In this episode of College Admissions Decoded, host Eddie Pickett is joined by NACAC members Emily Pacheco and Sebastian Brown to explore the impact of AI in higher education: from students using AI tools for essays and test prep, to admission offices using AI for student outreach and data analysis. Our guests provide insights on the opportunities and complex challenges that artificial intelligence presents, including questions about privacy, ethics, and access.
AI in College Admission: How is artificial intelligence impacting students and professionals?
In this episode of College Admissions Decoded, host Eddie Pickett is joined by NACAC members Emily Pacheco and Sebastian Brown to explore the impact of AI in higher education: from students using AI tools for essays and test prep, to admission offices using AI for student outreach and data analysis. Our guests provide insights on the opportunities and complex challenges that artificial intelligence presents, including questions about privacy, ethics, and access.
Guests: Emily Pacheco, Assistant Director of Undergraduate Admission at Loyola University Chicago, and Sebastian Brown, Regional Admissions Counselor at the University of Oregon
Host: Eddie Pickett, Senior Associate Dean of Admissions and Director of Recruitment at Pomona College.
___________________________________________________________________________________________
Eddie Pickett: Hello and welcome to the College Admissions Decoded podcast, an occasional series from the National Association for College Admission Counseling, or NACAC. NACAC is an association of more than 28, 000 professionals at high schools, colleges, universities, and non-profit organizations. as well as independent counselors who support and advise students and families through the college admission process.
I'm your host, Eddie Pickett, and my pronouns are he, him, his. I'm a longtime NACAC member and former member of the NACAC board of directors. In my day job, I'm a senior associate dean of admissions and director of recruitment at Pomona College in Claremont, California. Thanks for joining us. In this episode, we explore the evolving role of artificial intelligence, often referred to as AI, in higher education with two co-leaders of NACAC's AI Special Interest Group.
Together, we dive into how AI is impacting the work of college admission professionals, from streamlining repetitive tasks to enhancing communication and decision making. We hope you will gain a clearer understanding of how AI can empower higher education professionals. As well as its limitations and ethical considerations.
Whether you're new to AI or already experimenting with it, please join us. We're happy to welcome Emily Pacheco, Assistant Director of Undergraduate Admission at Loyola University Chicago. Welcome, Emily.
Emily Pacheco: Thank you. It's excellent to be here.
Eddie Pickett: Next, I'd like to welcome Sebastian Brown to the podcast. He is a Regional Admissions Counselor at the University of Oregon. Welcome, Sebastian.
Sebastian Brown: Thanks, Eddie. It's nice to be here.
Eddie Pickett: As a football player, it's also nice to be number one right now. So congratulations, Mr. Brown.
Sebastian Brown: Appreciate that.
Eddie Pickett: Thank you both for joining us. I know it's going to be a great episode. So let's hop in. I think the first question, just how do you define artificial intelligence?
Emily Pacheco: Yeah, great. I can take this question. I would define it as referring to the computer systems that are designed to stimulate human intelligence. This is a new system that can do things that previously only humans could do. So for educators, AI can really assist in a lot of different tasks, like automating grading, and personalizing learning experiences, and analyzing student performance data.
It does things that previously just required a lot of human input, so it can really enhance creativity and streamline processes. So it's just a really useful tool and has so many uses that we're just starting to explore in education.
Sebastian Brown: Yeah, I'd love that definition, Emily, and I think the only other thing I would add to it is a lot of folks are sort of conflating or using things like machine learning or automation interchangeably with AI.
And I think one of the biggest differences is that at some point like AI starts to almost learn from itself, right? It can be a little bit more adaptive. It's not just a really long set of code or strict, stringent rules that you're running consistently. And I think that's where we start to see more of like the generative AI piece.
So the fact that it can learn as it goes or sort of teach itself down the road is one of the important aspects. That's not just the email inbox where you're giving it a rule and it's setting aside certain sender into a folder. That's very much like an automated process. AI kind of takes things a step further.
