***DISCLAIMER: Please note that the following AI-generated transcript may not be 100% accurate and could contain misspellings or errors.***

[00:00:00] Mahan Tavakoli: Welcome to Partnering Leadership. I am thrilled to share with you my conversation with Conor Grennan on generative AI. As Dean of Students of NYU Stern and co host of the AI Applied Podcast, Conor brings tremendous business and AI insights into our conversation, where we talk about strategies for integrating AI through change management.

capability building, and ethical implementation. He makes the case that AI should be a priority given the many opportunities around workflows, and I totally agree with him. He also talks about employee effectiveness and value that generative AI can bring. Connor shares practical frameworks you can start applying.

Underpinned by compelling company examples. So I learned a lot from Connor. I'm sure you will as well. I also love hearing from you. Keep your comments coming. MahanadMahantavikoli. com There is also a microphone icon on PartneringLeadership. com Really enjoy getting those voice messages. Don't forget to follow the podcast on your favorite platform.

And leave a rating and review when you get a chance that will help more people find and benefit from these conversations. Now, here's my conversation with Connor Grennan. 

Connor, welcome to Partnering Leadership. I'm thrilled to have you in this conversation with me. So 

[00:01:32] Conor Grennan(: excited to be here. Great to see you, 

[00:01:34] Mahan Tavakoli: Connor. I am so appreciative of all the great work you are doing on generative ai, . Including the outstanding new podcast.

You started AI applied podcast, but before we get to talking about AI and generative AI would love to know about you, Connor, whereabouts did you grow up and how did your upbringing impacted who you've 

[00:01:54] Conor Grennan(: become? I love that question. So I grew up in Poughkeepsie, New York, which is where Vassar colleges in the United States in New York.

And then I was in Jersey. After that for a little while. So I had an exposure to suburbia and to extreme diversity. Then I went to university of Virginia, but then the important part was that I graduated from UVA and I moved abroad just seeking an adventure, not really having a job.

And I ended up going to Prague just finding a job in public policy, which is what I really wanted to do. I was there for six and a half years. Then I moved to Brussels for a year and a half to do the same kind of work. Then I went around the world by myself for a year. Then I found myself in Nepal and I lived in Nepal for like maybe two years and started a nonprofit there.

And that led to an organization I started and also a book that I wrote. And my wife's American, but I met her out there. So I was very blessed to have exposure to a ton of different cultures and experiences, I think. 

[00:02:48] Mahan Tavakoli: You do have a fascinating background and I encourage the audience to read up on it and your book as well.

We could spend hours just talking about that, but you have focused a lot of energy over the past year on generative AI, Connor, what got you started on that? Yeah, you 

[00:03:07] Conor Grennan(: know, I think that's such a good question because it's pretty new. And I feel really strongly about this aspect, is that I am not a machine learning, expert.

I'm not an AI expert. I'm not even a tech expert. And I really want to tell the audience that because I hope that it lowers the barrier to entry for people. So the way that I got here was. Back in probably February, as ChatGPT was released at the end of November last year. And, my wife actually does AI work for McKinsey and Company, the consulting firm.

And she saw ChatGPT coming out and she's Hey, you should really check this out. So I checked into it probably in January or something like that. And right away, I was like, Whoa, this is crazy. I have two kids. I have a 14 year old and a 12 year old. So my first thought was, this is going to really disrupt education in a very meaningful way.

And then my second thought was to my job, which was I'm Dean of MBA students at NYU Stern School of Business. And I thought, okay, so these young people or young leaders have to know this, and I don't know how prevalent it is anywhere. So I ended up creating a framework around. That to teach them which then morphed into teaching other companies and things like that.

But I didn't get into it because had been tracking AI for the last 10 years that's really important because again I want people to understand that you don't need some kind of special background. 

[00:04:24] Mahan Tavakoli: That's why I appreciate your content Conor, it's understanding the applications in whether it's education or in business, rather than sometimes the people who are well versed in AI are speaking machine language that doesn't translate to the rest of the world.

Now I'm with you that generative AI is going to be. transformative in all aspects of our lives. But we'd love to know how do you think it will reshape organizations and organizational leadership 

[00:04:53] Conor Grennan(: I like where you're going with this because everybody should be aware that first of all, this is not really digital transformation. It's change management. So why do we say that? Because a lot of changes in organization when we say, Oh, how will this change the organization?

It may be because there's a some kind of brand new technology like a sales force or something like that might redefine how a department works or how departments interact with each other. This is not that this is wildly different. This is so you. And it augments the capabilities of every single person, no matter what their job function that I believe it's going to have a real dramatic impact on the organization.

And I've done some stuff around this and my own theory. This is just a theory. But my belief is that it's going to really flatten the organization for this reason, when you see how this thing works, I never want to use the word automate because it's not exactly true, but it comes very close to automating.

