99. Addressing AI’s Impact on School Operations and Ethics with Evan J. Harris

The following content has been automatically generated by an AI system from the audio recording of this podcast. We cannot guarantee the accuracy, or completeness of the information provided, and we apologize for any errors.

Tara: Welcome to Mindful School Marketing. I’m Aubrey Bursch. And I’m Tara Klaes. Today, we’re joined by Evan J. Harris. Evan Harris is the founder and president of Pathos Consulting Group. With over a decade of experience as an English teacher and administrator, he holds a master’s in private school leadership from Teachers College and has furthered his studies in tech ethics at Stanford’s human centered AI Institute.

At Pathos, Evan focuses on empowering independent schools by helping staff departments integrate AI tools effectively, balancing efficiency with ethical responsibility. Welcome, Evan. We’re so glad to have you here today. 

Evan: Yeah, thank you so much for having me. I’ve been looking forward to this for a while, so I’m very excited to get started.

Aubrey: I can’t tell you how excited I am for this conversation, Evan. I am so excited. We’re going to talk about AI. We’re going to talk about all sorts of amazing things. Um, but before we get, you know, really dive in, I’d love for you to share how you became interested in AI ethics and applications specifically for independent schools.

Evan: Yeah. So, you know, I think that probably goes back to that Klingenstein experience, which is, if you know anything about it, a program that tends to produce a lot of like heads of school and people who are very ambitious tend to get into that. I think I knew pretty early on that that wasn’t my track, but I did want to make a difference in some way in the independent school world.

And then I think when chat GPT hit, you know, two years ago, I just thought, Oh, this is my moment. This is going to really shift things. And there has to be some connective tissue between this emerging tech and what we know about independent school operations and really good pedagogy. And so maybe I could kind of come into that niche.

Um, but then of course I had as you can imagine, just a ton of imposter syndrome for, well, I’m not a tech person. I’m an English teacher. Like I am as far away from a tech person as you could possibly be. Um, so then I just spent the year creating my own course of study, which included the Stanford program, but.

Also just getting a ton of certifications and, and trying to teach myself Python, which is just, I was just as bad as it, as you’d think an English teacher would be. Um, but I made the effort to really learn the tech, um, side of it. And then just spent a ton of time actually on the ground in schools, particularly with like the staff and operations side of the school, because I felt like that’s kind of what we’ve neglected in our conversations about the classroom and rethinking assessments, academic integrity.

Um, and some of the more sensational AI topics we hear again and again about AI tutors and teachers being replaced, the stuff that, frankly, just isn’t real, um, I thought kind of lost in all that was, well, wait a second, like, what about the implications for the other half of the school who are doing a lot of the mundane, time consuming work that can actually be automated and who are actually, I think, most at risk when it comes to some of these, um, Um, These AI safety issues.

Um, and so I just kind of felt like there was this whole other conversation to be had. And I’ve kind of made that a niche of a niche that I’ve become obsessed with over the last over a year now. 

Tara: It really is a niche and I think super important. And I know that our audience, especially it’s going to be really interested in it because we hear a lot about AI, especially as marketing people, right?

How to use it for marketing, how to use it for communication and that type of thing. But. You emphasize the importance of addressing AEI’s impact on school operations rather than just like marketing or academic applications. Could you talk a little bit more about why you believe this focus is more compelling and higher risk?

I mean, what, what challenges do you see in the school sector? 

Evan: Yeah. So I mean, to start, you know, I think if you look at your average independent school and how they manage risk, it’s, you know, often the CFO, maybe who takes a leading role, sometimes it’s kind of a board run committee, but it’s those staff departments like HR and Marcom to the degree that they’re doing crisis communications, they’re really involved in the risk management process and what happens after an incident and how do we mitigate some of these things and think about like the tabletop exercises we need to run.

Um, and yet. You know, I, I’ll tell you a quick story, which is I just did a, um, an event at Woodbury Forest in Virginia in front of 40 CFOs for prep schools. And I said, raise your hand if at some point in the last two years, you’ve done some kind of training for your faculty in AI. Every hand goes up. I said, keep your hand up if you’ve devoted the same amount of time, attention, resources to the staff and operations side of your school.

