131. Psychological Safety Is Your AI Strategy

I'm Tom Sloth, executive director of TSA. Welcome to Techtonic, the podcast where we explore what makes technology business models successful in today's world. I want to put something on the table that I keep coming back to with our members. When a tech company talks about AI, almost every one of them writes half a sentence. First half is how do we use AI to take cost out, get more efficient, do more fewer people? And look, that's a real half. That's a big story there. But it's not the complete sentence. The second half, the part most companies haven't gotten to yet, is what new value do we create that we never could before? And here's the kicker. The part we've been chipping on, you cannot write that second half of the sentence without your people. You can't grind your way to a workforce that says, "Hey, yes, let's go build something nobody ever built." That's the half of the sentence this episode is about. How to really inspire the workforce around AI? My guest today is Sarah Dumuccio, a social psychologist who works with global companies on the human side of AI transformation, not the technology rollout, the people rollout. And while a lot of the industry is assessing over upskilling and AI fluency, Sarah's doing something most leaders are skipping entirely. She's studying what actually makes people want to follow a leader into something this scary. So she's talking about psychological safety and the difference between bringing people along and dragging them along. And honestly, it's some of the sharpest thinking I've heard on the people side of AI in a while. And so I'm really looking forward to this conversation. Let's get the insight. Engine humming, Sarah, welcome to TikTok. - Thank you so much, Thomas. I'm excited to be here. - So this is gonna be fun. I watched this 11 minute keynote that you did talking about AI adoption and managing people through it. And what really struck me is this thing about resentment, the pushback that we're starting to see in employees and cuts across everything. Age, seniority, gender, job type, all the stuff that people say, well, people are acting differently. This cuts across all of that. So set the table here for everybody. What are you actually seeing in the wild right now when companies roll out AI and why is the human reaction the nervous is so universal? - Such a great question. As a psychologist, I think I can really bring a lot of insight here that we sometimes forget, which is that it's not resistance in the sort of typical sense of just people digging in their heels and saying, I don't wanna do this. It's really this kind of threat calculation that people are doing in real time, right? Across these levels and goals, people are asking themselves, does this make me less competent, less value, less needed, right? And it cuts across all of this because it's not threatening tasks. It's threatening how we think about ourselves, the story that we have about who we are at work. So it's not a demographic problem. It's a psychological one. It's a human problem, right? - Yeah. - And we all have to have a sense of, you know, we're protecting the same things, right? We need to feel competent. We need to feel like we're actually good at something. We need to feel like we're valuable and how can that be valuable if everyone has the same tools. And we also need to feel like we have a sense of identity, right? Am I still the person, the professional that I've been spending decades, years building? And if I'm not there yet, will I ever become that professional that I saw myself becoming? - Because I listened to what's interesting here is again, this concept of feeling valuable. AI is a unique technology because it threatens every knowledge of work position out there. It's not unique to a position. It's not like when spreadsheets came along and accountants felt threatened and that was unique, right? This is like, "Hey, I threatens everybody." It doesn't matter what kind of role you in. This is very unique there. And I'm gonna put some data on the table here. So we did this poll of our executive advisory board a couple of months ago and we were asking about how they were managing the rollout of AI and the training of it for that. And less than 15% of those companies, said they had a real structured program in place. And I was talking to a guy who does implementation software around AI and he was telling me he's watching companies invest in the technology side of AI but not the training side, not the enablement side. And to me, that's kind of crazy. So why is that? Why are companies pouring so much money into the technology and damn near nothing, into the training enablement side? - Yeah, that's a really good question. And I think it's a gap that a lot of people are really noticing, right? This huge gap between the billions of dollars that are being put trillions even into AI and the just minuscule amount that's being put into training. And I think the short answer is that one is quantifiable and the other is, right? You can put a number on a software license or an AI company, right? It gets funded because you can quantify it. Whereas human adaptation is messy, it's slow, it's hard to measure. And it also sort of is treated at this downstream free assumption, right? It's just gonna happen once we switch on the tool. So this is a huge category error, right? Leaders are sort of buying and using this as if it's a software and instead of as a work redesign used to be, you know, here's where the