Routine tasks and administrative tasks. So you can imagine just in a very simple example, writing an email or something like that. What that means for an organization is that you really have to rethink how you're hiring. Because if in the past, what we've always done the past meeting, like a few months ago entry level positions are by nature administrative because there's administrative work to be done.

The work that flows up is at a certain level. Do that, rooted in the skill level of the people in the entry level. And that goes up to what you might call a middle management strata. And then that work product gets taken and improved. Whatever your work product happens to be. This really changes that because with generative AI acting as a co pilot, we can talk all about this.

Of course, what it means is that the people you want to hire just need to have less administrative skills, which are. Going to go away anyway, to a certain extent and much more critical thinking skills, because what that will do is allow them to interact with this kind of copilot and really feed a much higher work product and more refined work product up to middle management.

And that's why I think it probably squeezes middle management to a certain extent. Middle management turns into more and this behind I hear you talking about this all the time. This is why I love your podcast. And why we connected, I think, because I love how you talk about what do the different layers of management actually do?

What does leadership actually do? And I think it's getting closer to what. I think is your vision a little bit from what I know and what I've heard from you speak, which is that middle management, senior management operate at a much higher level when they are, mentoring, when they're guiding, when they are, really drawing out the best in their people.

And I think that this is going to give that a real opportunity so that middle management turns into more guiding leadership. Vetting work product. And then by the time work product gets to senior leadership, it's almost done. And then senior leadership can really, and this is what I've heard on your podcast a lot.

Senior leadership can really focus on, culture, direction, business differentiation, things like that. So that's how I see the change in the organizational structure. 

[00:07:42] Mahan Tavakoli: It's a beautiful vision of the future and couldn't agree with you more therefore, the requirement for leadership to focus on those human elements, those culture elements 

I want to touch on a couple of things you mentioned there, Connor. One is that this is more of a change management initiative than anything else. I totally agree with you on that, but one of the challenges I see is in many instances change management organizations happens because the CEO or the board are the ones that see a need for that change.

I'm finding a lot of CEOs and boards are somewhat concerned about where things are headed. Don't fully grasp the scope of it to be able to lead this change management. How do you advise clients? On thinking through this change management.

[00:08:33] Conor Grennan(: wHen I work with organizations and I do trainings for organizations to integrate generative AI into their workflow the evolution of how I've done that has changed since, I started doing it back in the spring until now, because I've seen, what works and what doesn't.

[00:08:48] Mahan Tavakoli: Connor I wanna break in there to highlight one point I feel the same way myself. You said, The way I approach it has changed from past spring until now. Just think about the fact that you have reinvented your own approach over the past six months. I feel like I've done the same thing.

That by itself is a big statement. 

[00:09:10] Conor Grennan(: , it's such a big statement because first of all, we talk about like the olden days being like February, so there's that, but it's also when I train these companies, it's just as important for people to come away feeling enthusiastic about it.

So two things have to happen when I'm training a company. First of all, the senior leadership has to understand. What the capabilities are and what this actually does and how it's changed management, because, and I say change management instead of digital transformation, because there's nothing to learn.

Any digital transformation requires some kind of learning curve. But this is not French or calculus or Excel, this is like using your iPhone. In fact, it's not even that it's like talking to a human. So that's why it's changed management. And that's why I started at the top.

The reason I started at the top is because. You don't get to the top of your organization or even some kind of senior leadership team unless you understand people and what drives people. And that's the single most important thing you can do. So the reason why I talk, with that level first and really help them understand that first when I do big keynotes to senior leadership, things like that is because they have to understand that there's a whole new set of benchmarks.

Because if I just start with like the team. Some people will use this technology and some people won't. And that's bad because there's no accountability to what the new benchmark is. But then I have to move on to the whole organization because the really important thing is to get people excited about it.

And when people are excited about it, and I can do that fairly quickly now because I've done it. But I know that when people start seeing, oh, this is not just about how I'm going to make my company more money. This is how I'm going to change the way I work. It's how I'm going to be more productive with the purpose of getting promoted in my job.

And it's how I'm going to change my personal life as well, because it has such a big impact. But I work with senior leadership because. That's what senior leadership wants their people to be think they want that enthusiasm. So that's how my model like evolve, which I can talk about, but that's like the general overview of it.

[00:11:09] Mahan Tavakoli: yoU also mentioned Connor, the augmentation that this enables individuals inside the DC Beltway, there are more than 70, 000 attorneys. And the way I've been communicating it to clients is in many instances, attorneys were being hired to review contracts and do those things in training to really contribute.

So if you see this as Automating that, then what does it do with bringing people on board, whether it is for law firms or other organizations, how does it change sourcing and training of talent in organizations as people are augmented with generative AI? 