Every single hand went down except for one. And so I think this also kind of gets to a larger issue about how we think of PD and where our PD resources go. Um, and so, you know, there, there is so much opportunity on the staff side because there’s so much repetitive, mundane, time consuming labor, and also so much creativity from the people who are working in these various departments.

Um, and because there’s often no acceptable use policy or no mechanism by which they can get some of their ideas approved. Often they just kind of think, well, I’m not sure if I’m allowed to do that or not with AI. So I’ll just either do it secretly or not at all. And then they are, like you mentioned, really at risk in a couple of different ways.

The one I can think of that’s sort of the most immediate is kind of the data privacy risks because they work with a lot of financial information, student information donor information, all kinds of really sensitive data. Um, some of the A. I. Use cases, particularly for departments like H. R. And admissions get into what’s called algorithmic bias, which I’m happy to talk about.

Um, but yeah, I mean, there, there are just so many ways in which that side of the school is more exposed on risk and stands to benefit more in opportunities. And so it kind of just has, it drives me a tiny bit crazy that it’s really 99 percent of the conversation feels like it’s about classrooms and for good reason.

Like there are plenty of things to talk about, but it’s gosh, can we have some of this other conversation too? Cause there’s so much going on. On the other side of the school. 

Aubrey: I love this conversation and I’m so excited. We’re having it because this is what I see in schools, especially the schools I work with is like there’s this uncertainty of how to proceed, right?

How do we proceed? Um, on that side of things. And then are there any guidelines or are there, are there anything in place that we should, we should know? Um, and there is such opportunity there. Um, anyone who’s like really dived into AI knows that the opportunity is there to streamline so many things. Um, so I’m curious, like if we could dive a little bit deeper into that data privacy and including personal identity.

viable information in school departments like Marcom and admissions, like, can you tell us some common concerns there and how schools can effectively address these issues? 

Evan: Sure. So, you know, I think, um, it probably makes sense just from the 10, 000 foot view to outline what’s the connection between data privacy and a large language model, like chat GPT, like what’s even the connection there.

Um, so if you input a prompt or a document into a large language model, there’s a very good chance that that that information is getting brought into its training set and that that information can then leak out into someone else’s conversation, maybe halfway across the world. And so I think a lot of staff at independent schools are somewhat aware of this and kind of have this sense of like, Hey, probably not a good idea to put the kid’s social security number into the large language model.

What they’re tend to be less aware of is something called the indirect identifier. Um, so in 2000, they did this study at Harvard where they found that the majority of Americans could be identified just using their date of birth and zip code. That more of us than not are unique by those two identifiers.

And so the indirect identifier is this idea that Two pieces of information, which taken by themselves cannot identify a person, can be used in combination to identify a person. And so we don’t necessarily think of that when we’re working with data. We think, well, I don’t have any names in this, I don’t have any, um, you know, like I said, social security numbers, these things that obviously, very obviously seem sensitive, and so it must be okay.

Not realizing that these pieces of information in combination can be dangerous. Um, I think, you know, one thing that I would recommend for schools just as a really good kind of first step is to work with your tech department to figure out some kind of established procedure for, um, you know, kind of either accepting or denying AI use cases and new AI tools that people come up with, because your tech department is going to be good at looking at like, okay, what are their data storage procedures.

What kind of encryption are they using? Like, what, what kind of access controls exist for this tool? And they, they’re the ones who are maybe in the best position to kind of give a thumbs up or a thumbs down on a new tool or idea. But schools are in this position now where they don’t want to just say no across the board to everything new because, like I said, people are coming up with great ideas all the time and discovering new tools all the time.

But they need some mechanism to stop something bad happening before it happens. And then obviously you need some training for people to even be aware that these issues exist. 

Tara: Yeah, I mean, that training is so key and I think, you know, you’re probably among the few people who are addressing it in this particular way and there are, there are so many traps and so many cool tools that seem like they’d be fun and unless you dive in and figure out how they could be, um, problematic, I think a lot of people are nervous about it and people like me are like gung ho and ready to go for it.

So I’m sure you’re balancing those different, um, You know, the different approaches, the different responses to the AI initiatives that are out there. Um, when you’re working like with, from a top down perspective with, um, with the head of school and, and the, the higher level administrators, um, do you have any strategies that you use to engage them and to to maybe address what they’re, you know, personality is, how they are, you know, responding to this stuff.