Emily Pacheco: Yeah, that's exactly right. The incredible thing about it is each time you ask it to do something, you actually get a different answer. And so I think that's what you were mentioning there, Sebastian. It's pretty amazing that each time you're putting input into it, it's learning from what you're doing. It can also be frustrating because it doesn't repeat the exact same thing that it did last time, but it is continually learning from what you're doing with it.
Eddie Pickett: Excellent. How is it impacting the lives of the students we serve currently?
Sebastian Brown: One of the things for me with AI is just how quickly it's implemented into the lives of the students. Trying to think back when, like, I was in high school or in college and if there was something that really shook both our, like, social and educational areas as quickly as AI.
I think like a lot of the questions around the ethics and the authenticity of who is using AI and for what reasons. And I know and appreciate a lot of the work that Emily's doing right now around. Trying to prevent people from being afraid of AI and really how to use it as a powerful educational tool that students and admissions professionals can leverage.
Emily Pacheco: Yeah, I think what we find is that students are definitely using it. Whether they are talking about how they are using it is another story, but students are exploring these tools. They have them readily available to them. Unlike previous technology, students didn't have to purchase anything different to jump in and start using chat GPT.
The free version was rolled out and they could jump right in, and most students already had a computer after the pandemic. So there really was a very low bar to entry to use these tools. So students started using them. Unfortunately, what we've found is educators have been a little slower in adopting usage. And that just has to do with age, I think.
The younger you are, the more flexible your brain is. And so you just kind of are willing to try it and are open to the various usages. But what I've found is the more years of experience that an educator has in their role, oftentimes the less likely they are to have engaged with these tools.
And that's because they've already established systems for what they do. And so they just don't see the need. What I'm doing works. Why would I? incorporate new technology that requires me to learn and think in a new way. And so that is really, I see kind of a rough place where we are because students are using it both ethically and also in very unethical ways.
And because they don't have a place to talk about that, what's happening is that it's going silent in the classroom. Students are turning in assignments, maybe educators, teachers, counselors might have an inkling that something is, wow, the student's writing is improved dramatically in the last year, but they aren't quite sure how to approach that.
And few educators are creating safe spaces where students can talk about how they're using it. And so I really think the next step needs to be to listen to our students. What are you doing with this? Show me how you're using this because I think from looking at how they're using it, we can look at encouraging those ethical ways to use it and also say, “Hey, this is bypassing learning.” That's not what we want to do here and help have those conversations with students so we can help them navigate this new technology.
Eddie Pickett: Yeah, both of you said a couple of words I'm going to pick up on there. So Sebastian said afraid, and Emily said fear. And I work in the admission side just like you two, and we hosted a bunch of counselor programs this fall.
And in every single one of them, people asked about how we were using AI. And it was usually out of a fear or a negative instead of a positive. And you both were talking about it from a positive lens. So what would be your pitch to the educators to get on board with using AI in ethical ways?
Sebastian Brown: I love that question.
I mean, Emily and I joke like, yes, we're very AI positive minded folks, but AI, if you really boil it down and use it, a tool like spell check. I used to always tease that when I was in like third grade, my handwriting was terrible. And I'm left-handed and writing in pencil and like smearing. Like, I almost didn't pass third grade because I couldn't write in cursive well enough.
And I was really excited to work with computers in the future because of things like spell check. I wouldn't have to worry about even just legibly writing things, typing it out. That's never going to be a concern. And I think for a lot of folks with AI, again, leveraging it as a powerful tool, there's the education piece that we need to start with.
I think a lot of the counselors, for example. They want to give the right advice to their students. They don't want to be that counselor that says, “Oh yeah, you can use AI to write your essay.” And then that they apply to a school that has maybe a more restrictive AI policy. And that is a reason for not being admitted.
A counselor doesn't want to put a student in that situation. I think that's where universities need to step up. And they need to be much more transparent about their AI usage, their AI policies, so that students know what they are and are not allowed to do during the admissions process.
Emily Pacheco: What I think is important is that we bring educators to what I call the aha moment.
