 --> 00:00:47.119
Let's dive in.
00:00:50.159 --> 00:00:53.920
I'm joined by two outstanding guests and also friends.
00:00:54.240 --> 00:00:57.359
Phil, why don't you kick us off and introduce yourself?
00:00:57.840 --> 00:00:58.560
Thank you, Rebecca.
00:00:58.640 --> 00:00:59.920
And thank you, everyone, for joining.
00:01:00.079 --> 00:01:01.280
My name is Phil Kirschner.
00:01:01.359 --> 00:01:03.679
I'm a New York City-based modern work consultant.
00:01:03.840 --> 00:01:10.959
Um, have fallen into the world of employee experience by way of workplace strategy in the built environment, which we'll get into uh today.
00:01:11.120 --> 00:01:19.599
Um, I have my own consultancy, work with clients across industries, but usually at the intersection of groups like HR IT and real estate.
00:01:19.840 --> 00:01:28.719
And I write a newsletter called The Work Line, which is designed to help people kind of build their cross-functional courage and lead across those lines for uh better day at work.
00:01:29.439 --> 00:01:30.000
Wonderful.
00:01:30.159 --> 00:01:37.120
And for anyone who hasn't read The Work Line, it's one of my go-to resources for all things work and evolution of work.
00:01:37.200 --> 00:01:39.040
So check it out if you haven't already.
00:01:39.200 --> 00:01:41.040
Uh Mark, please introduce yourself.
00:01:41.519 --> 00:01:42.959
Yeah, Mark Christensen.
00:01:43.040 --> 00:01:48.799
I am uh here in sunny central Florida and uh I've been in tech for 30 years.
00:01:49.200 --> 00:01:57.439
For got into digital workplace and started to figure out uh like how the evolution of things happened post-COVID.
00:01:57.599 --> 00:02:12.639
So I wrote a book called The Productivity Paradox, and uh have been in kind of where tech meets the human condition and uh most recently uh was doing digital workplace and and putting Glean into things, and now I'm working for Glean.
00:02:13.120 --> 00:02:13.680
Wonderful.
00:02:13.840 --> 00:02:18.639
And I think the productivity paradox summarizes so much of what we're seeing right now with with AI.
00:02:18.719 --> 00:02:21.039
So I'm excited to dive into that.
00:02:21.599 --> 00:02:29.039
So, Phil and Mark, both of you think very carefully about AI in general, but in particular in very human terms, right?
00:02:29.199 --> 00:02:32.719
Not just as a technical problem, but as a question of work design.
00:02:32.960 --> 00:02:46.960
And what we're starting to see across the board is that many of our org structures, when we think about tasks and roles and trust and space as well, physical and virtual, they're being called into question in the world of AI.
00:02:47.199 --> 00:02:49.360
I want to start with roles.
00:02:49.599 --> 00:02:54.800
So, new roles within our organizations, evolved roles within our organizations.
00:02:55.039 --> 00:02:58.960
Phil, you've described the need for a chief work officer.
00:02:59.199 --> 00:03:00.639
Can you unpack that for us?
00:03:00.800 --> 00:03:02.080
What does that mean?
00:03:02.319 --> 00:03:04.080
What are they responsible for?
00:03:04.319 --> 00:03:07.680
And why do you think this role is important right now?
00:03:08.159 --> 00:03:08.560
Sure.
00:03:08.800 --> 00:03:24.560
So a couple months ago, when Tracy Franklin was uh sort of nominated as the head of HR and IT at Life Sciences for Moderna, it was a lot of like news about kind of future forward roles and what that means for marrying a human and agentic workforce.
00:03:24.800 --> 00:03:28.960
And having just started my newsletter, I kind of leapt to the page and said, This is excellent.
00:03:29.039 --> 00:03:33.039
I love this progress, but I think there's an element that's missing.
00:03:33.120 --> 00:03:36.560
And I cited a report that's now 12 years old.
00:03:36.639 --> 00:03:44.400
So like I did not come up with the term actually chief of work that I now hear more commonly as chief work officer, but you know, potato potato.
00:03:44.879 --> 00:03:53.439
Um, where the original definition was more about someone who is responsible for the holistic experience of working for a company.
00:03:53.520 --> 00:04:08.240
And that included a kind of policy and talent and cultural principles that may come from HR, uh, digital workplaces as Mark hinted at from uh from IT, the tools that we use, how we use the tools that we use, uh, and then the physical workplace.
00:04:08.319 --> 00:04:17.360
So all of those things having to be looked after by some number of people or group of people uh without any real specification about where that sits.
00:04:17.759 --> 00:04:27.120
But it sent me down a path of just realizing, like, especially as our day at work is getting gnarlier by the minute, less predictable by the minute.
00:04:27.199 --> 00:04:29.759
Um, we're working for multiple teams, we're working with different tools.
00:04:29.839 --> 00:04:36.079
You have people like me uh who has, I don't know, a half a dozen different email addresses for different client projects.
00:04:36.240 --> 00:04:46.639
Um, everything is getting more chaotic, and we haven't even really entered into the the next realm of future of work topics on the kind of gig and freelance point, right?
00:04:46.720 --> 00:04:50.560
We've we've covered remote and hybrid, not covered, but it's very much here.
00:04:50.720 --> 00:04:52.639
And AI is very much here.