And, and what kind of support do you offer for the successful integration of these things? 

Evan: Sure. So the, the dealing with the head of school thing is a really interesting question. Um, I find because heads of school are very used to kind of knowing what they’re talking about. They’ve, I mean, especially sort of seasoned heads of school that have been around a long time.

They’ve had the kind of the same kinds of conversations over and over for decades, a lot of them, and they kind of have a good sense of what they’re doing. And then here comes this massive curveball where all of a sudden they can feel really out of their depth. And for a certain type of person, that can kind of feel intimidating, like a bit of a challenge.

Um, and so I think, you know, um, giving them some sense of, of agency and, and like, we can build a, a set of strategies that align around their values and their mission, um, and kind of work backward from there. I find that’s actually really, um, an exciting way to start with a new head of school is I’ll actually, and you guys might appreciate this as, um, being from our comm is I’ll start with.

A press release dated six months from now talking about everything we accomplished in the last six months and create this kind of vision for them of what do we want to be able to say is true at the end of our work together at the, or at the end of this first phase, um, and give them some agency in crafting that vision together.

Um, I think that’s a really fun and exciting way to start. I love thinking about this as an organizational development challenge and not just a tech challenge. I think it’s one of the things that Marcom can do really, really well is do storytelling. And when we think about storytelling in a school context, it’s typically about, you know, how do we tell really great stories to our donors about the impact of their gifts?

How do we tell really great stories to prospective families to get them interested and excited about applying here? Rarely is it thought of. As a internal tool, and I’m kind of maybe stepping on your next question. I apologize. But, you know, people are a lot more willing to go along with the change if there are clear indications that the change is already working.

And that is, I think, especially true in an emotionally charged, um, topic like AI. And that’s just as true for administrators and heads of school as it is for faculty. So I think that’s another piece of it is scoring those early wins. And telling those stories internally to kind of build up the sense, Oh, there is change coming, but look, it’s, it’s happening already.

And the sky isn’t falling. It’s actually positive. 

Aubrey: I’m so glad you hit on this. I think this is a really a key thing I hear from schools a lot is, you know, well, you know, how do we talk to our faculty and staff about this? How are we showcasing, you know, that this is working and that’s nothing to be scared of and that there’s, you know, not going to be this huge negative outcome if we head this way, right?

Um, and then there’s also, there’s that piece that you touched on, but there’s, then there’s also, um, I think a lot of fear. Around, um, getting the buy in from that external commute school community, meaning like your parents, like they’re, they’re part of your community, but like, they’re not like the faculty and staff.

Like, how do we talk about that too? How do we, you know, position your school using AI as. a potential positive for parents and even for the external community. I’m curious, like, what are your thoughts on that? And like, what have you seen to be successful or perhaps like, you know, things that schools should consider on that, that path?

Evan: Sure. So I love that question. Um, you know, I think one of the main reasons, if not the main reason that a parent chooses an independent school is safety. Right. Physical safety, emotional safety, a certain amount of assurance that things are going to turn out okay for their child. And so much of what we hear from parents when they’re stressed is really fear.

Right. And that it’s, even if it comes out, even if it looks like anger, it’s very often fear, um, is my child going to be okay? Um, and so I think when it comes to AI, the way that that sort of manifests itself is, you know, is my child going to be prepared? Are you integrating this in such a way that they’re going to arrive into college or into their first job with a sense of how to use these tools and how to use them ethically?

Um, you know, I’m on a panel on Friday, um, and we’re responding to some of the incidents that happened recently at Lancaster Country Day and, frankly, at schools all over the country with non consensual deepfake pornography, which has just gone rampant. And so there are these other fears, too, about kind of the media landscape and what are you doing to protect my kids from some of these dangers.

And so this gets into the risk management piece as well. Um, Um, is, you know, are we creating kind of procedures to make sure we’re delivering appropriate consequences to perpetrators to support, um, victims of this and to signal really clearly that there is an equivalence between fake images and real ones and that we take these incidents just as seriously as any other form of sexual abuse.

