AI can help school leaders save time. But saved time is only valuable if leaders know where to spend it.

This week on AmpED to 11, Brett is joined by Paul Bambrick-Santoyo, founder and dean of the Leverage Leadership Institute and author of Leverage Leadership 3.0. Paul has spent twenty-five years studying the habits, systems, and leadership moves that help schools improve, and his answer keeps returning to the same place: the moments when leaders are directly helping teachers get better.

In this episode, Paul brings a grounded perspective to one of the biggest questions in education right now. If AI can reduce the administrative burden on school leaders, how do we make sure those recovered hours go toward coaching, observation, feedback, and the kind of instructional leadership that actually changes outcomes?

The conversation explores what makes teacher development different from evaluation, why excellent practice needs to be seen before it can be replicated, and why AI tools trained on the average of everything may not be enough for schools trying to become exceptional. It is a thoughtful, practical episode for anyone thinking seriously about AI, leadership, and school improvement. Rebecca is away this week, so there is no AI Tip or News segment.

Listen to the full episode to hear how AI could help leaders refocus on the work that changes schools.

Brett Roer: Relationships matter and is what I think most educators are most afraid of in the age of AI, that they will be displaced. And yet every time we go into schools and we ask any person in that building, why do they show up every day as a student or an educator or a staff member?

Brett Roer: It's the people and the relationships.

Paul Bambrick-Santoyo: Computer screen to human being learning works for a small percentage of people who thrive in that environment, but it doesn't work for the majority.

Brett Roer: Paul, how are you doing today? And welcome to the podcast.

Paul Bambrick-Santoyo: Thanks for having me. Doing well.

Brett Roer: Great. So for those out there that have not read all his books like myself, Paul is the founder and dean of the Leverage Leadership Institute. I was honored as an assistant principal that my principal gained entry into that institute, and a lot of what I learned about how to be an impactful and effective administrator came from the works of Paul's books.

Brett Roer: So Paul has written a number of books. It's had a major impact on schools worldwide. He's written most recently Leverage Leadership 3.0, which is what we are so excited to discuss with all of the transformations that are occurring in education, especially in the age of AI. Paul, thank you for all the work you've done.

Paul Bambrick-Santoyo: I'll share with you a story that some of the listeners might know.

Paul Bambrick-Santoyo: It's how I open every workshop 'cause I think it's the same why. You know, I grew up just outside Chicago, and as a young boy I was obsessed with imitating my heroes. And as a young boy who loved sports, that meant my heroes were people like Ryne Sandberg and Andre Dawson on the Chicago Cubs when the Cubs won nothing.

Paul Bambrick-Santoyo: Um, and the biggest one was Michael Jordan, and I remember after watching him play, I'd go outside and try to dribble like him or pass like him or shoot like him, and you can't see in this podcast, but given my size of 5'6" and my relative lack of talent, I was never that good. But I did get better And that idea of imitating heroes carried with me into the classroom.

Paul Bambrick-Santoyo: And when I'd walk into her classroom, at first it felt like magic. She could take the same students I just taught who had been bored or disengaged, had them completely participating and learning at the highest levels. But the more I watched, the more I saw there was a move she made or word she used, and I imitated them, and I got better.

Paul Bambrick-Santoyo: And that carried with me in my first days as a school leader in Newark, New Jersey, and it carried with me my first days as a principal supervisor. And fast-forward now 25 years later, and I've had the chance to work with over 100,000 school leaders across the globe, and in 100 different cities where they are now the highest-gaining or highest-achieving school in their district, state, or country.

Paul Bambrick-Santoyo: Particularly in schools that need it the most. All the schools I've worked with, 99% of them have been schools with high Free and Reduced Lunch populations, high percentage of students of color, special ed students, English language learners. Because I feel if we can succeed in those environments, we can succeed anywhere.

Paul Bambrick-Santoyo: And so that's been my journey up to today

Brett Roer: So thank you for always sharing the wisdom that you have. I will tell you that the chief of staff at Amplify and Elevate Innovation is Versha Munshi-South. We've had many great guests on the podcast. She said, "I cannot believe that Paul is joining us," because she treats your books, like many leaders, as, as their Bible.

Brett Roer: This is what they use to become effective, impactful urban leaders. So again, just wanna say on behalf of her and so many other leaders who have shared that with me, thank you for all your work.

