Most of Day 1 covered what has to change in the system. Patti Fletcher closes the day on what has to change in the leader. She starts with her grandmother, orphaned in the Armenian genocide and married off at 18, and the code of compliance that kept the family alive and then kept Fletcher herself employed long past its usefulness. Her argument is that AI absorbed the labor and left the accountability, and that a leader's real identity is visible in the calendar and the budget rather than in what they say they value. She walks five moves — recollect, reflect, rethink, redesign, refine — and cites GAO data showing 80 cents of every federal IT dollar going to keeping old systems alive, and McKinsey tracking 1,600 companies where last year's budget predicted this year's with 98% accuracy.

So have you ever stood at the edge of a decision that felt so big that it was too big because you knew that making that decision would cost you something? But underneath that fear, there was a question, a quiet one and a dangerous one. What if we get this right? Whoop, there we go. And notice this question is not what if I fail, not what will they think of me, but what if this is the one time it actually works? Now look, we all know hope is not a strategy and this question is not about hope. It's about intelligence. And that intelligence is your identity trying to tell you something that your fear is simply too loud to hear. I've been working and thinking inside of this question for a really, really long time. And I want to show you what happens when you start taking it seriously and not just for yourself, but when a room full of people take it seriously together.

We're going to start with the I, we're going to end with the we, and that is the whole arc of our time together.

Here's the thing. Everybody in this room knows what it's like to lead inside of a system that no longer fits the context in which it lives. And here's the thing, right? We all say the system's broken, there's a bug in the system, but we're all process people. We're product people, we're tech people. We break processes down. And here's what we know. The system is doing exactly what it was designed to do. So navigating it without losing yourself or the mission, that is arguably the hardest leadership problem in government right now. Let me give you an example of my own life, of working and living inside of a system not meant for the context in which it lives. My mom was a working mom before working moms were a thing. And so part of my childhood, I was raised by my grandmother. And my grandmother was orphaned in the Armenian genocide at six months old.

And so all of the decisions made for her and for her sisters were made from her brothers who were children and then young adults. And her brothers made the best decisions that they could for their sisters and for their youngest, my nana. But their focus was really that bottom of the Maslow's hierarchy with safety, with having food in your belly, a roof over your head. So some of the decisions they made for my grandmother were who she would marry. Back then, Ellis Island was closed to people who would be wards of the state. And so in order to get my grandmother over to the states where the rest of my family came, they put her to an arranged marriage. She was 18 years old, an arranged marriage with my grandfather who was in his 30s and not a very nice guy. And so they did this, of course, because they wanted her to have food in her belly and a roof over her head.

But here was the cost of that. In order for my nana to be able to survive, she had to completely stay quiet. She had no agency. She had no say in the decisions that would impact her life. That was the system and the culture that she grew up in, that she experienced her entire adult life. So in order to stay under that roof, in order to have her belly fed, she was conditioned to be obedient, to be compliant, to constantly prove why she belonged under that roof.

So here I am, born in America, full of beans, and thinking I'm at the top of Maslow's hierarchy, that I'm all self-actualized. And it wasn't until I was well into adulthood and working that I realized I was running on that same code, the code of being useful, the code of being worthy to be in the room. I had given up my power and agency to compliance in order to keep myself in the room, in order to earn the right to be there. The truth is that code kept my family alive and it kept me employed until it didn't. But a code built to survive catastrophe does not always know when the catastrophe is over. Back in my day and even still now, we called that professionalism, but we cannot call it that right now. So me being curious like the rest of you, I did some research to find out why, right?

Why are we running code that we never chose? So I found this study that followed 50 individuals throughout career change. And what they found with these individuals is that they would look at role models within the company that they had joined and see what made those folks successful, and then you would just copy them, right? They tried on like a new suit and they'd copy them and they'd do it over and over again, so much so that it became their identity. That same study found that 92% of those folks that took on someone else's identity did not realize that the identity, although worked once, was not working anymore. They simply could not see it. What I realized was that code I was running of being of use, it wasn't a flaw in my character. It was a role that I never asked to audition for.

And then, so actually, some of us, at least I did, I called it a mission. I proved my compliance before I proved outcomes. I earned the approval of the process before I earned the trust of the people of what the work I was doing, those people who was actually supposed to serve. And I was holding it together, really hoping that nobody saw the seams. And I did it so well for so long that the holding became my identity and the agency that was actually needed from me completely disappeared. And then something happened that probably everyone in this room and me who grew up in enterprise technology saw coming, but didn't necessarily see the impact or the speed of the impact. And that was AI and it did not come gently. Every report that took a Sunday to write, done in seconds.