buttons are. So we just need to show you that and you'll be good to go. But with AI enablement is, here's how your judgment, your expertise, your workflow, and your identity actually changed, which is obviously a huge lift. - Yeah, as I listen to you, it strikes me, there's now this sort of irony. In traditional software rollout, I get you rollout a new software tool for the workforce 'cause you said, you know, you gotta teach them what buttons to push. And people go, oh yeah, I gotta train people on that. I don't assume that this software lands on their desk, they magically know what buttons to push, right? So you would do a certain amount of dollars in time on training. AI comes along and the irony here is the fact that we're under investing in the training, just assuming they're gonna know what to do. And yet, a lot more up evil, a lot more training, 'cause like you said, it's about completely rethinking how you're doing your job, not pushing some buttons. So the lift is higher in terms of enablement and the investment ironically is proving to be lower right now. So it makes no sense to me whatsoever. So, you know, you made a distinction. I mean, first time you and I talked, stuck in my mind. We all talk about employees sabotaging AI, but you said the bigger thing isn't sabotage, it's detracting, the quiet, you know, it's not for me. Like, I don't wanna do this. So talk more about that, this sort of quiet checkout. So I'm not stalking, but quiet checkout. And why you think that that's perhaps the bigger issue? - It's a bigger issue, exactly for the reason that you say, is that it's quiet, it's way more common. It's not that sort of, hey, I don't wanna do this, or I'm actually gonna actively push back. It's, you know, that sort of polite opting out, it's, they're technically compliant, but they're not really changing how they work. And it doesn't look like resistance. So it's not, there was a recent study that was written about in Forbes that found that globally 43% of workers bypassed their company's AI tools in the past 30 days and completed the work manually instead. Which I think just some people sounds insane, right? Because this is something that's supposed to help them be more efficient. And this is again, back to another gap that we see, which is that up to 90% of executives say that their employees have the tools that they need, but only 20% of the workers actually agree. So the leaders are not even seeing it, even if they wanted to. They think that everybody is good to go and that makes it doubly difficult to manage, right? Because it's happening behind the scenes and they're not even able to see it based on their own biases. - Yeah, and I think, to me, what's interesting on how this is unfolding, I was talking to an executive, literally, after you and I spoke last time. And I was talking about this challenge of AI adoption and how they're dealing with it at their company. I shared with you a report, you sort of these categories of people, right? In terms of the sabotages versus whatever. And the executive looked at that and said, well, the people that are sabotaging, if we identify that, they're gone. That's how we deal with it. They are gone. And it's like, and I heard that and I get that. I get that, right? But it is to quiet not adoption. That is an executive to your point. If you can't really see that, it's not in your face, it's not super obvious. You can't point to and say, hey, you're out. But it is going to slow you down and you just don't know how much it's slowing you down. So is that costing you 10%, 20%, 40% in terms of how fast you could be living on this stuff? You don't have your arms around it. So I think it's a real problem out there. - Yeah, and you don't find out until it's too late, right? Because it breaks the learning loop and then you find out, unslater, when the numbers come in flat. - Yeah, you didn't get the RI, you didn't get what you thought you were going to get. And I think what a lot of leaders don't have their head around is that they think that enablement in general for AI is, hey, we got everybody a copial license. It's like, here's this really cool AI tool and it's so easy to use. You just talk to it and go do really cool things with it. But you can think something is cool, right? And so still sit there thinking, if I really lean into this, hey, what do I do with it? And also am I really just training my replacement, which is as you put on the table is a real issue. So how do leaders put these tools on the table and help their employees think in a good way about it? And I think also, the leaders have to be honest about the fact that they do expect efficiencies. I mean, I think that's another issue that we identify. Yeah, I first kind of the table. It's asking leaders, what do you think about AI? And the story to the employees was, I've used this term before in this podcast, rainbows and unicorns and it's all gonna be great. But then you talk to them off the record And they'd be like, oh yeah, we definitely are going to slow down hiring or reduce workforce, whatever that was already. thinking from day one. So how do leaders, again, put this on a way that is not just like here's this cool tool figured out, you know, by the way, don't worry about what this is going to do to our workforce configurations. What do you think there? I mean, it's so hard, right? Because there's always this balance that has to be struck in any organization by any leader about transparency and honesty and not actually terrifying people, right? You