Paul Bambrick-Santoyo: Thank you.

Brett Roer: You're welcome. Now, we find ourselves, right, you've had this impact in the world of education, you know, pre-AI being...

Brett Roer: While it's always-- While it's been there, now that we are in the age of generative AI, you now have the ability to use a new tool to potentially be an even more impactful leader. So I'd love if you could share with everyone from your newest work, Leverage Leadership 3.0, what is new and unique about this?

Paul Bambrick-Santoyo: My books are littered with a tool I use to make it easy to remember.

Paul Bambrick-Santoyo: I call it a core idea. And so it's a way to try to create pithy, summative statements that capture it. And the original core idea I wrote in my first book continues to remain true, that exceptional leaders succeed because of how they use their time, what they do, and how they do it. And what has been extraordinary since that first book is 16 years later, that is still true today.

Paul Bambrick-Santoyo: Many years ago, I began working in Latin America, and the World Bank invited me to meet with all the secretaries of education of all the countries of Latin America, from as far as South Uruguay to the Dominican Republic to Mexico, et cetera. And in the process, I began working with a number of the countries.

Paul Bambrick-Santoyo: We raised early literacy results in the Dominican Republic with the National Principals' Academy and things of that nature. And so the World Bank then followed up with me about 18 months ago and said, "Listen, we'd like to write about AI for school leaders in Latin America." So what's fascinating for you potentially is that I began my journey with AI in Latin America, not even in this country.

Paul Bambrick-Santoyo: And so the World Bank just said, "Okay, let's make a working group of one hundred of those leaders from about eight different countries, and let's work with them only to do that." And the, the, the goal was very concrete. Can we use AI to reduce the amount of administrative work by ten hours a week? And measuring it by what-- by leaders' self-assessment and by leaders saying, when did they go to bed?

Paul Bambrick-Santoyo: And we named that prompting framework... I'm gonna translate it now into English. The, the book we wrote, El Director Libre, is only in Spanish, but if you're a Spanish listener, feel free to download that from the World Bank site. It's a free access book, and you can read it. I've since refined it for an English context, and I would say the prompting framework was me, you, what, how Tell AI who you are, tell them your context, tell them what matters to you.

Paul Bambrick-Santoyo: Then tell AI who you want them to be, then tell them what you want and how you want it. And after you do that, always revise and add a personal touch. In today's day and age, that's pretty common sense for most people who use AI. They know that. Although many people forget to tell AI who you want it to be.

Paul Bambrick-Santoyo: A set of prompts to get an agenda to solve a problem, a set of prompts to stop absenteeism. And so the book is really just a set of prompts you can plug into generic AI, and it gets you better answers So the success was 10 hours gained. What was left undone was, okay, so what are we doing with those 10 hours that actually improves the quality of learning and teaching in the school?

Paul Bambrick-Santoyo: I consider them have two flaws. The first is the flaw of it just makes mistakes. And again, I'm preaching to the choir of many in this audience, but it was a news to me to realize that AI is totally different than a computer. 'Cause a computer is based on basic one and zero. Every choice you make is one or a zero, and so that's when you press send on an email, it sends 'cause there's no choice to be made.

Paul Bambrick-Santoyo: And so when you realize it only is probabilities, that's when you get crazy answers. That's when you get answers that are wrong or that are off. And most of us have experienced that. That's the risk of AI. That's why we don't use AI for mental health. We don't use AI recommended for healthcare questions, 'cause you can get answers that are just wrong.

Paul Bambrick-Santoyo: But the biggest-- the other biggest problem of generic AI is that it does not include proprietary content. It doesn't have my books. It doesn't have textbooks because it's not allowed to upload that material into AI. Some people have bootlegged parts of it, but you probably saw the lawsuit settled by Anthropic for billions of dollars when they tried to upload a bunch of books and hide it.

Paul Bambrick-Santoyo: My wife is a doctor. She uses Open Evidence every day. Open Evidence is a specialized language learning model that only has verified books, medical textbooks that the medical community has approved as standard procedure. And so when she's doing a diagnosis or working through test-- analyzing test results, she has a specialized expert in her corner who can help her get information.