[00:11:53] Conor Grennan(: Yeah, it changes everything, right? There have been a lot of studies now, really well done studies by Boston Consulting Group and other organizations like that, that have shown that when you are working with lower level folks, who are lower skilled, that will increase the level of competence by maybe 40%. And then even when you're working at the upper level, that increases the competence and speed and quality of work by about 17%. So let's take law firms, for example, 

 With law firms, this is going to be absolutely disruptive, but I would, I will say is that almost nothing more than legal is going to be disrupted by this.

So why is that? Because what this disrupts is actually not jobs. It disrupts tasks, so why do we say that? Jobs are not monolithic things. You don't go to work and do one giant, I'm putting in air quotes, job, right? Your job is made up of series of tasks. So if your job is only one or two tasks, then yes, that will get disrupted.

If your job is to Take giant documents, summarize them, create insights, and then give them to your boss. That will essentially be going away. And I don't like to talk about jobs going away, but that's not a job. That is a task that is going away because this does it within. minutes, you can do something that would otherwise take hours and you can do it much more productively.

You can do it to a much higher degree. You can continue to query it back and forth. You can be speaking to a document at this point, live on your laptop. That's how nuts it is. And so when we think about how this will disrupt it. We have to think about what are the tasks in there and the way that legal firms and law firms are built out.

 My wife is a lawyer a former lawyer Yeah, it's the billable hour and what happens to the billable hour when you can take the same amount of work and reduce that from 60 minutes down to say 14 or 15 minutes the first thing we need to do is understand how you take advantage of that and produce way more value for your clients.

That's what clients want. Clients want more value. So what I do when I work with companies is I show them how do you actually do that? What are you going to do differently the next day after we talk? So that's the big change for me. 

[00:13:51] Mahan Tavakoli: As you mentioned, there's tremendous additional value that comes from it.

Part of what I've also been able to show clients, whether it is just even working off of the minutes of their board meetings and recordings of board meetings there are so many insights that come from it. If the right questions are asked of it, therefore making the questioning even more important.

Now, part of what I find a lot of times people say, okay, I get it. I've played around with generative AI. It's powerful. It's going to transform everything. Now, in addition to change management, what do I do in my organization? Where do I look? How can we tap into the power of generative AI in my organization?

That's 

[00:14:38] Conor Grennan(: the question that everybody should be asking, so I'm glad you're going to that too, let me answer that by telling you how this evolved for me, because I think it's actually pertinent here. So when I first was teaching this to my, MBA students, I went out and reached out to big heads of, asset management firms, hedge funds, people that I knew and said, Hey, how are you using it?

I'm trying to teach our students. They would say we're not using it. I would go over there and I would end up teaching them instead. And after about an hour of teaching C suite, they'd be like, this is going to change everything. This is amazing. Just like you said. But what would happen then about a week or two later, I would check in.

I was like, Hey, how's it all going? They're like we're not really using it very much or it's pretty good. I'm like, okay, so I'm doing something wrong, but I wasn't. sure what I was doing wrong. And so what I ended up doing was taking it back to the kind of the lab of NYU, so to speak and trying to figure out what I was doing wrong.

And what it turned out, I started asking people who knew a bunch about neuroscience and executive coaches, which, Executive coaching is really about the brain and about , neuroscience and what people respond to.

It's very different from consulting, right? Coaching is having people come to their own realizations about things, et cetera. And what my colleagues were saying around this, were they like listen, Connor let's say you were trying to get two people in shape and when you gave a state of the art gym to, and the other gave a personal trainer.

Why is a personal trainer going to get you in shape more? The reason is that it's not actually about the tool. The tool doesn't matter. The behavior matters. If I buy a treadmill and I'm not in shape in a year, it is not the fault. I don't want to try to repair the treadmill. You know what I mean?

Like it's all about behavior. So I started to realize that my framework, my original framework from back in the spring of teaching would get people excited. And people would come out after an hour being like, this is great, but it wouldn't stick. And it wouldn't do your question. It wouldn't transform their organization.

So I started thinking what will transform their organization? So I changed my framework to teach. How do you teach in the flow of work? What do you actually have to do? And then bringing the tool into that, because people do know what their problems are. They know what drives value for their business.

They know what the obstacles are, the challenges are. They know it takes up a lot of time. And then if you focus on that, instead of focusing on the tool and then focusing on how well does the tool apply to each one of these discrete tasks? That's where you can change it. When we say how does it change business, if I'll go into, a very large healthcare company and the first question is how is this going to impact healthcare?

What I used to do was say, oh here are the 10 use cases which seems like a great way to do it, right? I'd say like use cases and that sounds brilliant. It was not. So what it turned out being was now I go in and say you're going to tell me what the 10 use cases are because when I go in and tell the use cases, they're like, great.

I'm like, wait no, I'm not a doctor. What I'm just telling you what I've heard, but when I say you tell me what it's going to be, then they're like what we really need to solve is this and this isn't okay, let me show you how this applies to do that. That's what the training is all about.

It's all about being in the flow of work. So that's why it's different for every organization. 