can't just be honest about everything. At the same time, you can't out demo fear, right? If somebody privately thinks that getting good at this tool is going to speed up their own redundancy, no amount of, look at how cool this is, is going to help that. So I do think that we should name the elephant in the room. I do think that the fear has to be named directly. We have this idea that if we just talk around it, then that will address the problem. But if we name it and then we meet that with honesty rather than the sort of false sense of reassurance, right? If the head count is going to be genuinely uncertain, then I think we have to say that not in a sense of there's something we can do, but more, people can handle uncertainly more than I think we think they can. And they're more likely to forgive that than to feel like they're being lied to or that they're being managed around the truth. So not saying, hey, look, haven't been might be fired tomorrow. But rather, I also am not sure where this is going to lead, but regardless, learning these tools is going to help you in your future. Yeah. I want to ship in seven a little bit because even myself personally, leadership team on a research for AI has changed dramatically the way that we do research. And we've had to navigate that for the past couple of years. And my personal approach on it, and I'm not saying it's the winning approach, but my personal approach on it was to basically get on the table and say, look, there is no doubt this technology is going to change the way that we operate. And if we want to be competitive, if we want to be relevant in the marketplace, we are going to have to dramatically reengineer the way we and nobody knows exactly what that's going to look like. So that's sort of the first message. I think to me, honesty about that, the honesty that this is going to change the way we operate. And it could change organizational structure, it could change headcount requirements. That's a journey we have to go on if we want to be relevant. So I think that honesty is important. The second message that I always had to our researchers was, this is how you personally stay relevant because I do believe this deeply is professional. And you know, this is a research for yourself. I mean, if you sat there and crossed your arms and said, I think AI stinks. I'm just not going to do that, right? Then two years from now, you're not a relevant researcher anymore. So those are the two messages I led with. And as you go through that, there are tough choices you're making around resourcing and everything else that unfold, but at least you're honest about the journey. But I just be curious about your reactions in terms of the messaging you're seeing that works with employees on that. Yeah, but I think that's the right direction, right? It's a human balance, right? It's also as a leader being a bit vulnerable. It's vulnerable to say, I'm not exactly sure what this looks like or where this goes. And that's one of the building blocks to creating trust. And really, this is a huge question of trust, right? If you are going to be brought along and not dragged along, you have to trust the people that are in the driver's seat. You have to think and understand and know that they do have your best interests at heart, but that doesn't mean that they can predict everything or that there aren't mandates coming from above that they can control. And yet, you're still going to be behind them because you feel that they're there, right? And that they are to be trusted and that they're not lying to you and that they're not scheming behind your back. And we know that lots of this is a question of communication. But I think that the approach that you're describing is really the right one. It's saying, I'm not exactly sure what the answer is, but I'm actually here for the ride, right? And I know that this is going to be the direction that we need to go in in the sense of we can only just be learning and experimenting more in order to try and learn these new tools and how it's going to change things together because we're on the same team, essentially. The term you put on the table here, I think is so important right now in the current environment. That is trust. And I think that a lot of leadership teamings over the past couple of years and even AI companies themselves, that is what they lost was trust with their employees. And this is the first technology I've ever seen roll out in my, you know, 35 years in technology. You know, the internet came and it wasn't like, trust me. The internet didn't shake the trust of employees. It was like, oh, this is cool, whatever. You put a new piece of software in and people use it and like, they're not shaken, but AI comes along. And this whole conversation about what does it really mean? The trust levels went way down and then executive teams do these really bonehead, had things of saying one thing and suddenly, oh, but we're really going to take out 30% of the people. We didn't really communicate that or that kind of stuff. And so I think trust in especially in the United States is in an all time low with leadership teams over this particular technology. And I don't see that getting better in the short term. And so I'm curious on now. And this is something we haven't talked about before. But if a leadership team feels like they've dug themselves a hole on this to date, right? Is there any tactics you've seen to help them start to climb out of that one of them you said is to be as honest as