Paul Bambrick-Santoyo: Same thing, ChatGPT Law, same thing for Bloomberg Finance. These are specialized LLMs that I think are the future of AI, is that you guarantee your information's coming from a smaller pool of verified information, and so the odds of it being accurate and high leverage go way up. 'Cause without it, the problem with generic AI is what I call the flaw of the average.

Paul Bambrick-Santoyo: And so I began the journey about a year ago of creating a specialized LLM, and I just uploaded all of my 12 books. And if you've ever taken a workshop with me, you know I'm famous for one-pagers, trying to synthesize it down to a page, exemplars, videos. We just put it all into the app and started to teach it.

Paul Bambrick-Santoyo: I think we've learned from the pandemic, we've learned that computer screen to human being learning works for a small percentage of people who thrive in that environment, but it doesn't work for the majority. The majority don't learn as well. And, and so the key driver I continue to see 16 years later that separates good schools from great is the amount of face-to-face coaching that teachers get, and the amount of face-to-face teaching that students get.

Paul Bambrick-Santoyo: And these are school leaders that had made 20, 30-point gains in achievement, um, or leaders that I'll talk later about Rick Romaine, who went from the 10th percentile in New York City to the 95th percentile on his state test results, serving kids that are nearly all free or reduced lunch. And we brought them together and said, "Let's coach this LLM, specialized LLM, to do things to your standard."

Paul Bambrick-Santoyo: And so we set out to create exemplar prompts, guides, and, and narrowed the coaching so that we could produce something that leaders could use that could raise the ceiling for their coaching and do it in less time. So we can go in many directions there, but I'll pause there, Brett, to let you take over.

Brett Roer: Well, I think, you know, I speak for everyone that, uh, I'm in community with. You know, we often talk about, and this is the, the biggest thing that I think that came out of ISTE this year, how do you keep humanity at the center of education in the age of AI?

Brett Roer: Because I think you are very good at telling, like, "Here's, like, the action or here's the synthesis of it." I think our listeners, it would be a great honor to hear from you what are some of those exact action moves that are now evidence-based that they could be doing themselves after, you know, they get access to your, uh, book and resources that they could be, they could be transforming for the upcoming school year.

Paul Bambrick-Santoyo: Like, rather than spend six hours on Sunday writing this, how to do that easier; problem-solving meetings; attendance tracking; quick meeting development; uh, security networks; and helping teachers be innovators with AI, like just creating a culture where they can innovate So those were the, those were the 10 user cases we solved, and we just designed a set of prompts for each one that you could say, "Here are the prompts you could use in Claude, Gemini, ChatGPT, whatever you want, that could immediately get you better answers in each of these ways and save you time in your planning."

Paul Bambrick-Santoyo: If you think about that me, you, what, how framework I gave you earlier. It was like, okay, I am the principal of a school of 800 students, all elementary students. That's the me. And we're planning a family fun day in the park for Saturday from 10 to 1. Okay, that's the me. You. Be my multimedia communication specialist.

Paul Bambrick-Santoyo: So it, it gave, for principals, really fast an ability to do this family fun day in 20 minutes rather than in three or four hours. And so that was just one simple example that was, like, a moment that leaders were really excited about to spend less time worrying about that.

Brett Roer: That's incredible. And I think what you really touched on, and, you know, just to actually validate this point, yesterday I spent two hours in a district in a, in a high school principal's office with his administrative team, and over, over his shoulder was many of your books, and we talked about these things.

Brett Roer: Those are the biggest drains on time. And if there's more efficient ways to do it using that methodology you just described, I mean, that's, that's the way forward because it, again, repurposes that time for what they all said. I asked, "What fills your cup?" It's being in classrooms, helping teachers get better, building relationships with students in the community.

Brett Roer: Now, what were they doing with that repurposed time, and any stories or evidence you could share would be, uh, so worthwhile for other leaders out there.

Paul Bambrick-Santoyo: So this is really just about time right now, and I will add the AI add-on after that. But, uh, Rick Romaine is a phenomenal story. He's in Brooklyn, New York.

Paul Bambrick-Santoyo: Like I said, he went from first to-- last to first in his district, from 10th percentile to 95th percentile, seven hundred and fifty schools in all of New York City. Incredible inspirational story. And yet, if you met him, he's like a quiet, shy college professor. He could be a librarian. You would never think he was the leader of the most, the most dynamic, evolving, improving school in the city.