00:04:52.800 --> 00:05:02.560
But as our team composition starts to include different ranges of uh employees or contractors or freelancers in addition to agents, the variability in our day will go up.
00:05:02.639 --> 00:05:31.920
And I believe that someone has to be sitting there off on the side, hoovering up all the signals about the journeys and the choices that we make, not just to make sure that it feels good in a you know, hugs and kittens kind of way, but are we making the best choices for us, for our teams, for the organization to balance kind of our objectives and their objectives, given all the signals that are available and the choices that other people may be making that I may not be aware of?
00:05:32.079 --> 00:05:35.519
A little bit of like a organizational Fitbit, so to speak.
00:05:36.240 --> 00:05:37.439
It's it's fascinating.
00:05:37.519 --> 00:05:48.639
And I want to drill into the physical environment piece, because as you mentioned, we have you know seen and heard the return to office debate for for years now.
00:05:48.879 --> 00:05:52.240
We're seeing AI infiltrate the workplace.
00:05:52.399 --> 00:05:59.279
Why is the physical environment so important when our world is becoming increasingly digital?
00:05:59.839 --> 00:06:00.560
That's a great idea.
00:06:00.720 --> 00:06:09.040
Well, uh one precondition maybe to keep in mind, um, most companies never asked people what they thought about the place where they spent half their day.
00:06:09.279 --> 00:06:14.639
And that largely comes from the fact that changes in the built environment are slow and expensive.
00:06:14.800 --> 00:06:18.639
If somebody doesn't like a wall, it's difficult to do something about that.
00:06:18.800 --> 00:06:24.639
Whereas if you say, oh gosh, this Google Doc is not very helpful now, you can just change it, right?
00:06:24.720 --> 00:06:28.959
Like you can start another Slack channel, you can try a different way to run a meeting.
00:06:29.120 --> 00:06:30.639
Um, that's free.
00:06:30.800 --> 00:06:34.319
Uh, but to move buildings or change things in the environment is not.
00:06:34.800 --> 00:06:42.879
And uh therefore a lot of facility teams even just come from a world of never asking for feedback, despite the fact that we were there all the time.
00:06:43.120 --> 00:06:44.879
But we didn't have a choice.
00:06:45.199 --> 00:06:56.000
Now, even for the people who are working for companies that have a very strong preference for their presence, no matter what, they have a choice.
00:06:56.160 --> 00:07:04.639
The bar is much lower now for when I'm gonna bail out because of you know delayed flights or traffic or weather or something, or my kid being sick.
00:07:04.959 --> 00:07:09.600
So um we don't we don't go to bad restaurants twice.
00:07:09.920 --> 00:07:13.439
And most offices are unfortunately like bad restaurants.
00:07:13.519 --> 00:07:16.959
Uh and yeah, I see in the the chat this is I'm talking about the private sector.
00:07:17.040 --> 00:07:23.199
It is definitely um more of a problem than the public sector where there's less capital spent on those environments.
00:07:23.439 --> 00:07:31.759
So it's maybe more important than now or than ever now because we're basically making retail-oriented choices about going to work.
00:07:32.000 --> 00:07:37.040
Um, and it's not that we don't travel and take time to do fun things.
00:07:37.199 --> 00:07:42.319
We we will go to great distances for theaters and to visit family and sporting events.
00:07:42.480 --> 00:07:45.120
It's not about that, it's like what is the value to us?
00:07:45.279 --> 00:08:04.079
And that puts a much higher priority on the office having to have a clearly defined and measurable purpose for why I should go there, um, not just because, and needing the infrastructure to allow facilities seems to measure all the choices that we're making there.
00:08:04.319 --> 00:08:28.560
But it just remains this bastion of the old way of thinking about work, uh, because even the most change positive, test and learn, curious, seeming leaders will maybe having having renovated a bathroom that one time at home, think they know what the built environment should be like for their company.
00:08:28.639 --> 00:08:39.200
So they'll be very open to test and learn for AI or for digital workplace, but then say, not only does the office have to look like this, but you have to sit there all the time.
00:08:39.360 --> 00:08:40.399
And that just doesn't clock.
00:08:40.480 --> 00:08:46.639
And maybe the last signal um I'm reading more and more books about future organizations and the future of work holistically.
00:08:46.879 --> 00:08:59.360
And I'd say nine out of ten don't even mention the existence of the built world, whether it's offices or warehouses or hospitals or homes, doesn't make it, it just doesn't come up.
00:08:59.840 --> 00:09:02.639
And I think that's crazy personally.
00:09:03.360 --> 00:09:04.559
It's it's really important.
00:09:04.720 --> 00:09:27.440
I was with a group of CHROs last week and uh talking with a brilliant CHRO who made the decision to relocate his physical office to be right next to the chief digital officer, recognizing that you know where we sit in the office influences, and there's so much research to suggest this is the case, influences how we communicate, how we collaborate.
00:09:27.600 --> 00:09:30.399
And um, it's it's really important to think about.
00:09:30.559 --> 00:09:31.519
So thank you.
00:09:31.840 --> 00:09:33.600
Mark, I want to turn to you.
00:09:33.840 --> 00:09:36.159
You've had such an interesting career.
00:09:36.320 --> 00:09:41.200
You've hold held multiple roles with AI in the title.