And, you know, these, these things are now popping up in the headlines and parents are seeing them. Um, the last piece I’ll say is. You know, we want to think, I think really carefully about those use cases where we don’t want to use AI in our communications because it interrupts what is really one of our core competencies as an independent school, which is our authentic communication and the trust that that builds.

And so there are places where I would say like, Hey, that donor just gave 10, 000 to the capital campaign. Write them a real email. Um, maybe AI checks the grammar or whatever, but you know, don’t, don’t automate that, um, or at least, you know, have a human in the loop somewhere for all this stuff. So I think we do want to be really, um, careful about those use cases where AI makes sense and, and where it hinders or interrupts.

What it is. We do so well. 

Aubrey: I love that you brought up authenticity because I think that’s key. Um, you know, I’ve heard, you know, where potential schools have gotten sort of in a sticky situation with parents where they feel. I mean, if you think about it, parents come to a school. Because, you know, they want that, usually it’s a personalized, you know, they want to be known by the community, they want their children to be known, embraced, loved, identified, you know, that the school knows their child really well, um, and yet, like, some instances with AI, like, parents might think, like, are they using this to reply to my emails?

Are they using this in, in a way where it’s part of the progress reports or like the report cards or something like that? So, and that can be a really, you know, potential negative situation for a school. So like, how, how have you seen schools navigate this balance or communicate about like this, this Like we want to empower our teachers and our faculty and staff who frankly are really overburdened in independent schools.

Um, and we want to empower them to be able to use this yet. Um, we also want to ensure that our parents, um, understand how it’s being used and understand the benefits to them. 

Evan: Yeah. So I, I, I love that you bring up the student comments example, um, cause it, it, our teacher comments rather. Um, cause like that comes up a lot.

And I actually, one of the first job offers I got from a school on contract was, Oh, can you build a bot that will sort of do this for us? And I had to sort of say, not a good idea. Um, would love to take your money, but can’t do that. Um, and I think it sort of starts with establishing why are we doing these comments to begin with?

And I know as a teacher, I was a teacher for 10 years. Yes, it takes in some cases days to write all these comments, but what happens in the process of writing them? It’s really a kind of metacognitive. I mean, it’s the mindful podcast, right? It’s a, it’s a mindfulness exercise about that student and how they’re doing and what they need, what you need to see from them.

And, and so often as teachers, we write those comments, or at least I did, and I would think, Why haven’t I said this to the kid? And if we automate that away, we’re losing that reflection time and that mindfulness time, and we’re robbing ourselves of that moment to really do nothing for five minutes, but think about one kid.

And that just doesn’t happen a ton. Cause as you know, it’s, it’s a marathon and a sprint all year long, and you just don’t get that five minutes. Um, where AI can be used, it can be used for clarity. It can be used for grammar. Um, it could even be used potentially kind of for tone issues, right? That would happen all the time where a student gets an A, but their comment is really rough or vice versa.

They get it and they failed, but the comments glowing and it’s like, this isn’t making sense. There are also teachers, frankly, who, um, don’t feel like strong writers. And so they don’t all come to the process the same way and might need different types of supports. Um, so I, I do think there’s a place for AI in something like.

Comment writing, but would suggest that it’s an opportunity for us to take a step back and say, well, why do we do this in the first place? Um, and how can AI support it without hindering what it is we’re actually engaged in here?

Tara: This is all fascinating to me. I think, you know, I am pretty far away from the parent perspective of this. And so the fact that this is yet another thing that parents are maybe thinking about when the communication is coming to them is really fascinating. fascinating and kind of frightening as well.

Like that’s where we are, that everybody’s questioning the validity or the legitimacy of the messaging that they’re getting. And as somebody who’s Who’s really becoming accustomed to using AI for communication. Um, it is such a fine line to walk, I think, because it can do a really good job in cleaning up a message.

You know, it can probably make your, your staff sound even better at times, you know, when they’re communicating, as long as they’re not just using it blindly, I think that’s really important. Aubrey, did you have another question you wanted to ask? 

Aubrey: Um, I’m just curious. I do think there’s huge benefit. I mean, I think about like, if you think about some of the, just like you were just saying, Tara, about the staff, interpersonal emails and stuff like that.