The intelligence analysis that took a week, the logistics that took a whole planning cell, the schedule that took four hours automated already in production. This is across every single industry, right? The systems that we were to deliver, the programs that we were told our responsibility to run, and the decisions that each of us have built our credibility on. But here is the thing that AI cannot do. It changed the return on where you spend your resources, your time and your money. And that's the leadership discussion we're going to have today, where you allocate your resources.

The organization's getting hollowed out right now and we're seeing this on what seems like almost a weekly basis. They're running on an old operating model where the mission's already moved, but the full budget is already working on an old operating model, right? They're not out of strategy. Leaders are not out of strategy. We hear it all the time. They're not. They're simply out of capital allocation. They're out of time allocation. And there's a harder line underneath all of that. What AI cannot do is often what you were told to stop doing if you wanted to be successful in the system in which you were operating.

AI can't command responsibility. It's that simple. It cannot stand in front of a committee and be accountable for a decision. It cannot hold the trust of a person who needs to know that a human being with a conscience and a career made the call. That is not bureaucratic overhead, right? It's legitimacy that makes the mission possible. In 2003, and some of you might already know this story, and when I talk about this in other industries, they're always just amazed by it. So in 2003, it took 2000 analysts to run the targeting cell that the Army called the most efficient in its industry. Last year, that same base ran the same mission with 20 people in a system, AI system called Maven, right? Lots of us have heard of it. That's a hundred times fewer people in minutes instead of hours. And the government did the right thing as a result.

The new federal policy now says that any high stakes AI decision needs to have a signature from a named individual who's going to accept that risk, and that signature cannot go down the food chain. It has to be from the person who's in charge. AI absorbed the labor, but it did not absorb the accountability. It never should, and I really hope that it never will, because someone still has to sign. Now let's talk about another industry, the airline industry. A major airline was sued because their chatbot gave the wrong information to one of the passengers and in their suit, back to them said, "Hey, wait a minute. That's a whole other entity and we can't possibly be responsible. It was the chatbot that did it." And the tribunal said pretty much, "That's a remarkable submission," and made the airline pay anyway. It is always going to come down to people, right?

Brick buildings don't do brick business with brick buildings. Machines do not do real business with machines. People do business with people, and that humanity matters. You can hand the task to a machine, but you cannot hand it the signature. The system rewarded speed and it rewarded output for decades, and it undervalued, often undervalued, the judgment. AI just automated the part we were told was the biggest deal, the things that we should focus on. What remains is the part that it undervalued, and that part is you. So now let's break it down on what that actually means. First, you've been working hard at the wrong version of yourself, right? The one that feeds the old system.

And meanwhile, the world that you were trained for has been automated, and the thing you were told was a liability is the only thing left that is scarce. And if you think you're behind, and I think pretty much everyone in this room knows that you're not, you are at an inflection point, right? That decision that might cost you something, that's exactly where you are. So here's the thing. What do you ask yourself? What if we get this right? Notice it's not, what if I get this right, but what if we get this right? And I mean that question seriously. And I want to show you how to answer it because the way to answer it, it's not a philosophy. It is a practice. And it's a repeatable one because inflection points don't happen once. In the AI economy, inflection points happen over and over again, right?

It's innovation at the speed of the machine, no longer at the speed of human thought. So we're going to go through five moves. You already know three of them and just know them by other names.

Okay. So we're all here at Prodacity to navigate, to forge, and to lead. And those aren't just tracks on this fantastic agenda over the next few days. They're the heart of the reset that everybody in this room must run. But before we navigate and after we lead, there are two moves that humans tend to fail on more than anything else. So we're actually going to run through all five. So let's start with recollect. Recollect is essentially the what. It's before we interpret anything, let's gather the facts. What happened? When did it happen? Where did it happen? Who was involved? And what was the time and money allocation associated with it? This is not the story that we use in the briefing. It's simply the record, but it's the one that most of us skip as leaders. It's the one that most skip in programs because it's human nature to go straight to the explanation and straight to the blame.

But the reset, it must start with an honest evaluation and that includes where that time and money actually went. The next is reflect, and that's all about why we navigated the way that we did. This is about tracing the identity you've been performing, your team has been performing to where it actually came from. And again, this is not about blame and shame. We do not solve problems that way. It's because you cannot redesign something unless you understand it, right? You got to break it down. Most of us are running a code that we never chose. It's the truth. Most of us are. And they were written by systems that needed us compliant and not sovereign. And AI demands that we change that. Rethink. Rethink is about how you forge. This is like, okay, where we question the old value logic of prove the process, produce the output, earn the approval.

But in the AI economy, that logic is completely obsolete because the thing you are pricing yourself on, it's been commoditized, right? We just gave some examples. That new logic is of course very human. It's judgment, it's consequence, and it's a willingness to stand by and be accountable for a decision regardless of where you sit on the value chain. So what do you bring that can't be automated, right? That becomes the question that all of us talk about all the time. And that's the second part of what if I get this right? So let's think about it this way. Redesign is how you lead and this is where it gets real. And here's the part that in every room I'm in, nobody ever wants to hear. Your identity is not what you say you believe and it's not what you say you value. It's about what you fund, plain and simple, right?