00:09:41.679 --> 00:09:49.840
From your time at Easy Cater and now at Glean, what does it look like to have a role with AI in the title?
00:09:49.919 --> 00:09:51.840
And what does that mean right now?
00:09:53.440 --> 00:09:55.919
Um I wish I didn't.
00:09:56.080 --> 00:10:02.559
The the funny part is that AI, and the joke is, and I've been using this and it doesn't come from me, but it's amazing.
00:10:02.639 --> 00:10:04.240
It's uh almost implemented.
00:10:04.320 --> 00:10:06.399
Like so AI isn't like a cooked thing.
00:10:06.480 --> 00:10:07.360
Like we haven't figured out.
00:10:07.440 --> 00:10:12.720
Every time we branch something off, we say, okay, this is a skill is an agent, this is a OCR, whatever the heck it is.
00:10:12.799 --> 00:10:16.559
Like when we productionize it, we give it a name and uh we brand it.
00:10:16.799 --> 00:10:20.879
AI is this nebulous, like smarter than something thing.
00:10:21.039 --> 00:10:29.360
And so uh, I think uh where why it ends up in the title now is because there's such there's so much unknown about what's going on in the market.
00:10:29.679 --> 00:10:37.279
Um for me, it opens doors, like okay, cool, it takes you seriously, it's not a side hustle.
00:10:37.360 --> 00:10:39.360
Oh, you also do the AI thing.
00:10:39.440 --> 00:10:49.679
Uh, so it's important to kind of have that on the placard, but really the the whole thing is is it should kind of melt away and just be part of the work uh or part of your role.
00:10:49.840 --> 00:11:03.360
I don't see it, I think the where we're gonna see it later is that AI is going to be on somebody who's going to be looking at it and saying, what are they doing for the entire company holistically, you know, in that space, you generating LLMs, you're doing stuff uh for that.
00:11:03.440 --> 00:11:21.840
But I think as far as like me, digital workplace, you know, or um yeah, AI outcomes manager is is good now, but really it should just be work outcomes or engagement outcomes because that's what we're doing here, is just kind of going through and saying, how do we make I hate the word work because it work is toil.
00:11:21.919 --> 00:11:25.600
You know, it's like in your brain, it's like the effort on shovel and stuff.
00:11:25.759 --> 00:11:29.360
Um, this is the kind of thing where AI is gonna take it away from there.
00:11:29.440 --> 00:11:35.039
So it's like then we're saying, like, what's the creativity thing that you really want to do, that you love to do, the reason why?
00:11:35.200 --> 00:11:37.200
Hopefully you got into the job you are.
00:11:37.360 --> 00:11:41.759
Uh, that's that's kind of where the where I really think that we're gonna see that shift.
00:11:42.639 --> 00:11:43.440
Fascinating.
00:11:43.600 --> 00:12:00.799
And one of the things, Mark, I've been struck by by you in particular, is you describe yourself as a new, as a near futurist, having that near future lens, someone who thinks about where work is headed in the short-term future, but not necessarily the long-term future.
00:12:01.039 --> 00:12:12.320
Can you talk to us a little bit about that framing and whether you think this near future lens is something that we all ought to adopt or can benefit from adopting?
00:12:12.879 --> 00:12:13.200
Yeah.
00:12:13.519 --> 00:12:17.759
So I joke and tell people the futurists are the Isaac Asma, right?
00:12:17.840 --> 00:12:27.440
So they're long dead before they just start judging what you what you said, uh, which is fine for their relatives, but you don't have to stand up and and kind of face it.
00:12:27.519 --> 00:12:32.960
Uh, and I look at it and say, my role is to look three to four years out.
00:12:33.039 --> 00:12:38.559
Uh, I think it's obviously it's compressing, I think, with with the state of AI and the speed of things.
00:12:38.879 --> 00:12:48.159
But it's important to look at it and say, uh there's a lot of change, but you have to look critically and say, how is this going to impact my role?
00:12:48.320 --> 00:12:50.639
How is this going to impact what we do?
00:12:50.879 --> 00:12:59.360
To Phil's point about the physical office space, I'm looking at it and saying, What how what does that evolution look like in the next few years?
00:12:59.519 --> 00:13:02.080
We're trying to force people back to the office, but what are we doing?
00:13:02.320 --> 00:13:03.039
Why are we doing it?
00:13:03.200 --> 00:13:04.720
Is it because of lack of engagement?
00:13:04.879 --> 00:13:08.720
Or is it because someone has to go and renegotiate a 10-year contract for the rental?
00:13:08.879 --> 00:13:14.559
And so, how do we make the office do a thing that it hasn't done or it hasn't had to do?
00:13:14.720 --> 00:13:16.320
It has to hold its own water.
00:13:16.799 --> 00:13:23.919
So I tend to try to frame everything and say, okay, if I take this out, the logical progression of where this is going, we branch off.
00:13:24.080 --> 00:13:25.840
What does that look like in a couple of years?
00:13:26.000 --> 00:13:37.279
It doesn't have to be perfect, it's just the constant thing I see with the C level executives just trying to work this out because they don't get the luxury of saying, I just want to make a decision that lasts for three, six, nine, twelve months.
00:13:37.519 --> 00:13:39.759
They're looking out two, three, four years.