possible. It's okay to be vulnerable and say, hey, we don't have all the answers. Those things help anything else help with building that trust back. Yeah, absolutely. I mean, the first is to actually include people in these decisions to the extent that it's possible, right? It's giving them some sense of agency and control over what feels like a completely uncontrollable situation to them. And again, this is the stress on this is the fear and the threats that are coming from above from all sides. So in what way can you balance the uncertainty with things that feel controllable uncertain? And that can be small things, right? That can be like, hey, let's try to experiment this with each other, right? Let's have a workshop where it's just us kind of poking around and figuring out what works with this, right? It's modeling that it's okay to not know. And this set of goes back to this question about psychological safety, right? AI adoption requires people to openly learn in public, right? It's trying out a new prompt. It's telling people that you used something in a way that some people still feel like is cheating, right? It's admitting that you don't trust the output. It's questioning where there are gaps and what they understand and what's going on with the AI and a precondition for that is that you feel safe to do those things. And a lot of what these rollouts are doing is basically saying, this needs to happen now and needs to happen fast and it needs to basically be perfect, right? I better be seeing these efficiency gains immediately. Meanwhile, the rest of us are just trying to figure out how in the world this new technology even works and what it means for us, right? So there has to be space for that because if it goes too fast, then that is going to jeopardize the rollout just as much as people doing it too slowly. So one thing I'm picking up on there that you said that has never struck me before, but it really aligns with the patterns I've seen within TSA, the researchers, as well as I've seen some of our member companies that have been very successful at AI. And that is a concept of co-creation because what I heard you just say is if a management team comes together as employees and says, look, we're going to co-create our future together. How do we use these tools? What are the best use cases? Employees get excited about, they get intellectually excited, like, on part of the solution here. While this is really cool, as opposed to what you just said, the exact opposite of that of saying, look, we've got the solution, you got to use it. Bam, it's got, like you said, up and running and whatever. That jamming in the crazy thing is with this technology, it is so powerful. It creates opportunity for co-creation in a way that no other technology ever has. Other technologies are like, here's the solution, here's what it does, implement it. This is a technology that you don't even know what the freaking use cases are. And it's smart employees starting to play with it. I'll tell you, again, I'm so proud of the TSI research team because they've gone on this journey of initially, hey, do I really have to use AI to, I mean, they are coming up with new use cases on a weekly basis that we share as a team. And I'm just like, my mind is blown like how so many creative ways, but they turn that bend of like, hey, I'm really part of the co-creating solution here. You know, this is not being imposed. So it's never really struck me before, but I think about again, I've watched other member companies and I've watched the employees get excited because they were like, hey, we could create this new solution for our support organization or for our PS organization or for whatever. And they were part of it. It wasn't being imposed, which please this other theme that you've referenced about. And I think it's Amy Edminson's work on psychological safety decided that people want experiment unless, you know, they feel it's safe doing it. So maybe talk a little bit more about how you make it super safe. And you mentioned some of the things that people it's okay to make mistakes, et cetera, anything else to basically build up that psychological safety. So they know it's okay to make mistakes, to try and fail, et cetera. What do you see there in the corporate room? Yeah, I think a lot of it is wrongling. So related to one of the things I said earlier is even having a meeting where the leader opens up whatever AI you're using. And it is like, I'm going to try something that I haven't tried before. And I'm going to basically do this with you as an audience so that you can see that it's not perfect, that I don't have all the answers and that I'm just testing and iterating with the use of this tool. So I think we're not very good at practicing what we preach. We tell other people, you know, you should definitely experiment and you should, you know, go off and kind of just mess with it, right? But then we're like, well, why didn't you finish this in time. So So making sure that you're actually showing that this is how we should do things, identifying on workflow that you can work on together where the whole point is just to test how you could make it better, rather than saying the workflow needs to be better. So it's about the process and it's about how we work with each other. And as you said, right, it's about including people making them feel like you actually want their brain still and that you're not just asking them to outsource their judgment to a machine, which is not what we want at all. So I