You know, if you type a staff email, like, sometimes you just reply, like, I’m remembering when I worked in schools and there would be the person who would reply. That was a very brusque email. Like, you knew their tone was off. But like, um, feeding that into chat GPT and saying, can you do this in a nonviolent communication tone?

Like, it’s amazing what that might do with communication. So there’s so many benefits, but I think my question here is like, so a lot of schools, you know, I know this is going to be a hot episode for a lot of our listeners, a lot of schools, especially our small schools who are kind of feeling a little overwhelmed, um, with everything AI.

They don’t know where to start, like, where do you start if, if like, where’s the first step, like the first step and like, what is like the bare minimum they can do, maybe that’s just something like, you know, this year we are going to do X, like. You know, what is it that they can do? Because a lot of, um, you know, both parents and faculty and staff, like, especially that younger generation of faculty and staff are coming in and being like, you know, we need to talk about this.

Like we do need to have movement around this. Um, and, and this is not going away that I could tell. Um, so where is, what are your thoughts on that in terms of like, just starting out? 

Evan: Sure. So, you know, one thing I’ll say about the small school that I think is unique to the small school is that I think in some ways they are more vulnerable to some of the AI snake oil and there’s a lot of it, right?

Um, and so if you see an application that promises that they can do the work of, you have to work with your HR department, we can do all the hiring and firing and, um, like resume filtering, like, I actually isn’t at all ready to do that, right? But if you feel like you are a department of one competing across town, you’ve got a department of five in your same department.

You might feel more willing to sort of believe some of that because it would seem convenient that that maybe this AI tech actually really does exist. Um, You know, I, it’s not a fun answer, but I think unfortunately it kind of has to be the answer, which is that the ethics and safety piece is the prerequisite for any of the fun stuff it just kind of is.

Um, and, and everyone governmental organizations, NGOs that have studied it they’ve all kind of come to the same conclusion. And so I think the first step is just to kind of look at some of these core safety issues. I would start with, um, the data privacy piece. I would look into algorithmic bias a bit, especially if you’re thinking about AI being used in admissions or in HR, that those are really sensitive areas for AI.

Um, but you know, again, it gets back to this thing about the change management approach and the organizational development approach, which is that scoring some small wins is a really great way to build some momentum. So I’ll give you an example specific to Marcom. One of the things I hear from Marcom teams all the time is we like using AI, but they.

Turns out this very bland sounding copy that doesn’t reflect our brand voice at all. Um, and your brand voice really needs to be consistent across all these different channels, because that’s, that’s really how you kind of engender a sense of trust and consistency, but you can upload your style guide.

Into a custom GPT that will automatically revise copy for newsletters and social media posts, especially if you have kind of that, like, you know, we sound like this, not like this kind of grid within your style guide, as I think a good style guide should. Um, because especially large teams communicating over a lot of channels, how often are they really consulting the style guide?

Probably not as much as they should. So this is a place where AI doesn’t. Kind of rob you of your voice, but it could actually keep you on track in terms of staying consistent with your brand voice. Um, so I love that one from our comm that’s a bot you could build in five minutes. It’s really easy. Watch YouTube, YouTube tutorial about making a custom GPT, upload your style guide.

Good to go. 

Tara: Yeah, I love that. Um, the style guide idea. I know that chat GPT, I played a little bit with creating your own bots and, and the way that it. Trained it trains itself just by using it too. Right. I mean, the more you use it, the more it learns your tone. Um, if you feed back in, do you feed back in what you’ve done, what you’ve added out of stuff 

Evan: increasingly has a memory and will remember everything you’ve said to it, which actually gets.

Pretty creepy. I asked chat GPT last month to guess my Myers Briggs personality type based on everything I’d ever said to it. And it guessed correctly, which is that’s weird, right? Like that’s, that’s weird. I’m pretty sure that’s weird. Um, it also, I asked it to do a portrait of me and had never seen a photo of me and it, it did it accurately.

And I said, how did you know I was bald? And it said, oh, nine months ago you were researching like a razor or something. So it remembers everything you’ve ever said to it. It’s a very creepy friend to have. 

Tara: Wow. Amazing. Oh, wow. Um, well, this has been so fascinating. I want to keep talking, but we are going to move on to a question about mindfulness.