00:13:39.840 --> 00:13:42.879
And that's a very it's getting harder and harder.
00:13:44.000 --> 00:13:46.320
That reminds me, or just as a riff on that.
00:13:46.480 --> 00:13:54.080
Um, I think it's important to hold a version of the future that is, you know, I judge like the minority report version, right?
00:13:54.159 --> 00:13:56.000
Like, will cars fly one day?
00:13:56.240 --> 00:13:56.960
Probably.
00:13:57.200 --> 00:14:01.759
Soon, no, but we could all kind of get our heads around, like, yeah, probably.
00:14:01.919 --> 00:14:16.159
And then just in making sure that the version that you're holding, whether it's physical or digital or AI, the plausible version in the three to five years is at least inspired by that 10 to 20 year version and not looking backwards.
00:14:16.240 --> 00:14:25.759
And and again on the play side, hybrid as a construct is almost by definition trying to hold on to something that we had before.
00:14:26.080 --> 00:14:27.519
It was like, oh, we did that then.
00:14:27.600 --> 00:14:30.159
And I'm like, oh, I kind of wish we still were, but we're not.
00:14:30.240 --> 00:14:31.600
So I'm gonna call it something new.
00:14:31.759 --> 00:14:36.960
Instead of saying mobility and moving around is almost inevitable.
00:14:37.039 --> 00:14:37.120
Yeah.
00:14:37.279 --> 00:14:45.279
Uh, whether that's sharing in our own space or moving between home and co-working or being driven around by driverless cars to meetings that we didn't know we were gonna have until that morning, right?
00:14:45.440 --> 00:14:47.200
That's probably going to happen.
00:14:47.360 --> 00:14:52.960
So it's ridiculous that we're fighting about, you know, how early in advance should I be able to book a desk, right?
00:14:53.039 --> 00:14:54.480
It just feels very restrained.
00:14:54.639 --> 00:14:56.639
And the same can be applied to AI, right?
00:14:56.720 --> 00:15:03.519
You can either look backwards and say, Oh, I'm gonna do the thing I used to do differently instead of reimagining the way you're gonna do it for a different.
00:15:04.240 --> 00:15:12.399
I think it's for me, it's the effectiveness, digital effectiveness, officer, because it's really then it's then it's whatever the technology or whatever the the thing is, really.
00:15:12.480 --> 00:15:14.399
It's how do you work effectively?
00:15:14.639 --> 00:15:23.360
Um but that's that that's the the nuance in this, is just to go through and say, hey, we have, but you have to have something of a North Star.
00:15:23.600 --> 00:15:26.240
You have to say, look, in three years, we're kind of heading in this direction.
00:15:26.399 --> 00:15:37.279
Otherwise, you just it's the anxiety, it's ADD theater where you know the new thing comes and you shift and shift and shift, and change management is is getting horribly overlooked.
00:15:37.440 --> 00:15:41.759
Um, AI is is more about the human condition than it is the technical.
00:15:41.840 --> 00:15:43.200
The technical is relatively easy.
00:15:43.279 --> 00:15:44.240
How do we get people to use it?
00:15:44.320 --> 00:15:46.159
How do we get the people to use it too effectively?
00:15:46.320 --> 00:15:48.960
And then what happens when they use it effectively?
00:15:49.039 --> 00:15:50.320
What do we do with the time?
00:15:50.480 --> 00:15:52.960
So it's like there's a whole bunch of things we have to do here.
00:15:53.120 --> 00:16:01.600
So effectiveness is really what I would say is if I was looking for a title, that'd be you know, like digital effectiveness or chief effectiveness officer.
00:16:01.759 --> 00:16:05.840
Because as weird as that sounds, it's about where do you need to go to be effective?
00:16:05.919 --> 00:16:07.200
What do you need to do to be effective?
00:16:07.279 --> 00:16:09.519
And what what what do you need to be supported?
00:16:10.480 --> 00:16:19.840
And I would argue that this you know near future lens is especially important in a world of a lot of uncertainty, too, because employees are looking for that clarity.
00:16:20.000 --> 00:16:31.279
We see it all the time in AI policies and AI principles, having a policy and a principle and acknowledging that it's a working, living, breathing document that is going to evolve.
00:16:31.440 --> 00:16:42.480
Employees, you know, in almost every case would much rather have that than nothing because it gives them, you know, the clarity in terms of when can I experiment versus, you know, when can I not?
00:16:42.639 --> 00:16:45.840
How do I use this technology and push the boundaries?
00:16:47.039 --> 00:16:55.039
So I want to turn to one of the big themes of our AI Transformation 100 report, which is around structure and ownership.
00:16:55.200 --> 00:16:56.480
Where does AI live?
00:16:56.639 --> 00:17:01.919
And in particular, when do you centralize versus when do you decentralize?
00:17:02.159 --> 00:17:03.440
Mark, I'll start with you.
00:17:03.600 --> 00:17:08.559
How do you think about this tension between centralization and decentralization?
00:17:09.119 --> 00:17:12.960
Do you have examples of organizations that have gotten this right?
00:17:13.119 --> 00:17:15.599
And when does it tend to break down?
00:17:17.200 --> 00:17:19.359
I think everyone's kind of working it out.