[00:17:24] Mahan Tavakoli: I love your humility your growth mindset and willingness to continually learn, which I think is going to be even more important in the AI age. So you're saying I tested something, people were nodding, they were agreeing, but they weren't doing anything as a result.

Therefore, I adjusted and I changed. And I think we need to do more of that. For ourselves and in our organizations. Additionally, what I heard from you is that it's the business issues that you're having and seeing how can generative AI help solve those business issues rather than the use of generative AI, just for the sake of it.

[00:18:00] Conor Grennan(: Yeah, and it's so tempting because it's such a cool product. When I do a keynote or a training or something like that, I go through these demos and I show people, but I try to keep it sort of high level. And then I tailor it to the organization I'm working with so they can see, but you have to start high level in the same way that if you're trying to get into shape and exercise is such a good analogy here.

You can't just say go out and run, right? Because what you don't know is that it's hard to get the willpower. So what do you do? You have to have a system whereby you get up in the morning. The first thing you do is put on running shoes or you get rid of the Doritos in your house, or you at least don't have them on the counter.

There's certain things that you have to do. And why? Because your brain is the thing that you have to change, not really your body. And when your Brain gets these triggers prompts and things like that. That's what does this. And it's the same with this, right? You have to have the dopamine hitting your brain.

And when people see what it can do, and then they can envision, Oh, now I see because it can do this task or whatever. I can now see how I could save an hour of my day. That moment has to come, and it doesn't come if you're just saying, look at this wow factor, look at this wow factor, and instead you have to say tell me what your day is like.

, and with that All of a sudden, if you can take back some hours in the day those are all the things that you can get people excited about.

[00:19:12] Mahan Tavakoli: As people are getting excited about this, Connor, this past week, I was having a conversation with a CEO client. And she said, every time I bring it up, my chief legal officer says. We don't know the legal ramifications of people using this, so on and so forth. So what is your advice on that front before we get into how to experiment?

Because in many instances, I've seen people who have heard. Negative stories, whether it's from data, safety, responsible use of AI, which has caused them to be cautious or not initiated within their organizations. 

[00:19:50] Conor Grennan(: Yeah. I'm glad you bring it up because a lot of people sometimes hear all this.

 But so they're only like half listening. Cause they're like I can't do it in my organization. So what's the point. I want to try to reassure people here on this in a couple of ways. So number one, Of course, caveat all this with, you should not put proprietary data into chatGBT.

You don't even have to know why. But let's explore a little bit about what that means, right? The big reason not to do it is not because if you put your spreadsheet you work at J. P. Morgan and you upload a spreadsheet. It is not the same as like leaving that spreadsheet with your company secrets on a table in Starbucks and somebody else.

It doesn't work like that. It's trained on language and how language fits together and things like that. And even that you can turn off the training of it, but more importantly, people need to know that first of all, if you've already uploaded proprietary debt, it's fine, nothing's wrong.

You know what I mean? Like it doesn't do anything. Nothing's out there. It's like pouring a Coke into a swimming pool and expecting the swimming pool to taste like Coke. It just doesn't work like that. So that's number one. Number two, the things you really have to be careful about are if you're in a regulated industry, right?

And the reason for that is that you do not want to violate third party agreements. You don't want to violate legal agreements, things like that. That's a real thing. You cannot share. This is sharing it, right? So now having said all that, the danger is actually not, I think, what people think. When they hear about Samsung uploaded the data, nothing happened with that Samsung data.

It's just Made a lot of waves in the spring, because they uploaded their secret. It didn't really, yes, maybe somebody on OpenAI saw it. If they looked at that document out of a hundred trillion people, and they believed that those people really were from Samsung, do you know what I'm saying?

It's not like that. So here's what is frustrating to me is that even in like hedge funds and places where it's the proprietary data is what they make billions on. They ban it on their office computers, but then everybody's using it on their mobile devices. And so this is what I, implore you because you are so good at talking about this.

And I've heard you, you haven't said maybe literally these words, but do you trust your people or do you not? And what does that look like? And so if you trust your people, then what you do is you give them guardrails and you say, this is what proprietary data is. Don't. Upload this data because they already have restrictions on what they could, upload or bring out of the building or share 

they already know how that looks. Companies that make their living on risk, like they walk right up to that electric fence, they live on the edge, all that kind of stuff, have all of a sudden turned into companies that see risk in black and white, and it's there's either risk or there's no risk, and that is not how companies get ahead.

 So I would say it's probably safer than you think. Talk with your legal team and understand what the actual risks are. 

[00:22:28] Mahan Tavakoli: As you said, set some boundaries and within those boundaries, let people experiment and use it because first of all, they are using it anyway.

And the power that it brings to the team, the insights that it brings and the capabilities it brings falling behind is a bigger risk so let's say we set some guidelines and boundaries in my team and organization, then what is it you set up a committee to decide what are functions that lend themselves to generative AI use?