What you fund.

I often hear the, "You know, Patty, I really do value these things. I really believe these things." And I think, "That's great, Mr. And Mrs. Leader. Let's take a look at your calendar and let's take a look at your budget." I see a lot of head nodding, right? Absolutely. That's the identity you are running. That's the reality that you are running. And that is the best place to start because no matter what you say, it doesn't matter what you value if you're not living in that new identity. Reflect happens in your mind. I'm sorry. Yeah. Reflect happens in your mind, rethink happens in your mind, but redesign, this is where we come to allocation. This is where the rubber meets the road. Nothing has reset until the capital has reset. Nothing has reset until the time has reset. You invest your time, your money, and your energy in what matters.

And this is the hard part. You divest in what cannot matter anymore, right? It's Eisenhower Matrix in practice. And yes, that second one of divesting, that's the hardest one for leaders. It's so hard, but it's also the most critical. And it's also essentially the whole play.

Every transformation postmortem that I have ever seen in a company, in a government program, it reaches the same finding, capital and time problem. The organization kept funding the stuff that no longer mattered. Things like gatekeeping, funding the gatekeepers, the people who decided who got access to information, who got access to be able to solve a problem for the few, gatekeeping for the few. But in AI, we all know this as solution providers and people who are purchasing solutions, AI requires that we fund getting the people closest to the problem to be the architects of opportunity for the many. And many of you I've talked to are doing exactly that. We're getting tech into the hands of the people who are actually facing the bottleneck problems. That's what we fund and it requires a completely different shift in what we spend our money on and how we spend our time.

So I was taking a look at GAO data and what the GAO data found in 2016, 2019, 2023, and again in January of this year is that for every dollar the US spends on IT, Fed IT, 80 cents of it goes to keeping the old systems alive. 80 cents. That's only 20 cents that's funding the future. And of course it's not just government, right? McKinsey tracked over 1600 companies over the last 25 years. And those of us who have been in decision making rooms are not going to be surprised by this. With a 98% accuracy, McKinsey found across all of these companies, across all of those years, that last year's budget predicted this year's budget. Sure, the strategy decks were updated for the investors, for the customers, for everyone in their supply chain, for their employees, but the spreadsheet essentially stayed the same.

Okay, so let's talk about refine because nothing you do just stays reset, right? The moment you stop reallocating, you have no reinforcement mechanisms to keep you in the new identity, what happens? We regress back, right? We go back to the way things were. Entropy funds the past. It is what it is. So refine is the discipline of coming back to this reset at every milestone, at every view, and every time you see the signal that the mission is outpacing what the org can actually do. You never reset once, it's ongoing. I kind of think about it this way. Resetting is the new resiliency where you are successful because of the disruption and not despite it. That's a lot of agility and we can do it. So everything I just walked you through, you could do it alone, right? You could make the decision maybe within your program, you could make the decision within your team, but it's not enough because the systems that each of us work in, they're not personal, they're not individual, they're collective.

One person resetting inside of a program that has not reset in the broader program, if they're lucky, they just reset what's right in front of them. So the reset does what it's supposed to do only, only when each of you can run it at the same time, the collective. And this is not a kumbaya statement, right? This is not all of us singing in a circle. This is an actual mechanical fact, right? We are pushing boulders up hills and it literally takes all of us to make this kind of change. A single leader who allocates her capital inside a broken program, it best improves one program, but not for long. A room full of leaders who do it together have the opportunity to move an entire institution.

The entire difference between an inspiring talk and actual change and how the government builds and implements and adopts software and the impact it realizes is dependent on the we, which is why the question that you walked in with was always the smaller version of the real one. You walked in asking, "What if I get this right?" And that makes sense, right? Change starts with the person that you see in the mirror. It's the right starting question, but it's the wrong ending one. The question should be, what if we get this right? Not what if we survive the disruption with our programs intact, but what if enough of us reset at the same time so that the institution itself starts to move? What if the people in this room stop funding the identity that the mission has already outgrown together on purpose and starting right now?

And here's the thing, I hear a lot about, I don't feel ready. I used to say that, but here's what's really important. Ready is not a feeling. It's a decision, period, end of story. And ready is a decision that every single person in this room is able to make. So let's end with this. Whoop, my slides are going the wrong way. I'm missing a slide, so I'll just read it to you. Recollect, reflect, rethink, refine. The mission has been waiting for us to be sovereign, accountable, and impossible to automate. It doesn't want us to manage it. It wants us to lead it and we have everything we need to lead it. Thanks everybody.

‍