00:17:19.519 --> 00:17:29.119
Some of them I would say it's difficult to say what's right and what's wrong because of the fact that, you know, how long do you give them before you figure out, you know, did it work or did it not?
00:17:29.440 --> 00:17:34.640
I see things that are definitely better than others, and I like the patterning.
00:17:34.799 --> 00:17:46.319
Um, I did a responsibility pledge to go in and look at it and say, how do we deal with the fear and the resistance by the uh the company employees?
00:17:46.400 --> 00:17:47.440
It's something that you don't do.
00:17:47.519 --> 00:17:50.480
You normally you come in with the software and say, hey, we're putting a Microsoft Office.
00:17:50.720 --> 00:17:53.039
Nobody argues it's just like train them and go.
00:17:53.200 --> 00:17:54.799
Well, AI is a scary thing.
00:17:54.880 --> 00:17:58.880
Like some people might have a myth in their head, or they, this is gonna take my job.
00:17:59.039 --> 00:18:01.039
And so you have to deal with some of those things.
00:18:01.119 --> 00:18:07.759
And so you have to have centralization to manage kind of the overarching discussion.
00:18:07.920 --> 00:18:08.960
Like, what is this gonna do?
00:18:09.039 --> 00:18:10.480
And how are you going to apply it?
00:18:10.559 --> 00:18:13.519
And how do you communicate that to the employees?
00:18:13.680 --> 00:18:14.559
It's not as scary.
00:18:14.799 --> 00:18:24.400
Are you are you going to like some of these big companies say, okay, well, we're gonna do this thing and we're not gonna we're not gonna raise headcount, and hey, maybe we might do layoffs uh if they take ownership of that.
00:18:24.559 --> 00:18:28.240
If they don't, then you leave it up to the employees and they get nervous.
00:18:28.400 --> 00:18:40.319
And so I look at it and say, it's where I think it's best and when I'm seeing better activity is ones that are getting out in front of that and saying, what is AI doing for us and what it what is the expectation?
00:18:40.480 --> 00:18:45.599
I want it to augment ours our people so that we don't have to raise headcount to grow.
00:18:45.839 --> 00:18:50.640
Um, so it's not about not headcount, it's just smart headcount.
00:18:50.720 --> 00:18:54.799
Uh, get out in front of it, talk about it, and then give TASS a permission to drive that.
00:18:54.880 --> 00:19:00.799
And then individually, and this is where the decentralization has come in, give them the tools and then get out of their way.
00:19:00.960 --> 00:19:03.519
Because there's these are people that have been in these roles for a long time.
00:19:03.599 --> 00:19:14.400
And if they had the time to step back and look at what they do and what other people do, I guarantee you they already have 10 ideas that they just haven't had the time or the wherewithal to expand on.
00:19:14.559 --> 00:19:22.480
Now, if you say, hey, you can create a skill, you can create a workflow, you could you could think differently about how you do work, uh, do this job better.
00:19:22.799 --> 00:19:26.160
You've basically created RD at every level.
00:19:27.359 --> 00:19:34.240
And Phil, before I turn to you, Mark, you alluded to this idea or this concept of an AI responsibility pledge.
00:19:34.400 --> 00:19:35.920
I think it's very novel.
00:19:36.079 --> 00:19:40.559
Can you describe a little bit about what that was and what the motivation behind it was?
00:19:41.039 --> 00:19:47.599
Yeah, when I first joined Easy Kidder, I I weirdly went to the wait, I had to start to implement AI.
00:19:47.680 --> 00:19:54.160
They were behind it, and uh I saw the early signs of that tacit resistance.
00:19:54.240 --> 00:20:03.519
Okay, because the unlike anything else, if like they could say they use it, like AI in in the workplace can be used or can be ignored, they can continue to do what they do.
00:20:03.680 --> 00:20:09.519
Um, and so they don't have to be vocally resistant, they could just be not using it as much or not effectively.
00:20:09.599 --> 00:20:17.359
And and you could see it as like, okay, they're doing something, they're summarizing a document that's not really effective uses, or just kind of uh checking the box.
00:20:17.519 --> 00:20:19.519
So, Willie, what is the effective use?
00:20:19.759 --> 00:20:21.039
They can be resistant.
00:20:21.119 --> 00:20:27.039
So, what we wanted to do is say uh at EasyKitter was uh very customer service heavy, and that was the gold standard, right?
00:20:27.200 --> 00:20:28.319
Customer service.
00:20:28.559 --> 00:20:44.400
Um, so it was important to go out and say, look, the idea here is AI is is here to make our people better, to spend more time with the customer because I took away a bunch of the things that they didn't have to do, all those note-taking or all this digging for information or whatever.
00:20:44.559 --> 00:20:52.000
How do we support the team so that they can take the extra time and make sure that that white glove, the human experience is better?
00:20:52.160 --> 00:20:56.720
It wasn't about how do we take AI and shove it in front of the customer and back away from that.
00:20:56.799 --> 00:20:58.079
So make it more human.
00:20:58.240 --> 00:21:05.920
And so the pledge was just a way for the executives to kind of go and say, look, guys, we're not doing this to cut people.
00:21:06.000 --> 00:21:07.039
We want to make you better.
00:21:07.119 --> 00:21:11.279
We want you to have time, we want our customers to feel extra supported.
00:21:11.440 --> 00:21:14.799
And it just outlined kind of that agreement.