Or does each person start using generative AI? What are best approaches with making sure that it's incorporated into the systems and approaches in knowledge working to organization? I would 

[00:23:17] Conor Grennan(: say that number one, and this may feel like a luxury to people, but I'm going to throw it out first, is get training on it,

because it is not as intuitive as you think. The problem that I've seen a lot is that organizations will do a couple of things. Number one, they'll say hey, this person really knows how to use it. Let's have that person teach everybody else. That's not really how it works. Only because, you can't get somebody who plays guitar really well and say hey, teach me.

That's why you need a guitar teacher. Somebody who understands like getting into the mindset of the people who don't know how to do it. It's not an expertise level thing. It's a, do you understand how to convey knowledge thing, right? So that's why I really like your podcast because you focus on not, this is amazing, but how do I convey this knowledge?

So find somebody like find somebody you trust. Like somebody who does this kind of work and say, Hey, like. How would this work? So that's number one. Number two. I do agree with putting together a committee or having a few people dedicated to it just to make sure that they are really going deep and understanding and then sharing what they've learned.

It's a little different from training, obviously, but even just sharing and saying, Hey, listen, and I don't believe in really prompt engineering. I think it's overblown for the reason that first of all, I think it's probably going to go away eventually because the A. I will know better than you what you need.

But second of all, I think because I'm very anti putting any kind of barrier to entry to anything, right? So when people like, Oh yeah, you got to do this, but you have to be able to prompt that will knock 75 percent of people off their bicycle. Do you know what I mean? They'll be like I'm never going to learn how to prompt.

And my argument is all you need to do is talk to it like a human, but that's hard. That's hard because the brain does not let you do that because your brain sees this computer screen and the same way. It's hard to talk to a baby like a college professor all day. It's gonna be hard to talk to your screen like a human.

You have to get used to it. So then people say just practice. And what I say is yes, but practice what? If you want to learn how to play volleyball, just practice. But practice? What, how do I do it? So I don't believe in the quote unquote, just practice.

What I believe is a systemized approach where you have people who are really dedicated to it, understanding what the new technologies are and understanding your business uses understanding what drives value for your organization? What are your stumbling blocks and checking out and just being responsible for just.

Messing around and just being like, Hey, I wonder if this will work. I wonder if this will work. I don't know if this, and asking that over and over again, and then sharing what they have learned with senior management, 

[00:25:36] Mahan Tavakoli: what you just mentioned there, Connor, in that the training is essential. It's one of those things that I find is lacking in so many organizations.

And you mentioned a couple of points that I was smiling because I have heard from you, but I would imagine the audience is not as familiar with. Because there's so much evolving in AI. You said, talk to it like human. One of your most recent episodes of AI applied podcast, you had a conversation around that, that why these generative AI, whether it's chat, GPT, or Claude, in many instances, respond a lot better to emotional conversations.

Whatever it is, that might sound like a simple thing, but it's not something that people by themselves would find easily. So there is need for training. It's not prompt engineering, but it's understanding how the generative AI tools work. 

[00:26:38] Conor Grennan(: That's exactly right. I send people down the path of going over the top.

So when I demo this, I don't just talk to it like a human. I literally talk to it like a friend. So I'll literally be typing Hey man, I don't even understand how string theory works. You got to help me. Yeah. I would say literally that not, can you explain And why do I do that?

I do that because it, Starts to trick my brain into thinking this is a human. It speaks so much like a human that you cannot detect whether you are speaking to a human or not. They've proven this over and over again. And when you speak to it like a human, it responds like a human. And the reason we don't is that we're so used to this command and response model of Google.

You have to put something in, you have to search for just the right search terms. Everybody knows how to do that. And then you get something back and then you take the link and you do your own research. This is the opposite of that, right? If you just ask what you're looking for. So what I'll do sometimes is I'll be sitting with somebody and they'll say what I really want to do is figure out like a new performance management system for my, asset management firmware.

And by the way, I say those, but like I do every single industry pretty much, I think at this point, but I start typing that in there. No, I just want to tell you first. I'm like no. Think of this as our third person here. And you'll see just, I'm going to literally write down everything you're saying.

And it could be like, Yeah, I don't know. Maybe like performance management or something like that, but I'm not totally sure because write those words into chat GPT there's just some little fun things that are happening of late, which is they're finding that if you say.

Hey, I really need your help really badly. It responds with a higher quality stuff. If you say, Hey, my career depends on this. It responds to the higher quality thing. And my colleague on AI applied podcast, Jaden has a great theory on this. Because it draws so much from Reddit and everything else.

People probably got better and more fulfilling responses. If they're like, guys help, I need to figure this out. My career depends on it. And everybody in the community jumps in. And so it's like a baby learning how to speak. It's oh when somebody says that I really better do a good job and I really better give a better answer.