And me and Sebastian have had a number of these together and this often happens in a conference when we share with them a pretty basic usage of AI and it shows them how it could really make their professional life better, easier. It could make them what we call a superhuman, kind of give them extra powers in their professional life.
And that is a really important thing. Because before that moment, they don’t see how these tools could be useful for them. And if you're hearing a lot of the hype about it, and then what has often happened is, they've taken a Google prompt and they've put it into ChatGPT. The outcome is obviously terrible because you can't use ChatGPT like Google. And then their conclusion from that is that these tools are completely useless.
I've tried them and what I got was so awful and so unusable, I don't see any place for this for me. And often they'll follow that up and say, maybe other people can find places to use this, but there's not going to be a place in my professional life. And we've been able to show them one of the apps that we like to use for this aha moment is called Gamma.
I recommend anybody that hasn't played with AI and wants to start, go to Gamma.app. Create a presentation. It takes one line. You put in one line, you say, “I need a presentation. It's going to be for high school juniors who want to study out of state. I want it to talk about financial aid,” and that's all you have to do.
You give it, how many slides do you want? Do you want a lot of text, a little bit of text within a minute you will have a great starting point for that presentation. So that's one we love to show because when educators see that, they're like, wow, I either A, don't make presentations and say no when people ask me for a presentation, or B, I spend 15, 20 hours doing what you just showed me it could do in five minutes.
So that is a really basic step there, because after that, then they start playing with it. Then they go back to their desk. Usually when we've presented, Sebastian and I, we have multiple people come up to us afterwards and say, I'm so excited to go back to my desk and try what you showed me. And I want to take that and use it in a project that I'm working on right now.
I can see the usage. of this for me today for something that I'm going to show my boss on Friday. So I think that step helps to start to diminish that fear. Because if you can see the usage and that there are positive ways to use these tools, I think your fear starts diminishing. And then beyond that, it's exactly education.
It's talking about kind of making sure that we're all on the same page, that there are giant ethical concerns with these AI tools. We're not just fanatics, just pushing this on everybody. We really believe we need to talk about when you should use it and when you shouldn't use it. And we need more educators to help us figure out those good ways that it can be used and to help get the message out there to other educators, to other students about how it should not be used to bypass the learning.
Eddie Pickett: Sounds like multiple PhD programs or dissertations right there happening, or at least starting. I love how you talked about just go play with it. As children, we learn by playing, you know. As we become adults, you learn by reading. And there are lots of different ways of learning. And so I love how you said, you just have to go play and try it.
Some things will work, some things won't. So I really liked that idea. And thinking also, when it comes to college essays, people are always like, well, hmm, like it's just writing it for them. And what you said there, it's a good starting place. There's no way we can guarantee that a kid was writing their essay before.
We can't guarantee that because we're not sitting there with them. And so it's not about that piece. It's more about like, is this the authentic piece of the applicant's voice? So it could democratize help, realistically. So that's one way of seeing it.
Emily Pacheco: Yes. Yeah. A hundred percent, Eddie. Students have for years been getting outside help with their essays.
I'm pretty sure from the first essay that was ever turned in, there has been help there, either from outside sources, IECs, from excellent high school college counselors. But as we all know, the masses have been left behind in this process because the average student doesn't have access to quality college counseling.
And so these tools are not going to replace humans in this process because if you have a choice, you'd choose a human to work with. I think the majority of parents want that for their students, but the majority of students just don't have access to it. to this. So that's exactly right, Eddie. We have a lot of hope that this could help democratize quality college counseling, which is up until this point has just been super, super expensive.
We just can't possibly provide every student with a quality human. So this can take a human and be able to have that person provide services to a larger number of students or tools that on their own can mimic really quality college counseling. So that's the future that we see and the hope that we see in these tools for sure.
Sebastian Brown: Yeah, I was going to say, Emily, it is ironic. We encourage students to seek outside resources when writing their essays. We tell them, don't be the only person to have read and written your own essay. And then when the AI, you know, conversation comes into town, then folks sort of change their ethics around that, right?