00:21:14.880 --> 00:21:21.200
It was just like going out and saying the quiet part out loud, but being very intentional about it.
00:21:21.359 --> 00:21:25.680
I think it gives a lot of people like, okay, cool, they're not coming for my job.
00:21:25.839 --> 00:21:27.039
I feel supported.
00:21:27.279 --> 00:21:28.559
This is what they're saying.
00:21:28.640 --> 00:21:32.000
And and then you can go in it and it's a much healthier environment.
00:21:32.880 --> 00:21:38.400
I think it's it's really important and something we can all take pieces from, I think, as inspiration.
00:21:38.799 --> 00:21:50.480
So, Phil, one of the things that both you and Mark have already alluded to today is that, you know, the more AI is embedded in the flow of work, the more that's a sign things are working.
00:21:50.559 --> 00:21:56.240
That, you know, you in particular have said we shouldn't think about doing AI just as we don't think about doing the internet.
00:21:56.559 --> 00:21:59.759
Instead, organizations should clearly state goals.
00:22:00.079 --> 00:22:04.079
And eliminate the micro behaviors that slow work down.
00:22:04.559 --> 00:22:06.559
What changes does that require?
00:22:06.720 --> 00:22:08.960
How do you do this in practice?
00:22:09.359 --> 00:22:09.599
Yeah.
00:22:09.839 --> 00:22:21.839
So first, uh all credit due to Anish Raman, who's an economist at LinkedIn, who I think is the first person I heard say that on a stage, the version of like, we don't say we do internet, and like just like we don't uh we say, like, oh, I'm gonna book tickets now.
00:22:22.000 --> 00:22:23.119
I'm gonna book my vacation.
00:22:23.200 --> 00:22:24.559
We don't say I'm like going to the internet.
00:22:24.799 --> 00:22:26.960
And we'll the same thing will be true for uh for AI.
00:22:27.119 --> 00:22:36.000
And I think a lot of the answer and what we have to change, uh Rebecca, comes back to what what all of us have been flirting with a little bit is both change management and clarity.
00:22:36.400 --> 00:22:45.119
And change management today, like I think employees are so sensitive to like low trust issues with organizations.
00:22:45.279 --> 00:22:50.960
We were all treated genuinely like humans in 2020 and 2021.
00:22:51.039 --> 00:22:54.880
It was the first time when all the, you know, the politics came down.
00:22:54.960 --> 00:23:00.000
Like I'm a human who feels unsafe, you are a human who feels unsafe, and we saw our leaders in a different way.
00:23:00.079 --> 00:23:08.880
And I think we're very sensitive now to how far the pendulum has swung the other way back to kind of corporate speak and uh wanted to be really careful what you say for shareholders.
00:23:09.039 --> 00:23:19.440
But a well-managed change, uh, or one that is really set up for sustainable success starts with kind of brutal honesty of the definition of what's wrong.
00:23:20.079 --> 00:23:24.640
So when employees are told, Oh, like Chat GPT is here, you have to do things differently.
00:23:25.440 --> 00:23:32.880
Uh, they're kind of asking, before you heard about this today, what yesterday did you think was actually broken?
00:23:33.359 --> 00:23:35.680
Like, did you think our sales numbers were not high enough?
00:23:35.759 --> 00:23:37.599
Did you think we were spending too much time in meetings?
00:23:37.759 --> 00:23:41.119
Did you think we had too much space in the office or whatever it was?
00:23:41.279 --> 00:23:51.839
If you had no preconceived notion that anything was wrong yesterday, and now you're telling me, you're trying to sell me on a future that I'm not aligned with, I'm going to, as Mark said, resist.
00:23:52.240 --> 00:23:58.960
And that that second part is uh maybe setting more of a North Star vision is very similar to a responsibility pledge.
00:23:59.039 --> 00:24:03.200
Like I'm going to articulate for you first what's broken.
00:24:03.359 --> 00:24:09.920
And that should be clear enough with uh backed by data and who made the decision that something's actually broken.
00:24:10.079 --> 00:24:42.400
So that when we step into the phase of, I'm now going to tell you where we're going over the next two or three years and the different thematic ways we're going to do it, how I'm going to know we're making that progress, and pledge to you that if little changes we make along the way, which could be we try a new tool or we combine two teams or we move you around, uh, if something doesn't work in service of that broader vision, like we'll back off and say, all right, the micro stream didn't work, but we're we've got lots of, we have a whole portfolio of changes going on.
00:24:42.480 --> 00:24:44.880
Um, but you know why we're doing it, I know why we're doing it.
00:24:44.960 --> 00:24:51.519
It has a long enough vision that the day in and day out bumps and bruises are not gonna kind of uh shift the whole game.
00:24:51.839 --> 00:24:59.359
And that gives people a feeling of psychological safety, a feeling of being able to participate, know that their little actions ladder up to something bigger.
00:24:59.599 --> 00:25:01.200
Now we're just missing that.
00:25:01.359 --> 00:25:09.599
Um, every every even change management role that I've seen kind of come across my feed is 90% communications.
00:25:09.759 --> 00:25:11.839
It feels like internal communications role.
00:25:11.920 --> 00:25:15.200
I'm going to tell you what we have decided behind closed doors.