So that's, what's so fascinating to me. 

[00:28:39] Mahan Tavakoli: It is fascinating, but that also goes back to the point you were making for the need for some training. In that you set the guardrails in order to allow people to experiment and then have some training gets people to take advantage of opportunities like this. As you said, emotional conversation is not what we are used to when we are interacting, whether with Google or with computers.

But in this instance, whatever the rationale behind it, whether the Reddit threads or other reasons, it performs much better. Now within the organization. Connor, are there certain functions that lend themselves more? So the CEO and the executive team have set some parameters and boundaries, encourage people to use it, have trained them.

Should there be particular focus in certain parts of the organization that can most benefit from generative AI? 

[00:29:37] Conor Grennan(: This is a really good question. I think every organization is different because every organization, is driven by personality. Like you coach and work with a lot of CEOs.

And I know that because I've heard you talk about it, like every CEO has their own style. Some are very hands on. That caveat aside, I think if you're really generalizing, I think the C suite needs to first understand the capability and that is part of a training.

 I always start with that. I'm like, look, this is what it can do. Because what we come out with that is not, I'm going to use this every day, although they tend to, it's more oh we were expecting eight hours of work to look like this. Now we understand that's more like.

Five and a half to six hours of work, and now there's a new expectation for our people. It might make certain positions redundant. I don't like to talk about that, but it's not positions. It's just tasks. Redundant is probably a better way to say it. So that's why the top level needs to know.

I think the second line down whether it's the head of, HR, but people who have a lot of direct reports. These are the folks that really have to understand what the jobs under them look like. And what it would look like with that.

That's why I always encourage people say should I bring my whole team? I'm like, yeah, bring your whole team because people get really excited when they see it. It's much different from, hey, we just had a big training. Now you guys have to learn how to use that. The whole thing about the power of this is that the dopamine hits everybody once they get this kind of training, because it's not just about, Yeah.

How I'm going to make more money for my company. It's how am I going to change my life? How am I going to get promoted? Everything like that. So I would say that C suite has to understand that this is a priority and what it can do under that whatever kind of like VP or management role, whatever that role is, has to really understand how this impacts jobs, and then the sort of if we just call it like the kind of the third layer of people actually executing on the work.

Need to know so they can carry it out, so at the top, they need to know the line managers need to have an understanding of how to impacts work. And then the next level, the employee quote unquote level needs to see it in practice so they can get excited about it. 

[00:31:31] Mahan Tavakoli: Now in. Needing to know Connor part of the challenge that I've had, and I see you running even faster than me is the generative AI is evolving by. They, drastically, almost every week there are announcements that boggle the mind with the capabilities. So how can executives and leaders keep up with enough of the changes in order to be able to take advantage of those opportunities in their organizations?

[00:32:05] Conor Grennan(: Yeah, it's hard. You know what I mean? Like it's hard for me. I spent a lot of time on it. So the first question is, do we need to feel like we do? So the reason why, we started this AI applied podcast, for example, is my big reason was because I didn't want people to feel like this AI FOMO that everybody feels like, right?

You see all this news whizzing by you and you're like, what was that? Oh, what was that? And you feel like, why wasn't I on that train? And why didn't I catch that? So first of all, everybody can take a breath. If you understand how to use chat GPT, you're doing great. You know what I mean?

Like you're doing great. I promise you. So don't feel like you're way behind. Now the opportunity on the table is huge and there are other competitors using it better. So I do want to say that. However, in terms of the brand new technology, you'll catch up, so I really recommend you or people who are like really excited about AI find the people that they follow online.

So I post on LinkedIn every day about this, but there's other great people. There's Allie Miller. There's Ethan Malik. There's like a ton of really great people out there. Paul Reitzer, who has another podcast as well. There's folks just doing great stuff. There's great podcasts out there.

Jaden Schafer has another really great podcast calls AI chat, which is like 10 minutes a day telling you what you need to know. And that's the kind of thing where it just like alerts you to what's happening. So maybe, this open AI dev day, developers day that happened earlier this week.

Everybody's buzzing or everybody in my bubble is buzzing about that. Cause there's new thing called GPTs, which are almost like a precursor to autonomous agents, what that means is that it's not just helping you brainstorm and figure things out. It would actually carry out things.

So it's not just Hey, help me find flights. It would then book you the flight and book you the hotel. Now that's not something people have to worry about right now. Cause it's not really available right now. What is available though, are these GPTs on open AI, which is the pro account for chat, GBT 20 a month, highly recommended. But what those are like customized bots, you can now say Hey, I want to create this, my own little chat, GBT. Based on my own data and you know my own HR function or something like that So those things are out there, but they're so brand new that if you don't know about them, you're not behind 

there's other things like, humane came out november 9th with their new wearable pin. That's a new tech thing. But I don't want people to feel FOMO. I want you to think about what are your business use cases, get your head around chat GBT and other large language models, and you're going to be doing great.