And I remember chatting with a group of students. I love the term, write your ethics in stone, but your opinions in sand. And I think this is one of those things that as something so profound in this technology, like there are certainly some folks that are going to be very concrete in their beliefs in terms of who gets access, right, to college counseling.
But I think. A. I. Is powerful enough to kind of push past that. One of the things I love, and Emily alluded to just having fun with this part of the process, to write your essay shouldn't feel as dreadful for as many students as it actually is. It's an opportunity for them to really explore parts of themselves.
I tell a lot of the students that I work with that they need to do that internal search before they do this external search. They need to have a better understanding of who they are, who they want to be, and then finding that right college fit based on the sorts of decisions. And I love AI and I know my parents wish that it was a thing when I was a child because I was very much what I've called like a wild child and that's W.H. Y.
I was that kid was always asking, “Why this? Well, what about that? Can we take this direction?” And one of the amazing things about AI is it doesn't have any sort of patience to break. You can continually pester it and let's take this a new direction and it will always meet you where you're at. And I think for a lot of students that feel the time crunch when it comes to things like the essay, they just haven't put in a lot of the effort or the ability to further develop some of those really complex ideas.
AI can kind of take them into that next step in a shorter amount of time. And I think that's where it allows for a lot more of that creativity. We really, truly get to know students a bit better because they fleshed out those ideas further. If they're doing it properly, there is sort of that incentive for students just to get the first rendition of their essay and be like, okay, that's enough.
I'm done. I always share the story with Emily and other folks. Last year, I was a part of an essay writing workshop at a private school in California and sitting one on one with the student, reading their essay and giving them all these praises. The first paragraph is beautiful. It's poetic. It's well structured.
And then the first sentence of the second paragraph says, “As an AI model, I cannot accurately convey what it means to love.” And kind of in that moment, I see the student's face just get real red, and it's like, you can have the smartest technology in the world, but if you're not proofreading what it's giving you, like, that's on you.
And I think that's, even those kinds of things where students need to realize, okay, this is something I still need to be a, have my hand in this process, and it's still my writing, it can't do everything for me.
Emily Pacheco: Yeah, it really should be used to help with the process, not for the output. And that's what I constantly stress to students.
If you are putting a prompt into AI and asking for it to create the product, then you are not doing the learning. And really, you're cheating yourself. And I think the majority of students understand that. That's not a hard concept. They understand that the point of learning is that eventually they hopefully will be in a happy career and have a happy, successful life.
And using this tool to just do that learning for them is really not going to help them get to their ultimate goal. And so, when you talk with them about how these tools could be used for the learning process, they see that. They understand that. And I think cheaters will be cheaters.
We're always going to have that very similar small number of students who will use these tools to try to cheat, but the majority of students, if we guide them and show them how we can use these and great ways to help with the learning process, then we are going to have really great outcomes with this.
Eddie Pickett: Excellent. So we've talked about the students so far. We've talked about the counseling side, but the three of us all work in admissions offices. So let's turn a little bit to the admissions offices. Emily, you gave one example about the presentations in gamma.app. But are there other ways that. AI is impacting higher education and the work of college admission professionals.
Sebastian Brown: Yeah, absolutely. I think this is where a lot of my like fun interests come into with AI, you know, that enrollment management piece. We talked so much about data and a lot of folks use the term like data driven decision making. It's kind of a big buzzword, especially in the higher up VP director levels. And I think Emily and I have really tried to push folks more towards the data informed decision making.
It's still a very human decision-making process, but it really allows us to splice through data in a much faster fashion. It's incredible what AI is able to pick up that sometimes in really large data sets, the human eye might miss. Even if you have some of those beautifully coded Excel spreadsheets that something may go under the radar.
I'll give an example. Emily and I are both a part of the RACC network, so the regional admissions counselors of California. And we surveyed our membership. And we were able to identify pretty quickly that, hey, someone that is supervising is getting drastically underpaid. And we should really reach out to that person and help them advocate for better pay within their institution.