00:25:15.359 --> 00:25:23.039
And that just does not sit well with employees today, I think, especially in a moment of sort of economic uncertainty where we are.
00:25:23.759 --> 00:25:26.799
And if I may, piggyback on that just for a moment.
00:25:26.880 --> 00:25:37.359
I try to put everything in a lens, and I'm not certified, but I would encourage a lot of people, and it's not not an ad for ProSi, but like ad car is uh awareness, desire, like knowledge, right?
00:25:37.519 --> 00:25:44.720
The change management is probably going to be bigger than we've ever seen it because there the pace of change is is there.
00:25:44.880 --> 00:25:47.039
We have to create that awareness and desire.
00:25:47.200 --> 00:25:47.519
Why?
00:25:47.680 --> 00:25:48.880
What value does this have?
00:25:48.960 --> 00:25:51.599
You can't just come in and like push AI into every role.
00:25:51.680 --> 00:25:54.319
I see someone talking about that in the comments.
00:25:54.480 --> 00:25:56.480
It it's it may not be appropriate.
00:25:56.640 --> 00:25:58.319
Like what you do might be effective.
00:25:58.400 --> 00:26:02.000
And so what you have to do is kind of say, hey, this is a tool to get there.
00:26:02.160 --> 00:26:07.440
So it's like cool, you give it to them and let them do it, but don't say, hey, you have to do it.
00:26:07.599 --> 00:26:14.799
And then and then, but you have to give them the tools, the understanding, and the desire to want to address things and say, like, is there an opportunity?
00:26:14.960 --> 00:26:16.480
If it's an opportunity, go for it.
00:26:16.720 --> 00:26:20.240
We need a new discipline, I think, for change definition before we get to change management.
00:26:20.319 --> 00:26:21.680
And maybe that's the problem.
00:26:22.079 --> 00:26:24.240
I I think it's it's incredibly important.
00:26:24.319 --> 00:26:27.920
And that clarity, the why is something we consistently see.
00:26:28.079 --> 00:26:29.119
And this isn't new.
00:26:29.279 --> 00:26:32.559
We've known, you know, we saw it all the time with hybrid and remote work.
00:26:32.880 --> 00:26:39.519
When people understand the why behind the policy, they're much more likely to agree with the policy or at least support it.
00:26:39.599 --> 00:26:41.440
And I think that's true with AI.
00:26:41.599 --> 00:26:50.160
Change management is hard, it's multifaceted, but you know, a bare minimum is giving people clarity in terms of what is the why behind this and ownership.
00:26:50.319 --> 00:26:50.559
Yeah.
00:26:50.720 --> 00:26:50.880
Yeah.
00:26:51.519 --> 00:26:53.599
And um, okay, but carry on.
00:26:53.680 --> 00:26:54.799
I know we're gonna be quite moving.
00:26:55.440 --> 00:27:05.200
Okay, well, I'll move on to a related topic, which is you know, this tension that we're seeing between top-down change and bottom-up change.
00:27:05.279 --> 00:27:12.559
And Phil, you just alluded um so eloquently to how we can't rely too heavily on top-down change.
00:27:12.799 --> 00:27:14.559
Bottom-up change is also very important.
00:27:14.720 --> 00:27:23.759
And one of the ways that organizations have done this, and Mark, you've done it in your roles, is activating AI champions.
00:27:23.920 --> 00:27:27.680
AI champions, sometimes they're called AI influencers within the organization.
00:27:28.240 --> 00:27:34.319
Mark, how in your past role and current role do you identify those champions?
00:27:34.400 --> 00:27:39.680
And what is their actual mandate day to day in terms of AI transformation?
00:27:40.640 --> 00:27:58.559
Um It's it's the top 18, 20%, the people who are they're they're the people who want to be in on the beta, they're the people who want like intrigued, they're they'll eat a sandwich over a keyboard, not because they have to, but because there's something interesting.
00:27:58.799 --> 00:28:05.920
Um they genuinely want to help themselves, and then they're they're the kind of people that vote are vocal about when they win.
00:28:06.079 --> 00:28:11.119
And so I'm always looking for the people who want I I who I can make the hero of the story.
00:28:11.359 --> 00:28:14.000
Um, I can go in and I can say this is easy.
00:28:14.160 --> 00:28:15.839
And everybody goes, Yeah, it's easy.
00:28:15.920 --> 00:28:18.000
Yeah, you're you know, PhD and nerd them.
00:28:18.160 --> 00:28:26.240
And so, like I go in and say that I really look at it and go try to find for me, a champion is someone I can give them the information.
00:28:26.400 --> 00:28:33.039
They try it, they use it on themselves, they they validate the situation, and then they go in and they they talk to their team.
00:28:33.119 --> 00:28:35.119
And I call it the rule of the sixes, right?
00:28:35.200 --> 00:28:37.680
So that person probably touches six people.
00:28:37.839 --> 00:28:42.559
Those six people, if you look at fall off, those they touch three and those three one.
00:28:42.880 --> 00:28:48.799
And so you look at it and just say, hey, how many champions do I need to move an organization at some level?
00:28:48.960 --> 00:28:54.079
Uh, so I'm looking for that percentage of people that I can just go in there and say, Look, you can come back with your questions.
00:28:54.160 --> 00:28:55.279
I feed the hungry.