Paul Bambrick-Santoyo: I'm like, "But how did you do it, Rick?" And he said, "I'll tell you a secret. I'm not really good with my time, but I had a superb secretary. And so I said to the secretary, 'Listen, I want to meet with every grade-level team every week, and I want to observe this, uh, uh, teachers on this frequency. Help me build a schedule.'

Paul Bambrick-Santoyo: And they built it together. He said, 'And I want you to hold me to it.' And she was relentless. She'd knock on his door. 'Rick, you're supposed to be observing right now. Get out.' When a parent shows up and said, 'I have to meet right now,' said, 'Nope, Rick is observing right now. I can set up another time for you to meet with him.'

Paul Bambrick-Santoyo: Learning took off." And so I like to say to leaders, "Find yourself a hound Who's the person who can, you can give them charge of your schedule, and you can say, "Your job is to keep me in classrooms and meeting with teachers." And, and, and you have to empower them. There are very few secretaries in a school that feel that they have that power and authority to talk to their leader the way Rick's secretary does.

Paul Bambrick-Santoyo: It changes your life. I still remember watching one of my children get coached, and this little tip that they, to not be distracted with your laptop, that the coach taught my child and said, "Listen, when you're on your laptop doing schoolwork, face towards this wall. When you wanna check the internet, turn your chair to the other side of the table, turn your laptop 90 degrees, and then check the internet."

Paul Bambrick-Santoyo: And she's like, "You'll train your brain. One way you face, you do work. The other way you face, you play." It's like, that's just brilliant. Well, this is just executive function for school leaders, that we need a coach to help us r- change our time. Now, many of the success stories I talk about are leaders who figure that out first and then just need to know what to do.

Paul Bambrick-Santoyo: So rather than just say only for the ideal week, we adjust, and then she'll make all the calls. She'll email all the teachers. She'll tell them the new schedule. Rick doesn't have to do that. Rick just shows up at the right time that the leader, t- the secretary tells him to show up, and he shows up prepared It is such a simple formula, Brett, but I think we could dramatically improve student achievement immediately if we did those two things.

Paul Bambrick-Santoyo: Someone who hounded us to do s- time the right way. They have to know what's the right use of time, and then give leaders the tools to how to use that time once properly all- allocated effectively. I think those two things alone would dramatically change student achievement.

Brett Roer: Again, what I'm learning about AI, and it's validating to hear from leaders such as yourself, is there are some very common sense moves.

Brett Roer: What is your pathway for success? You just named, like, shaping the path, who you need to have on the, on the journey with you. How does one go about actually doing that?

Paul Bambrick-Santoyo: Do you wanna go on the how you spend your time, or do you wanna go on the how do you shape?

Brett Roer: So imagine-- Yeah, no, that's a great question, and you're right. So here's the first one, like, conceptualize for a leader who's, you know, they're listening to this podcast, they're like, "Okay, I got this. I even know the right person for this."

Brett Roer: What are the exact-- There's no, you know, if there is a prescribed way, like a lot of people like to know, okay, step one, do this. Step two, do this. If there are like a progression of steps that you would recommend on how to even build that initial system, maybe we start there. And then there's what are the steps where you're actually doing that are gonna transform student outcomes?

Brett Roer: Like, what does Rick do in those teacher meetings that's so impactful and effective, right? Because holding that space is crucial, and changing that habit is crucial, and yet without additional refinement and training, those could also be an inefficient use of time and maybe not impactful. So Rick did something right.

Paul Bambrick-Santoyo: Yeah. So let's start with your schedule. So how do you build your schedule? So the first should not be secret, but the first secret or core idea to great leadership is that- Teachers grow the best with one primary coach, not with multiple people giving them feedback.

Paul Bambrick-Santoyo: And the first big error schools make is they have a large leadership team, or they have even just a principal and assistant principal, and they're like, "We're all gonna all observe everybody, and we're all gonna give feedback to everybody." And what that means is you see people infrequently, and from one leader they hear this, from the other leader they hear this, and then they have to figure out how to reconcile those two pieces of feedback.