And actually kind of, um, I wonder if, if AI itself is mindful. It sounds like it kind of is, but, um, so we like to talk about mindfulness on this podcast as part of our title. So how do you define mindfulness and how would you say it applies to, um, to this this topic. 

Evan: Sure. You know, um, I take and this is going to sound a little self indulgent, but it, I promise it’s not.

I take probably an hour long walk every day. Um, there’s a pretty long walk. I, that’s probably my most productive hour of the day. I get so much thinking done. And I was, I was thinking the other day about why is it that it’s so productive? Um, and I’m, I’m sure that part of it is that I just don’t have my phone and I’m, I’ve just unplugged.

Um, someone made a sort of a similar point about, um, like shower thoughts, like the thoughts that we have in the shower, there’s like a subreddit for shower thoughts, like how come all these amazing ideas just pop into our head while we’re taking a shower? Well, what is like one of the few rooms in the house where you don’t bring your phone with you, right?

So I do think there’s something just to kind of unplugging, um, and focusing over an extended period on, on a single thing, um, and not doing the kind of whack a mole thing that we do for the rest of the day. So that’s, that’s probably a part of how I would define mindfulness at least for me is a kind of, you know, a slowing down and an intentional, um, focus that lasts 

Aubrey: Thank you.

I love that. I love the long walks. I too, I used to listen to audio books on my walks and I realized that I was actually not giving my brain a break and now they’re much more creative when you go on the long walks, right? Um, so thank you for sharing. That’s really great. Um, we’re about ready to head into one of my favorite parts of the episode, which is our rapid fire questions.

Are you ready? 

Evan: Yes, let’s do it. 

Aubrey: Let’s rock and roll. All right. So if you could put one book as mandatory reading in the high school curriculum, what would it be? 

Evan: So I was a former English teacher. I thought about this question. I thought, Oh, I should, there’s some nonfiction book that would really educate our, and, and, you know, instill these students with a great sense of something, um, but I’m just going to pick my favorite novel, which is the remains of the day by Kazuo Um, it’s a beautiful book.

It’s one of the best books of the second half of the 20th century. It, you know, Ushiguro won a Nobel prize in literature. It won the Booker prize, the movie adaptation with Emma Thompson and Anthony Hopkins won an Academy Award. It’s a beautiful book. Every time I taught that book, all the kids loved it.

It’s an incredible, it’s about a butler. So it sounds like it’s going to be boring. I promise. It’s not, it’s very good. 

Tara: I love that. Thank you. Um, that’s great. I have not read it. I’ve seen the movie, but I’ve added it to my list as well. What is one app you could not live without? 

Evan: So I just, in the last couple of weeks, taught myself how to use CapCut.

Um, I don’t know if you all use CapCut. I turned myself into a YouTuber in like 48 hours. I’m doing zooms. I’m doing like key frames. It’s so fun. It’s really easy. Like all this, all these kinds of special effects that you see YouTubers doing. It turns out it’s actually very, very easy to do that stuff with CapCut.

Um, and so it just really ups your production. So if you’re a content creator or you’re making videos. For your school and you’re a little intimidated by like, you know, kind of upping the production value CapCut and, you know, not sponsored by CapCut. I’m really doing a good job of acting like I am, but it’s so easy and so fun.

And I kind of got addicted to it. 

Tara: Does it use 

Aubrey: AI? 

Evan: They have some AI tools. Yeah. 

Aubrey: Yeah. Yeah. That’s awesome. Um, this comes up every time, um, I do a small school leaders meetup monthly. Every time someone will ask, what are you using for like video editing? What are you using for like anything creative or content creation?

And this comes up too. So I’m glad you mentioned it. And, um, I have not, I played with it, but I haven’t dealt, you know, kind of really David and like, yeah, played, played with it. So I’m putting that on my list to do over winter break. So thanks for the reminder. Um, what are you reading right now? 

Evan: So right now I am just buried in policy papers that I got from an organization called encode justice.

And I know I should talk about a book I’m reading, but I kind of wanted to use this as a chance to name drop encode justice because I think they’re so incredible. Um, so I’m doing a couple of projects with them at the moment, but encode justice is a Gen Z led. AI policy think tank. They have 40 chapters all over the world.