00:28:55.440 --> 00:29:00.079
Go in there and tell them, don't bother with the main body who are their head down, they're busy.
00:29:00.160 --> 00:29:06.799
Uh, but when someone shows up in their team and says, Look what I just created, and I did it in three and a half minutes, and we don't have to do that anymore.
00:29:06.880 --> 00:29:07.759
And this is awesome.
00:29:08.000 --> 00:29:09.440
People know them, they like them.
00:29:09.599 --> 00:29:13.680
This is again that the that whole thing about the desire and awareness.
00:29:13.839 --> 00:29:20.079
Um, and they're more likely to adopt change because that person is probably at a similar technology level.
00:29:21.279 --> 00:29:31.839
And either Mark or Phil, I'm curious because I am seeing this strategy in more and more organizations where previously I think there was a narrative and a strategy around let's get everyone adopting AI.
00:29:31.920 --> 00:29:42.160
We need all our employees adopting versus let's focus on this 20% and make sure they're champions so that they can, you know, spread the word, spread these actions to others.
00:29:42.559 --> 00:29:45.039
How do you get then that 80%?
00:29:45.599 --> 00:29:49.680
Do you rely on the good intentions, the goodwill of the 20%?
00:29:50.079 --> 00:30:04.960
Or are there specific strategies that you've seen either of you effective in once you've identified your champions, then spreading that, spreading adoption, excitement, enthusiasm around the technology to the, you know, the other folks within the organization?
00:30:06.160 --> 00:30:14.160
I think the, you know, the the carrot is the other part of successful changes of any kind that is often overlooked.
00:30:14.240 --> 00:30:19.920
And that's the sustaining measures, which could include things like performance management, right?
00:30:20.160 --> 00:30:28.640
Um, you tell people I want you to behave in a different way, but you don't actually put, you know, give it like real teeth.
00:30:28.799 --> 00:30:34.240
Um, in the end, the you know, the the organ, uh, the like the body will reject the organ.
00:30:34.480 --> 00:30:45.359
There's lots of programs that just kind of go back when that one person who was holding together the change agent network who was the most passionate and had the greatest activism and influence, they leave for whatever reason.
00:30:45.519 --> 00:30:47.279
And everyone sort of forgets.
00:30:47.440 --> 00:30:48.559
You go back to the old way.
00:30:48.720 --> 00:31:03.839
So it's back to you know, define the change originally, have a clear vision that people can align on, getting leaders' role modeling the behavior, getting a change agent network that you've got, and then those sort of sustaining change mechanisms and metrics to know that it's all working.
00:31:04.000 --> 00:31:06.400
That's true for any change.
00:31:06.640 --> 00:31:25.839
Um, and failure to go all the way through that chain is very similar to failure in identifying the right sponsors for whatever you're doing, telling uh whole cohorts usually of second and third tier managers that whether they know it or not, you are a sponsor of this change.
00:31:25.920 --> 00:31:26.960
And that's a job.
00:31:27.039 --> 00:31:36.880
And we have to train you on what that means, just like identifying the right change agents and just like uh really onboarding everyone who is influenced by the change, who has a job too.
00:31:37.119 --> 00:31:41.839
You may not have to like it, but you can't totally bury your head in the sand.
00:31:42.000 --> 00:31:52.480
And some of the other change management methodologies, uh, in addition to ProSi, also spend more or less time on really like knighting everybody into those roles.
00:31:52.799 --> 00:31:57.839
And instead of just going, you're supposed to cascade this message, uh, and kind of back to something Mark said.
00:31:57.920 --> 00:32:07.759
In particular, with changes like AI, those second and third level managers may very much be in the um like their own panic and and haven't gotten over the change themselves.
00:32:08.000 --> 00:32:12.400
So they can't uh what is it like put your own oxygen mask on first before helping others?
00:32:12.480 --> 00:32:20.079
Like they're told help others tomorrow and have not had any amount of time to sit with it, and they've got jobs and they've got fears and concerns.
00:32:20.160 --> 00:32:24.480
And this is one of those uh resistible changes.
00:32:24.720 --> 00:32:26.480
So it's it's just all that scaffolding.
00:32:26.559 --> 00:32:32.079
And if you have that and you can align people, and this is how we're going to compensate you, this is how the organization is going to shift.
00:32:32.160 --> 00:32:35.519
Uh, if you don't like it, it will come back to bite you at some point.
00:32:35.599 --> 00:32:42.640
And everyone else hopefully can swing in the more positive, uh, clarity-oriented direction about why we are doing this.
00:32:43.920 --> 00:32:45.200
It's it's so important.
00:32:45.440 --> 00:32:46.079
Thank you, Mark.
00:32:46.160 --> 00:32:46.720
Thank you, Phil.
00:32:46.880 --> 00:32:48.160
Thank you, everyone, for joining us.
00:32:48.240 --> 00:32:52.559
Uh, we'll send along the recording and join us uh next time for the last episode.
00:32:52.720 --> 00:32:53.359
Thanks so much.
00:32:53.599 --> 00:32:58.079
If you enjoyed that conversation, please subscribe so you don't miss another episode.
00:32:58.240 --> 00:33:03.440
And download the AI Transformation 100 report using the link in the show notes.
00:33:03.599 --> 00:33:06.319
Thanks for listening, and we'll see you next time.