Paul Bambrick-Santoyo: But the main coach who's developing each individual player is your specialized coach for your area, and the head coach makes sure all those coaches are aligned to a same vision and plan. I think the same thing should be happening in schools. Simple leadership team, principal and assistant principal.

Paul Bambrick-Santoyo: Divide your school in half. You got half the teachers, I have half the teachers. Even if the principal's the only one who can write a formal evaluation. Formal evaluations do not count as teacher coaching. I'll say that very firmly. They don't, they don't coach. Those are evaluative tools. They're requirements.

Paul Bambrick-Santoyo: They're doing coaching. And so larger leadership team, you've got a high school principal as assistant principal and multiple department chairs. Distribute your staff over all of those instructional leaders. You have not many instructional leaders? Pick two teachers who are extraordinary, who can give a stretch assignment.

Paul Bambrick-Santoyo: "Will you coach two of your peers while also teaching?" And most leaders I do find like some small stipend they can give that teacher, and the teacher's thrilled to be honored as like a veteran teacher. When you do that, ninety percent of schools in the United States can get to a fifteen to one ratio of teacher to leader People don't believe that number because they think, "No, I've got 60 teachers," but they forget the number of people who could actually coach.

Paul Bambrick-Santoyo: And so if you get to a ratio of 15 to one or less, you can coach every teacher in the school every week. And this is the bombshell moment for a school leader who thinks, "No, I can only coach a teacher once every few months, and I'm in the best." You know, Rich Marzano wrote in his book, Teacher Supervision and Evaluation, that like the average teacher's observed once every two years, and the average rook- rookies observed twice a year.

Paul Bambrick-Santoyo: So 90% can get to 15 to one. 100% can normally get to 30 to one. I've-- It's 99%. I've met very few leaders in the United States who can't get to 30 to one. But if you get to that ratio, you can build a meeting, you can build a schedule where you're now meeting with them face to face every week or every other week if you're above 15 to one, and observing them every week.

Paul Bambrick-Santoyo: And how? The, because you start by locking in those meetings into your schedule. You start from a blank piece of paper and say, "Okay, let me lock in my meetings first. Here's when my faculty meeting is." And normally you can schedule a leadership meeting at various points, so wait on that. "And here's when I could meet with each of these teachers I'm responsible for on their prep periods.

Paul Bambrick-Santoyo: Then you put your leadership team meeting on your schedule, and then when you're done with that, uh, we also put what we call student culture time. Block out time is when student crises or issues are gonna happen that'll likely blow up your meeting, and it's pretty easy to know when those are. That's breakfast, lunch, and dismissal.

Paul Bambrick-Santoyo: And, and then as a leadership team, you decide, "No, we can't have each of us responsible for every breakfast, every lunch, and every dismissal th- five days a week, or we don't get anything done." So you say, "Who's the buck-stops-here person? All right. We'll all be when they arrive, but only the AP is in charge of Monday arrival if something big happens."

Paul Bambrick-Santoyo: And what we teach in observation is in 15 minutes you can learn everything you need to know to coach a teacher. Move away from full block observations and get to three to four teachers in a single teaching block. And I can ta- talk about how if we want to, but that'd be a sidebar for the moment. But that means with four observation blocks a week, you're getting to 15 teachers So you start to do the math.

Paul Bambrick-Santoyo: It sounds like such a reasonable number, and yet it is a game changer because normally what happens is an instructional leader has one week of the year that looks like that, and every other year doesn't, every other week doesn't. Every other week gets blown up, or we don't hold to it, or we're not consistent.

Paul Bambrick-Santoyo: But by locking those meetings in, what do you do? A teacher does-- You don't have to look for a teacher to get them feedback. You don't have to send a follow-up email to give them feedback. They know when they're gonna find you, and you feel the pressure. They're meeting with me at Tuesday at ten. I better have observed them before then.

Paul Bambrick-Santoyo: You know, I better have taken a look at their student work before then 'cause I'm accountable. And if you have that hound that makes sure you don't cancel your meetings and says, "Okay, Tuesday, the district called you into a special meeting. Let's reschedule your teacher. Her other prep period is Wednesday."

Paul Bambrick-Santoyo: This is when you can have two hours uninterrupted, two to three, to prep for your meetings, um, when you need to think, when you're looking at student work, when you're looking at a lesson plan, when you're thinking of your observation. And the, the reality is for many a school leader, those hours cannot occur during the school day.