think a lot of it role modeling, showing actually including processes a lot of psychological safety is a sense. It's like this feeling that I take for granted that people want to hear my opinion. I take for granted that it's not a risk to speak up. And so having that kind of culture where you're asking people, where you have a meeting, where the whole point is to say, let's talk about a mistake that you made with AI, right? I accidentally didn't read the email to the client that AI created and that shouldn't have done that, right? I learned from it though, right? So the whole point of this is that you're learning that you're getting better. And it's the same thing as it's always been. It's just that we've skipped that whole step when it came to AI because it seemed like AI would fill the gap. And that's been the mistake, right? It's thinking that we didn't need the same basic foundations of learning and iterating and experimenting and having radical candor. And that means being honest about when something didn't work because you used the AI the wrong way, right? And you're learning about this and you're sharing those things and you have to role model that as leaders and you have to implement norms and meeting structures and new ways of doing things that actually include the experimentation within the structure. - So you open up there with a leader saying, hey, this is how I use AI or how I'm experimenting with it or whatever, right? So just showing that they're in the game with the employees, right? They're not dictating it. I'm a big believer, you know, at least I'm a player coach, right? I have a research team, but I'm also a researcher. And I was a very early adopter of screw-in-round with AI and see how it could use as to do things and then sharing that. And I think that that helped and in hindsight, I mean, really is how important that actually was. And there was a study that just came out and I think last week I forget where I read it that corporations that are having more success with AI, that that initiative's quote owned by the CEO, right? It's not an AI office, it's not an AI guru, you know, it's the CEO directly says, I am going to stew with this. And my gut is that the most success within that are CEOs that are not just dictating, thou shalt use AI, they're probably in there, understanding the use cases, really, you know, empathizing, really, they have rolled up their sleeves and they're getting their hands dirty. I bet that is the example for the reasons you said because employees can say, okay, they feel the pain, they understand, it's working, it's not working, they're part, again, of co-creating with us and there's not this gap. That I think with this technology more than ever, that type of approach is going to prove way more successful than the top down from on high, thou shalt type of approach for sure, yeah. Yeah, it's great. You know, it's so interesting as well because Amy Edmondson is also directly applying her work to the AI world and has written lots about it in the last year. And she has this term she calls trust ambiguity where it's basically that people have the sense that they're supposed to trust the AI, but also not fully. And then, but they can't talk about the mistrust, right? So then they stop voicing their concerns and then they're not catching errors and that either leads to the detraction or it leads to the AI slot, right? They're just using it for everything and people are getting just also, you know, responses, that slows everyone down because it's actually not good. You need a human in the loop. And so this sort of ambiguity of like, you want me to be vulnerable, but I don't think I can be vulnerable about this thing because there were companies that came out just banning use of AI, which is also those things where even now looking back or going, well, that was a little bit silly. Exactly. A little dramatic kill. You should use it at all. And so there's still people who are going well, I'm not really sure, and we're dealing with, you know, confidential information. And we have a lot of client information. I'm not really sure how to use it. And our in-house one is really terrible. So all of these kinds of ambiguous, what am I supposed to do with this is also one of the things that a CEO who goes, "Hey, let me show you what I'm doing," which gives them permission to do the same thing. It's really just like, as you said, hits the nail on that. - Yeah, well, that scenario of companies saying, we don't trust it. We've been told you they're not to use it or we're just, we're not going to, 'cause we try to use it and we got better results. I don't hear that as much, but I heard that in early days. Sometimes I still stumble across that with a company. And again, my position on that is, that is such a losing strategy. I mean, for all the things we're talking about around the challenges of rolling out AI and building the trust and doing an admin. I mean, there's a lot of stuff that has to be done there, but the strategy of simply crossing your arms and saying, we don't trust it yet is beyond, there's no doubt in the history. That is a losing strategy for MLK, what you're doing for a living, I don't care what your company does. But it is amazing, you can still see that, right? Because they just don't know what to do. I want to shift gears here. You have this three circle diagram. The attributes that have to be in place from leadership for people that actually want to follow them on this journey. And I love this diagram. So it put it on