And I think that's something that had been, I mean, they'd been in RACC for a long time. And that was something that was sort of always under the radar because when we're looking at a couple dozen to maybe 100 responses. It just didn't really stand out right away. But I remember asking an AI model, identify anything that seems either kind of on the high or on the low end, and it was able to pick that up pretty instantaneous, which is great.
Emily Pacheco: I've seen it a lot of interest in several areas on the higher education side. One of those areas is definitely with chatbots and a lot of schools that have rolled out these pretty impressive chatbots. And we all think back to the days before AI and chatbots weren't good. It was very obvious we're talking with a computer.
And pretty quickly when you were asking it questions, it got to the point where it just was not helpful unless you were asking it very basic questions. That is not the case anymore. Universities like NYU has an amazing chatbot that they're using and that is there 24/7 for students and it's not pretending like it's a human.
It makes it very clear up front that this is a chatbot, but the way that you can converse with this, the way you know, that the depth of the questions that you can present to this chatbot and the really quality answers that it provides; It's just a whole other level of recruitment. Previously, we were around 8 a.m. to 5 p. m. Monday through Friday.
And at a lot of universities still to this day, if a student sends a text to the admission office on a Friday at 3 p. m., you'll be lucky if you hear back from that before Monday or maybe even Tuesday of the next week. So we know that at the pace of technology right now, that's not acceptable for students.
So there's been a lot of ways that AI has just made it so that we can interact better, quicker, and provide quality answers, much, much faster for students. The other place that I've seen it being used is definitely in how we are marketing and who we are marketing to. So previously we purchased large lists of names, and they were just kind of a blanket approach.
Let's try to find every 17-year-old kid who's out there and market to them. But now we can dive a lot deeper with AI for the reason that data is just much easier to analyze data. And because of that, now you can find students who are a much better fit for your university. And so you can allocate more resources towards a smaller number of students where you can market to them much deeper and provide much more meaningful marketing materials to them.
You can just do it much better now. And so I think what we're finding is we're finding universities who are doing that, but we have a lot of universities who have not yet delved into how that's done. So I think we're going to really see in the coming years how that plays out for some universities who are just really going to excel in marketing because of the use of artificial intelligence.
So those are two definite areas that I'm hearing a lot about. I'd say the biggest area of interest for those outside of the enrollment world is definitely, are you using this to admit students? And that's the fear that we, we have students using AI to write essays that are then being graded by AI. And then the decision about whether we admit the student is made by AI.
And from what I'm finding, that's not the case. And I think most universities are very, very, very far away from that. They might be using these tools in portions of the application process. There's a lot of data that we have that can be sorted that we get through applicants that can be much more easily sorted with AI, but it's definitely not being used in the majority of universities to actually make that decision of who would be a good fit for that school. That is still being made by humans.
Eddie Pickett: And as you think about just making decisions, that's the big decision is to be admitted or not to be admitted. But there are also multiple tiny decisions that we make daily and as admissions officers. And so in what ways can AI help admissions enrollment leaders make better decisions? Sebastian, you gave an example about the salaries earlier, but there are other examples of that.
Sebastian Brown: Yeah. I think one that folks are starting to lean into. It's certainly like a passion of mine. Anyone that's ever talked to me probably can't go more than five minutes without hearing me talk about mapping and how we visualize spatial data. But I think that's something that is really empowering for even our like newest admissions counselors, folks that are brand new to the profession, where if you are using AI tools to help them better understand, not just the numbers, but the people, the students that live in their geo-markets.
We can be much more effective and efficient when it comes to things like travel planning. You know a lot of folks that start as those road warriors certainly wear that badge of pride and how much they can do. And then we all look back on it and I'm like, “Why did I put myself through such an arduous fall travel schedule? Why did I visit 160 individual schools in just a few weeks?”
I always am really happy when I can plan truly sometimes six or seven schools in a day, because I feel like I've perfected my efficiency, but I'm also looking at which way traffic is flowing in San Diego and how easy is it to get in and out of this parking lot?