Paul Bambrick-Santoyo: For many of us, it occurs before school starts or after school ends. It is a harsh reality of leadership. But there is one exception that helps many a leader, when a leadership team agrees together to cover for each other. So the AP says, "From ten to twelve on Tuesday, I'm gonna schedule no meetings, so no matter what happens in the school, I can cover it.

Paul Bambrick-Santoyo: 'Cause CEOs have, like, a phalanx of 17 secretaries blocking you from access. You know, you can't get to Taylor Swift unless, like, you have direct line. But a principal doesn't have that, and so they get interrupted every day. And so I coach principals, "Go to the janitor's closet, or if the social worker's not on site, or the nurse's office isn't there, or go across the street to a coffee shop, and tell your AP, 'Only call me if there's blood on the floor.'

Paul Bambrick-Santoyo: Like, like, "If it's an emergency. If not, you handle it." And if you can do that, you can get prep work done during the day. If you can't, for some leaders, you're getting to school early or you're doing it after school, but it's gotta be a time where you can't be interrupted That's the building of a schedule.

Paul Bambrick-Santoyo: A Monday meeting is where the leadership team sits down and say, "Did you meet your schedule this past week? What kept you from your observations or your coaching? What adjustments do we have to make this week, either 'cause of time or because you got to follow up with somebody?" So we say, "I did third grade weekly data meeting.

Paul Bambrick-Santoyo: We planned a critical reteach lesson around fractions of a whole. They're teaching it on Wednesday at nine AM. I'd like to observe." "Great. You were supposed to do this. I'll cover you so you can observe that class." Like, they adjust their schedules every week to actually respond to what teachers and students need, rather than trying to fixate on a locked schedule that never changes, and to acknowledge the life as a school year.

Paul Bambrick-Santoyo: And also 'cause our work around executive function, we know as well, color-coded schedules we follow more than when they don't have color. And so, you know, if you color-code your Outlook calendar, you're much more likely to be successful in, in keeping it. And so we build schedules by hand, and then you can use the app to make it quicker and to adjust it.

Paul Bambrick-Santoyo: But I find that until you can visualize how doable it is in front of you, you can't even understand the power of an app to or AI to help you make it better.

Brett Roer: The whole time you're explaining this, you know, I'm going back in time to my first year as an assistant principal. I wanna give some flowers to Greg Dutton, who is the reason we were fortunate enough to have this podcast.

Brett Roer: Love Greg.

Paul Bambrick-Santoyo: Yep. Hello, Greg. It's good to talk to you via the internet.

Brett Roer: And for that one year at least, you had to observe all of the attributes of Danielson, not just like the core eight. And I remember thinking, "How is this possible that we could ever do this?" But it's what you said. The first step was build your schedule. And I think I had 18 teachers that year I was directly responsible for coaching and evaluating.

Brett Roer: It's like, right, every two weeks you have to have been in every one of their classrooms. You had to have done a coaching session. You had to have done a pre. And it wasn't about evaluation. I think people were just so shocked from the New York City Department of Education, but really everyone, like, that you needed this amount of level of coaching, but it really did create a culture of coaching, you know, overnight there.

Brett Roer: And f- you know, the, the outcomes did show that. So yeah, listening to that and then thinking now of those parts that were so challenging as a first-year AP and anyone really trying to adopt your system, having AI to support some of that work would've been truly, like, a perfect marriage of your impact that you know works from all the research you've done and the ways AI, as you've already shown, can make the, the workload more efficient.

Brett Roer: So just again, like, thank you for sharing that wisdom.

Paul Bambrick-Santoyo: Well, thank you.

Brett Roer: Yeah. As we begin to wrap up here, some of the things we love to allow our guests, there's two things that we like to end with. One is Rebecca and I usually turn the reins of The AmpED 211 Podcast over to our guests. So in this case, Paul, you're now the host of The AmpED 211 Podcast for just one question.

Brett Roer: What's something that's a question that's been top of mind for you that you'd love to discuss and ask, and we could have a quick conversation about? Anything that comes to mind that you'd like to introduce as a topic that we could be discussing. You're now the host, so you take it away.

Paul Bambrick-Santoyo: So you've been doing this podcast for two years, three years now?