And then follow me and others, I try to post every day about what's going on and how to help people understand what's going on. But there's other great people doing that as well. 

[00:34:32] Mahan Tavakoli: There is a lot happening. And the reason I'm thrilled to be having this conversation with you, Conor, is that you are one of the people I follow and I recommend for people to follow 

jaden has a great podcast on his own Paul Ritzer. I've had a conversation with him for the podcast as well. And Ethan Malnick is outstanding, especially on the education impact of it.

Before getting your thoughts on education, which I also think will be truly transformed, would love to know what are the pitfalls that you've seen or the challenges you've seen. With either potential clients or others as they have tried to experiment with generative AI in their organizations. So our listeners and the CEOs can stay away from those pitfalls.

[00:35:18] Conor Grennan(: I Think the big pitfall is obviously you want to stay away from uploading things that you're not legally allowed to upload. So that's number one. Let's just put that right out there. But I don't want to harp on that because as long as you just know that you're fine. There's not much more that you need to know rather than can I share this with a third party where I say this on the corner of the streets of New York, whatever, right?

So I would say to be totally honest, the pitfalls these days is not seizing on an enormous value proposition that is literally on everybody's laptop right now. And then what you don't know is which one of your competitors. Is grabbing this and is about to explode because not only are they integrating it into their features, which, a lot of big tech companies, the canvas of the world and, Microsoft of the world and Google's of the world are integrating these features in what I'm seeing a lot is that people will say once we get Microsoft copilot will be.

Fine. Microsoft Co-pilot for people that I don't know is an extension of their office suite. It's gonna be $30 a month per user. It is just rolling out now and it has AI built in, so it has AI and Google Labs does a little bit of this as well, where it's in your email, it'll help you write that email in your Excel.

It'll help you do Excel. That's a huge game changer. The problem is, and I've already talked to some companies that have that integrated and they're really. Site insightful ones are like, Hey, this is great. But now people are just using that and not thinking about the grand scope of generative AI. And that's why I think training is so huge.

Just to give you an example, what generative AI has done is that it's literally ripped the user interface off. Artificial intelligence, so it's essentially now a solution without a problem, which is sounds awful, but it's actually unbelievable. It's like going back in time and giving somebody a light bulb and people being like, Oh, my gosh, this is amazing.

But then you go back in time later and say, actually, I have something even better. I have electricity. Now, if you're the CEO of a company and you can go to that moment and be like do I want the light bulb? Or do I want, electricity, which can also power the light bulb, but all those other things, well, now that you probably want electricity, but back then they'd probably be like what do I use electricity for?

And you'd say, I don't know. What do you need to do, right? That's like generative AI. So I want people to think about generative AI as electricity. Think of it like, it's so much more powerful than just Microsoft building in one tool. And I think people just think Oh, now we're using it for emails and so we're way ahead.

 It's electricity. It's going to do so much more than that. And if you can grab that, it's going to get you so far ahead. And in a way that's going to be very hard for other competitors to follow. So I think the risk is just leaving so much value on the table. And that's right now, that's not.

in the future. That's right now. 

[00:37:46] Mahan Tavakoli: That's a wonderful analogy. And when you are experimenting with that electricity, you will see a lot of opportunities and potential in addressing business issues that you won't when you are just experimenting with the light bulb. 

Now, Connor, you are dean of students at. The MBA program at New York Stern. This is going to also be transformative on education. I've been playing around with conmigo and thinking about how it's going to impact not just education, but also training and development. So we'd love to get your thoughts on the impact of generative AI on education will actually something happen because there are systems that are resistant to change.

Education is one of those. So where do you think the future of education will go with generative AI? Will it be more of the same as this being like a calculator that is used on a side, or will it transform education in your view? That's a 

[00:38:49] Conor Grennan(: good question. It's hard to say about the analogy of the calculator.

It's hard to say if it will disrupt in that way or in a different kind of way. It certainly disrupts with writing, for example, it's english and and i'm a writer by trade, right? So that's very painful to think that kids are gonna have a harder time learning how to write now In terms of how it will transform.

It's funny, right? Because what we're talking about with ceos is look there's Huge value on the table, it's for you to grab right now to keep up or to get ahead whatever education it's more like a giant wave has just come in and knocked out the power and, your community, whatever, or it's like a tornado has just landed. There was no preparing for it. And also it's not like you can pick up your town and move it out of the way of the tornado, 

education is a very stayed, very solid function. Why is that? It's because we've been teaching the same way for In actuality, probably two or three thousand years,

it's just teacher says something, student goes away, thinks about it, comes back, says, this is what I think. Teacher's you have gotten an A, or a B, or a C, or D, very simple. That's all changed. Overnight. Imagine that. Just completely, it's like changing the way you use your lungs.