the table here. Should explain those circles and why they're so important. Because again, well, you and I are discussing is really, how do you get people to lean in and want to be part of it and not resist this? - Yeah, absolutely. Well, it's a great question to have now because I think it really brings together what we've been talking about. The idea is that we created this sort of trifecta of the future ready leader. It's a thin diagram where these three competencies don't just have to exist in isolation, but they have to actually exist together signultaneously. The first is AI fluency, right? It seems really simple, but it's obviously, it's more than prompting. It's yes, knowing how to prompt effectively, but it's also about strategic delegation and actually rigorously evaluating outputs and using the technology responsibly and knowing that it's not about getting answers, but what questions you're asking. And that takes practice. It does take upskilling and it's necessary, right? You have to be able to do that. The other one is strategic agility. So this is sort of your business acumen, your forward exponential thinking. It's being able to have sound judgment through the fog, through the uncertainty, really knowing and being able to think about where the real opportunities are and adapting those with confidence when everything changes. And then the last one is relational intelligence, which is a lot of what we've been talking about. It's a people part, right? It's this genuine willingness to understand what your people are experiencing to not scoff at the fear, right? To actually understand these threats and then actively create that environment. The one that is psychologically safe where people are speaking honestly, they're taking risks, they're telling you when they've used AI and how they've used it, they're admitting their mistakes with it. And you're really building that kind of trust that is what makes people choose to converse you. And I think what a lot of people miss is that if you're missing one of them, then the whole thing falls apart in a pretty predictable way, right? Like, here's AI, go make magical things happen. So they have vision. They probably have the technological understanding, but they're missing that last piece, right? They're announcing, they're not consulting as we talked about, they're sort of optimizing, they're not including, and their team is really feeling drive-to-long. So, and you can imagine the problems when you're missing the other ones as well. So I think this is the leader that knows how to drive the car, right, the fluency. They know where they're going, the strategy. And they've made sure that there's not just room in the car, but there's a reason for everyone else to get in and actually help them get there. So you have to have all three or you're gonna go off-roading, right? - So let me play that back because it is a tough time to be a leader, you know, the technology industry for sure was, you know, what I focused on, but probably any company. Actually, for the very reasons you put on the table there. So you should think about it and let's say, if I'm a vice president and an executive vice president, now I'm in the C-suite. I've been working for 20, 25, 30 years, right? I didn't grow up with AI. But now, in this current environment, I'm being asked to be an executive, and I agree with the attributes that has these three things. Number one, I can speak AI. I've really clicked in, I'm fluent on it, I understand it. I'm just not reading words or again, I'm just not dictating something I don't really understand. So that's number one. So that's a curve for a bunch of leaders out there. Number two, I have to have some credible strategy around what this means to our business. And again, that's so fluid right now, that's super hard. And in number three, I have to have this, you know, sort of emotional capability here to understand that this is a technology that is, unlike any other technology I've ever put out, and that my employees are not going to feel safe, but that I've got to create an environment here that I've probably, again, in my entire career, I never had to worry about it at that level. So if you put all that on the table, that bar is high. I mean, I do not envy the executive up there. Having said that, that is the bar, because I agree with what you're saying. If you want to be a leader in 2026, in 2030, whatever, in this, you know, we are in the AI era, whether you like it or not, then you are going to have to be fluid with AI. Number one, you are going to have to start to figure out, and again, I'm co-creating that strategy with your employees is a good thing talking about, being transparent, all the things we talked about, but you are going to have to evolve the strategy of your company. And then you are going to, I think, have to raise, You know, you're. or understanding of how you take people on this journey. I think all three of those notes are going to be critical, but I think we do have to pause and acknowledge. It's a high bar, and I'm curious, 'cause this is what you do for living. You help management teams think about this. I'm sure you go in and you meet executive teams and they're probably zero and all three of those are right. So I'll talk a little bit about how you help them on that journey to say, look, here's where you are. This is where you eventually need to get to. If you really want to be an effective leader in today's environment, talk a little bit about how you take them on that journey. - Yeah, that's