They have about coming up on 1500 members. Everyone on the leadership team is under the age of 23. Um, and they have passed legislation all over the world. I went to an event on Capitol Hill with Dick Durbin and Amy Klobuchar and Senator Cruz. And they’re like, you know, They are pushing legislation for AI safety all over the world.

And they’re doing it because they’re like, Hey, we’re the generation most affected by this stuff. So we’re going to lead the charge. Um, and so I just wanted to kind of, if anyone’s looking for resources on AI safety, I think in code justice does an incredible job and the fact that they are Gen Z to me, like, I just think they’re the most impressive people I’ve met, you know, since in recent memory, they’re, they’re incredible.

Aubrey: So I have a follow up question because I’m so intrigued. So I’ve not heard of this organization. I’m thrilled. So both as a parent, so as a parent and also as a son, I have a son in middle school and like, I work closely with their school, their middle school and high school. Um, I’m curious, is this a good avenue for those both educators and parents to kind of dive in and kind of have an understanding of things and get some Some information around that 

Evan: a hundred percent.

So they’ve, they’ve got a lot of resources. It’s also a really good Avenue for a young person that wants to get involved in this, but isn’t maybe quite sure how they can do a fellowship through and code justice or start a chapter at their school. Most of the chapters are either at high schools or colleges.

Again, it’s all over the world. They also come into schools and educate. They’ve done. I forget how many workshops, but they’ve educated 30, 000 young people over the last two years in AI safety. So it’s young people teaching young people about these issues and especially on the, um, non consensual deepfakes issue, they’ve made a ton of progress.

The legislation that’s working its way through the Senate right now. is largely due to their efforts. So the fact that they decided like, we can’t wait for adults to save us. We’re going to do it ourselves. To me, they’re just like the coolest people. I know. 

Tara: Very cool. I love that. I will put that in our show notes for sure.

I will put a link. I’ve just looked it up. So thank you for sharing that. Last question. What is one great piece of advice that you’d like to leave with our audience? 

Evan: Sure. So, um, I used to work in the music industry before I was a teacher. Um, and I, I think I, I, I made some mistake. I used to work for a booking agent.

I was 19 years old and had made some egregious mistake with a tour booking or something and felt so embarrassed. And the agent that I was working for, um, just kind of pulled me aside and this is very simple advice, but I, I still think about it all the time, which is he said, um, you know, everyone makes mistakes, just don’t make the same one twice.

If you don’t make the same one twice, you will inherently get better. Um, and that was like the most liberating thing in the world for me is just kind of letting go of this idea of being perfect and just kind of realizing like, yes, I am 100 percent going to make mistakes. The only thing I’m responsible for is not making the same one twice.

And I will, by the, by the nature of that, by the very logic of it, I will improve. Um, and so I just found that really liberating. And I think for a topic as convoluted and scary. As AI giving yourself permission to not really know and to, and to kind of, you know, be like, yeah, of course I’m going to feel like an imposter.

This is not my area of expertise, but I’m going to try anyway, because it’s important. Um, I think that’s really liberating. So 

Aubrey: that’s amazing. Great advice. Um, it’s like the beginner’s mindset. Like we’re, we’re all learning. We’re going to make mistakes. That’s part of life. That’s what I tell my kids all the time, like, but the same mistake twice is a really powerful advice to give.

Thank you so much. Thank you. 

Evan: Yeah. Thanks for having me. This was fun. 

Tara: Yeah. Where can people find you online? 

Evan: Sure. So, um, I’m pathosgroup. ai. I’m Evan Harris AI on LinkedIn. Um, and I’m actually working right now within Code Justice on an online course specifically for independent school staff that covers some of these AI safety issues because we just kind of felt like it didn’t exist and it needed to.

Um, and so I’m, that should be rolling out I think in the next 30 to 60 days. And small schools, we’re going to make it almost free for them just to make sure that no matter what sort of the PD budget is for your school, like you will have access to this. 

Aubrey: Great. That’s amazing. Wow. Yeah. Much needed and much appreciated.

Tara: Yeah. Thank you so much. It’s been great. Thanks for having me. Great chatting with you. Chatting. Bye. Bye. Bye.

And we’re clear. We are. Well, oh, there it is. Oh, it moved.