Brett Roer: Yeah. Mm-hmm.

Paul Bambrick-Santoyo: Okay. What are the top three biggest insights you have learned from your guests so far?

Brett Roer: And yet every time we go into schools and we ask any person in that building why do they show up every day as a student or an educator or a s- staff member, it's the people and the relationships. So the biggest one insight is our task in the age of AI is to make sure relationships stay the primary reason we show up, and rigor doesn't slip as a result.

Brett Roer: The tool is agnostic. There are better tools than others that do certain tasks. So we always go back to the driving analogy. Once you learn how to drive a car, then you decide what the task is, then you pick the best car for that specific task. But the idea of knowing how to drive a multiple types of vehicles is the important thing, and that you have the aptitude and you know how to, you know, navigate different circumstances.

Brett Roer: So some version of that analogy. I think the last one is similar to what, how you just described, you know, a model that you have found is most successful. Whatever our guest has kind of found is their area of expertise and passion. I just heard Richard Culatta say this, so I'm gonna attribute this to him.

Brett Roer: And so I think one of the things that many of our guests have said that I hope continues to be the way forward in the age of, you know, education, AI, innovation, edtech, et cetera, is when we come up with new solutions and tools, we don't just take the check, drop it off, have one, you know, surface PD and say, "All right, y'all got this."

Paul Bambrick-Santoyo: Those are, those are fantastic, and on, on-the-spot answers, Brett. Very impressed. Could I respond to number two? Please. You are the host. You do you. Yeah. So I completely agree with relationships matter. I love the analogy of Boeing. And you said the tool is agnostic. My only response is when we get to specialized M- LLMs, we can take away the agnosticism.

Paul Bambrick-Santoyo: After I'd shown a bunch of videos of high-achieving leaders, we'd done a bunch of exercises of how to look at student work, and they said to me, "I think I'm understanding why this is daunting for me. I've never attended a good school in my life. Not in preschool, elementary school, middle school, high school, college.

Paul Bambrick-Santoyo: I don't know what a great school looks like, and so this is all brand new for me And it just resonated with me so deeply that we lean on our own experience of school to lead. And yet, if we had a school that was on the surface great, but like actually just relied on private tutors or what you did outside of school, like many of us don't have any experience of what excellent teaching can really look and feel like.

Paul Bambrick-Santoyo: Nearly everywhere in the country is using HQIM, but HQIM is not improving student learning. And so we were teaching them how to do a planning meeting, how you take the lesson. In our Leverage Leadership app, when you log in, you state what state you're in, and then you can immediately upload any lesson plan.

Paul Bambrick-Santoyo: It will evaluate its rigor against your state standards and your state tests and tell you how to make adjustments. And so I was saying, "Okay, high school principal, you're getting ready to coach your biology teacher, 10th grade biology, around this lesson." And I taught him a framework for planning meetings.

Paul Bambrick-Santoyo: Like the whole room, it was like l- everybody was able to be vulnerable and like feel this. And so we said, "Okay, let's try it." So we plugged it into the Leverage Leadership app, said, "Plan a planning meeting around this lesson plan for biology." It gave him a no-show chart to unpack the standards, so he learned some about the biology, what students have to know and be able to do to master the standards.

Paul Bambrick-Santoyo: So he learned for himself. Gave him the adjusted lesson plan, and then gave him a script to meet with the biology teacher. We proceeded to role-play the meeting. I put him out in what I call the hot seat in front of all the twenty-five people in the room, and he led the meeting and gave him feedback. And the meeting was the best, like the highest quality of stuff he had ever done before, what his peers said to him.

Paul Bambrick-Santoyo: And that's where I think AI can be, not just a agnostic tool, but putting expertise in their hands, but requires specialized LLMs. And if we do that, I think AI can become an amplifier of excellence, not just an augmenter of time.

Brett Roer: May I, uh, may I respond to your response?

Paul Bambrick-Santoyo: Yes. Yes, you may.

Brett Roer: Much appreciated. So I actually wanna commend you for pushing that thought deeper, and here's probably an addendum I should've made the first time.