It's just, very hard for teachers to get their heads around that. I'm very sympathetic. What I say to teachers and, especially people with kids is, Don't just think of this as like a plagiarism tool, which by the way, it totally is a plagiarism tool. Don't get me wrong, right?

Let's not be Pollyanna about this, that's number one. But number two is, when I tell teachers, and this is why I even, work with teachers as well. I'm like, you also have this, Ironman suit, you can create this brand new stuff. So what I do is, work in biology with.

teachers and be like, Hey, listen, there's a whole new way of teaching the cell. The cell, which seems like it should be very interesting, is very boring, people are like, here's the mitochondria, here's the protocol. But in fact, if you made, the parts of the cell, different Marvel characters, and you made the bad things like the cancers, like into Marvel villains, and then you wrote a giant screenplay, people would probably pay attention.

And you can do that in seconds with chat GPT. It's so sticky and so memorable. So We just have to really rethink what's the new proxy for grading? What's the new rubric? All that kind of stuff. Now, luckily this is outstanding at that because you can ask Chattopadhyay hey, Chattopadhyay exists.

I don't want my students to cheat. What's a good way of teaching them? And it will help you come up with that new way of teaching. You're not alone. So I don't want people to fear it. I just want people to know that this is something that's going to help people learn. I have learned so much more with chat GPT than I ever did in school, but just that it's a pace of learning that I can't even tell you.

And the reason is because you can learn in your own context. If I love, the Miami Dolphins and I'll be like, Hey can you teach me string theory but use the analogy of how the Miami Dolphins play football. You know what I mean? That's a great way to learn. So I think that the possibilities are huge.

Don't get caught up in this is going to be terrible. The opportunities are really vast. 

[00:41:38] Mahan Tavakoli: They are. And that's why even the analogy that you just use goes back to the power of those questions. Teach me string theory using Miami dolphins. Football play as a way to do that.

There are great ways to learn and it customizes to your style, which can be really impactful either in augmenting education or I think transforming education still to this day has a strong sage on a stage perspective to it, as opposed to real engaging, real innovation. Kids 

primarily based on age group are in same classes, regardless of how fast they learn. Now, Connor, I know none of us have crystal balls, but as you look ahead at the next few years with generative AI, what are the transformations in the business?

World and in the way we operate in organizations as a result of generative AI. 

[00:42:37] Conor Grennan(: I think that we're just going to be streamlining jobs. They'd be broken down more into tasks. Look, if people want to be forward thinking in their company, and this is not taken for granted to be forward thinking, but if you really want to be forward thinking, what I would do, this takes a little bit of work with chat.

GPT goes much faster is take things like job descriptions, and really understand what are the tasks that go into that job description. , let's say you have somebody in an assistant director role or something.

There's probably 20 percent right now today that you could take off that person's Plate And I think more and more companies will start to do that it's only now just starting to happen and I think it's very slow companies are not really doing that right now But it is what I recommend and I think that's How people look at it rather than work flows through this person, then flows through this person.

It's going to now speed through this person and speed through this person. So what does that mean for your human capital? And that's where I think everything has to start in business, which is who do you need to complete your value propositions, of course, there's also.

How do we increase our value? 

And that's the thing that every company has to answer individually. There's no sort of one big answer, but understand that these tasks are going to be sped up and understand what you're going to do next. 

[00:43:42] Mahan Tavakoli: AnD I find Connor, even in my own work with clients, it has sped up parts of what I used to do.

In essence, it's been augmented or automated. It is really allowing me to tap into deeper insights that I wouldn't have been able to tap into otherwise. So now I end up doing more work, but the work is very different. As you said, there are certain tasks that are totally automated, but it does then provide more opportunity for.

Much more of the cognitive skills, much more of the judgment, much more of the things that can only come from human thinking. Exactly. 

[00:44:21] Conor Grennan(: Totally agree. So 

[00:44:23] Mahan Tavakoli: Connor how can the audience find out more about you, your podcast, your writing, all of the great work that you do, including with clients.

[00:44:31] Conor Grennan(: Thanks for asking. Yeah, I'm on LinkedIn a lot. I probably post on LinkedIn every day. It's pretty easy to get in touch with me through there. If you write to me, I'll write you back. If this does seem like the kind of thing that people are interested in, like a training, I'm happy to at least talk you through it, whether or not it's a good fit.

And yeah, and our AI applied podcast, which I think is up to three times a week now . That's just a fun way to just hear, some things that are going on in this space. But but yeah it's reach out. I'd love to hear from folks. 

[00:44:55] Mahan Tavakoli: It is a lot of fun listening to you, Connor.

It has been a joy learning from you. I love both the speed you are running in understanding generative AI, then translating it. Two applicable approaches for business executives and leaders, which is why I really enjoyed this conversation and really appreciate all of your great work. Thank you so much.

Same. 

[00:45:21] Conor Grennan(: Thank you so much for having me. I've been looking forward to this for a long time.