such a good question. I mean, a lot of what I start out by doing is, where are they, right? Are they really zero on all of them? Where are their strengths? And where are the gaps? And of course, it's complex within each of these bubbles, right? AI fluency is so much more as we talked about them prompting. So it'd be they're really good at using cool pilot to write their emails, but they're not using it in a strategic way. Or maybe they have really, really strong strategic foresight, but I think as I said in the keynote, they're running a one-man relay race where they're trying to pass on the baton to a non-existent group of people because they haven't brought them with them, right? So I think my experience is that people are worst at the relational part, which is so funny to me. It's ironic because a leader is a relational role, right? The whole definition is in relationship to someone else. And when you ask people who is the best leader you've ever had and why, they never say they were strategic or they were really good at their job. They say they listened to me, right? They heard what I had to say, they built me up, they guided me, right? And that's because such a big part of it is relational. Now with that being said, I mean, what we do is of course, we have experts who come in and help with each of these components, right? You have the Harvard professors who know exactly what to do and how to teach disruption and strategy and then you have the AI people. And I am more on the relational side, but I think what I do is I try to help them attach these to each other, right? What does it mean to teach AI fluency and to communicate that to your team with the relational perspective, which is just like I've been talking about today, is first acknowledging that this is actually not a technology problem. This is a, how can I get you even on board to being willing to learn this skill? That's a relational problem. And the strategy as well, right? People are not going to follow somebody who has no strategic foresight, right? Again, it's the trifecta. You can't just be good at their relations and not having there the other two, or you have to be able to know where you're going and people are not going to think it's nice to follow somebody who seems like they don't have their head on their shoulders. So thinking about how they're related to each other, I think is one of the things that is my specialty and the sense of connecting the dots and making people realize that these things are interconnected and in what way that actually translates to Monday morning. - Yeah, yeah. Well, it's interesting on that and there was a question I had a huge here that I think really relates to this. 'Cause again, you talk about these three boxes. In one thing I see specifically in the technology industry, is you do have leaders in these tech companies who are starting to crystallize a vision about what they think AI means to their company, right? So they've got this strategic. They themselves have become very AI savvy, they've fluent, but they a complete dead spot, complete dead spot on the relational part, like you're talking about. In example, I will use, and I hate to pick on companies in a negative way, but I'm going to do it right now, meta, and what you're reading publicly about meta, right? So here's a company that clearly the leadership team is all in on AI, they're thinking about how it's going to change their business model, they're making all these investments. But they're AI fluent for sure, they've got strategy around it, and then they're pushing this out to their workforce, saying we're going to capture your keystrokes. Everything that you read, and again, I'm not in that company, I don't know the real deal, but everything that you read secondhand is that these employees culturally are miserable now. They just feel like they're training their replacements, and I would argue that they're not resigning, 'cause they sort of have the golden handcuffs, right? They're probably super well being paid right now. So, is that a sustainable strategy, right? To have two of those things going, and say, well, I'm not going to worry about their relational stuff, and people don't like it, go somewhere else, 'cause I can just hire somebody else and just bring 'em in here, they're interchangeable. When's the bill come due there? What's your perspective on? - So we all know that fear can be a motivator, but it's very short term, and it will keep people on their seats for a while, especially if leaving feels brisky, but what it doesn't do is it doesn't produce the things that AI success actually requires, right? It doesn't produce the kind of honesty when something doesn't look right. It doesn't produce a willingness to experiment publicly. It doesn't produce motivation, right? It doesn't produce this sort of discretionary effort that it takes that sort of at a bare minimum, so you get people who are just a drain. They don't want to leave, because it's a good place to be, but they're not helping you succeed in the way that you need them to, and I think AI is a really potent example of this in a way that may be passed as we've talked about technologies, haven't really been. So the fear is destroying the conditions that actually create the success that you need. So that's the gap between I think staying in the seat and actually contributing from where you're staying, and I would say that maybe the golden handcuffs are solving an exit problem, but definitely not a value problem. - Yeah, I mean, I