Brett Roer: And then we say, "You already have enough power there to do what," you know, like what you've just mentioned. But now you need to train that model on expertise. So we call it the AI sandwich strategy. What is all of the human knowledge and expertise pre-AI that you've already written? So instructional frameworks, handbooks, mission and vision, beliefs, your best observations that you think are worthy of emulating and replicating.

Brett Roer: And we say like, until you have every single piece of that, and we keep pushing and pushing or train an AI model to push you to say, "I still wouldn't know how to answer this," and that's the most important part, that wisdom collection and really refining what you're gonna create your outputs on. Then we click a button in AI.

Brett Roer: 'Cause if you're training it on the best patterns in the United States and what's most likely used in education, it's-- the results of not-- are not what you're getting. So therefore, be really careful about before you use click-- I say before you click a button, is that the best version of all the resources that you have available, and have you really thought that through?

Brett Roer: But it is the most, it is the most important part to realize like, again, you're someone like Greg who builds a community of feedback and, you know, we're gonna push each other, but we're gonna be there for each other. So challenging at first, and getting teachers to role-play in like a debrief. But once you move past that, it makes them realize, right, like get it right in here because the alternative is you get it wrong in there in front of kids.

Brett Roer: And so if we can all agree like, "Hey, this might feel weird, but it's for the kids," usually then people can move forward. But that definitely-- I'm glad to hear that story, how validating it was to make sure that person left feeling they are smart. But, uh, that's something you definitely wanna push on and frame a culture of trust and like why we do it.

Brett Roer: Otherwise, it could feel like artificial and not helpful, but it's so insightful and helpful. So thanks, Paul, for letting me respond

Paul Bambrick-Santoyo: If we value education, we'll practice, but that is counter-cultural to our most, most of our own lived experience.

Brett Roer: Well, we close with this always, and you've done a great job of, you know, giving flowers to, you know, that, that school in New York City and to people like Greg. Before we leave, if you could let listeners know who else out there in the space, and again, you know, don't feel like you have to name every single person.

Brett Roer: This is just maybe who's top of mind today for you. Who are other folks out there if people listening want to learn about how to utilize, be a better leader, how to utilize AI to help support them in their roles? Obviously, besides going and finding the Leveraged Leadership 3.0, who else should they be looking towards for guidance, support, and wisdom?

Paul Bambrick-Santoyo: I'd say it's an insufficient but necessary step on the journey to college readiness and beyond. And so find the places that are using AI to improve student achievement outcomes. And when you ask that question that way, there aren't many places to look yet. You know, there's a lot of innovation without impact.

Paul Bambrick-Santoyo: One of the groups I'm most excited to, you know, work with is, like, the Teaching Lab, which re-recently merged with the Re-Re-Relay Graduate School of Education. But the Teaching Lab has experimented a lot with on-site coaching of teachers, and they've also got some AI tools they've put alongside it, and I think that's a really fun place to see the work they've done.

Paul Bambrick-Santoyo: And so go to the places where the learning has gotten better, and I think you'll cut through the noise, and you'll find how is it being used as one tool of many to actually improve outcomes for kids.

Brett Roer: That is such a great-- Again, many of the ti- things that you're saying are the most common sense, and yet least replicated.

Brett Roer: So again, I hope folks just, just, just heard how clear that message is. Thank you for shouting out some of the orgs and some of the districts that people could start looking into and finding, you know, research on. And Paul, last but

Paul Bambrick-Santoyo: They'll be, they'll be all across the country. And check us out at the Lever Leadership app, and you can see it in action, and join a community of other leaders who wanna figure it out, not just to do AI, to actually improve learning, and we'd love to, love to have you.

Brett Roer: That was what I was just about to ask.

Brett Roer: And Paul, thank you because that is the most important thing from these podcasts, right? When we bring on leaders in the field, those that have proven transformational outcomes and impact in schools around the world, around the country, often with the hardest to reach students based on historical demographics.

Brett Roer: That's who we need to be listening and learning from. Paul, thank you for continuing to revise and refine, you know, all the work you've been doing to make sure it keeps meeting us where we are in this current age of AI. So again, on behalf of the AmpED to Eleven listeners, thank you for everything you do, and thank you for a really engaging and informative podcast today.

Paul Bambrick-Santoyo: Thank you for having me on your show, Brett. Thank you for having me on your show. It was a gift. Thanks so much.

Brett Roer: Thanks everyone.