completely avoid that, because again, as we've already discussed, I mean, again, just AI efficiency, in terms of AI efficiencies, and then there's AI opportunity. Now, opportunity is all the things that you could do that you never realized you could do, as a company, terms of value creation, et cetera. And those employees for that environment are not going to be about AI opportunity. They're not going to be contributing. They're just going to be doing the bare minimum they have to do, not to lose their job or whatever. And I'm watching this unfold. We went through the pandemic, and we went through a point in time in tech, where there just wasn't enough people to go around. Everybody was hiring employees that put all the power, right? And then with AI, it's swung back really hard, and you're watching a lot of these companies act, like employees are completely expendable, and it's not a problem. If you don't like it, again, go somewhere else. We don't care. This is the way it's going to be, right? And you watch that pendulum go back and forth, and the reality is that the truth is right in the middle. Employees need companies. There's no doubt about it. Like companies need quality employees. And there's never enough quality employees to go around. So as soon as companies start with that mentality, I predict something like a meta, they are going to bleed their best talent. Their most creative best talent is not going to stay there. If the culture is truly being reported, they're just going to say, look, life's too short, and I'm employed, and they're going to go. And then what does that brain drain mean to them in terms of future competitiveness? I think a lot of companies leadership teams are not doing the proper calculus right now on this. I really deeply believe that. I think they just swung it too hard, and they've got to get back to the middle. And employees should appreciate the opportunities. But companies have to realize, you know, I don't have the right people. I'm not going to be super competitive. That's just the real thing. And also knowing who is the right people, right, as we've been talking about this uber focus on AI skills. I think we talked about this in our conversation, right, is that in a year or two, that's going to be like saying, I'm really proficient at the internet, or at my iPhone, right? That's going to be table stakes. And what the differentiator is, the human judgment, right? It's knowing what problem is actually worth solving, right? What questions to ask? And recognizing crap, AI generated output in the irony is that the better AI gets at generating those answers, the more valuable it's going to become to have people who know when not to trust those answers, right? And the other thing is that judgment is actually not an individual capability. I know we think it is, but it's something that we express socially. It requires you to do it externally and with other people, and that is an interpersonal risk. So if you don't feel safe on a social level, on a cultural level, then that judgment is not going to be something that you're really exercising. And it's funny because I think this example of the metas and the Elon Musk's and the Zuckerbergs is, for decades, organizations have created value by rewarding people for having answers, right? And we know that the organizations that are going to succeed are creating value by rewarding people for asking better questions, for account assumptions and spotting risks and knowing when it's wrong. And those things are dependent on curiosity and candor and the willingness to speak up, right? So I don't think the question is whether fear can build a successful company because sometimes it clearly can. The question is whether fear creates the kind of human thinking that organizations actually meet in an AI era. And I think the answer to that question is no, because we know that the companies that pull ahead are not going to be the ones that have the most AI. They're going to be the ones who have the people who know how to think about it and challenge it and create something that's actually to be able to be valuable. - Yeah, I agree, 100% with that. And again, I think companies have to find their way there to appreciate that middle ground of the balance of technology and all the people around it. So I'm watching the clock here. This has been a great conversation. I loved it, it was really good catching up with you. I think you're on a really good square. And again, the keynote that you did that caught my eye really resonated with me because companies, I think are missing the mark right here right now in terms of their rollout, and enablement, and building, all the things we've tied up, building the trust, and co-creating with their employees, et cetera. But for companies to get this right, I mean, it's going to be phenomenal. I'm very bullish on what this AI era is going to look like for employees and their creativity, and all that kind of fun stuff. But we've got to stick the landing. - Yeah. - How we bring employees along the way. And again, I think you're one of us. work is super relevant there. So I appreciate you coming in and sharing your perspective. And I always end these sessions with a question of the day. And everybody's focused on the first half of AI, worried about efficiencies, et cetera. But it is both half of the sentences here. And so I want to challenge our leadership teams. Are you really thinking about not just rolling AI out, but bringing the employees along the way? Here's everybody.