And I think when you start to allow AI to help with some of that decision making and it's cutting through sometimes what we carry is a lot of biases in certain areas. It can be a little less influenced than we are You can now have a one or two year experienced admissions counselor operating with five or six years’ worth of experience.
And then the folks that have been doing this in that five to eight year range are now starting to make almost more of those associate or director level decisions when it comes to things like travel planning. And it kind of just keeps moving the spectrum further down the line. And I think that's one of the really exciting things. We're not quite there yet, but I think a lot of institutions are starting to head into that world.
Emily Pacheco: I've seen it really useful in just helping to approach day to day problems. And I'll give an example of this just from this last week. I received an email from one of the directors in our office. They had a problem, and they wanted to present to us, and they wanted us to kind of brainstorm solutions to this problem.
We have Saturday visits and right now we have some students who are both admitted and students who are not admitted. So we'll have juniors who will be on a Saturday visit, and we'll have a student who's been admitted. And up until this point, we haven't treated those two students any differently on a Saturday.
And we saw this problem because the student's admitted and we want to kind of cultivate that relationship in a different way. So the supervisor sent this out, said, “What can we do? We don't want to spend too much extra money. We don't want to put another person who has to work on a Saturday, but what could we do to make it special for that admitted senior who's visiting on a Saturday?”
And I first read that, and I was ready to close the email because it was kind of an optional call out. It said, you know, “If you have some time, can you think through this? I'd love to hear your suggestions.” But then I remembered that I have this intern next to me, this amazing tool, chat GPT, I gave that problem to AI.
And I said, “Here's the problem I need to brainstorm. Give me eight solutions. What are eight possible solutions to this?” And I gave it a little bit more details about the way we operate, but not much. I basically took that email, put that in there, added a few extra details. It came back with three terrible ideas and four excellent ones.
And I took those excellent ideas. I did brief read them. I pasted those into the email and within about 10 minutes later, I responded. The next morning when I woke up in my inbox was a response from the director profusely thanking me for those ideas and just saying that was just so, so helpful.
I will preface it. I did say in the email, I said, “I used some really great tools to help me come up with these ideas.” I didn't want to take, they weren't fully my ideas. But those ideas, I don't think it mattered in the end. I don't think that director cared where those ideas came from. She saw that these could really help solve this problem.
And we're working on implementing one of those ideas now. So that is something I wouldn't have done that. I would not have taken that step. I didn't have the energy at that point to do that. But with these tools, it kind of took me above and beyond and made me a much stronger employee at that moment, made me much more valuable to my institution because I had access to those tools.
Eddie Pickett: Yeah. What are some of those tools that college admissions professionals should familiarize themselves with?
Sebastian Brown: That's a good question. I feel like this could almost be a sponsored podcast, right? Insert name here. Could at least get some money out of it. I think, and it really is becoming quite a flooded market right now, sort of that ed tech space.
And there's a lot of companies that are saying that, like, we're an AI company, and I think they need to do a better job of fully explaining in what ways is AI a part of their process. We, you know, had a great demo during our SIG membership meeting at the NACAC conference in LA and CollegeVine had a great example of something that is an AI tool that can help students and counselors.
I had a conversation with Score the other day, and it was really cool to see that on the high school side they have an implementation that's allowing counselors to sort of see an entire ranked list of students based on the types of schools that are applying to, and they've culminated all this data that helps with starting that conversation around expectations for that student that's maybe below a 3. 0 GPA, but really wants to go to an Ivy League school, right? And like, how do you talk about what is also changing expectations from institutions. As our institutional priorities change, how is that going to affect students?
Eddie’s like, “God forbid if Oregon wins the national championship, like we're going to see applications skyrocket. And that as no influence on my work as an admissions counselor, but it's certainly going to give me a lot more to do in the spring.
Emily Pacheco: Uh, yeah, I would definitely recommend if you've never touched these tools, a great place to start is with Perplexity.
That's not where I started and I actually use ChatGPT more because now ChatGPT and me, we have a relationship. It knows me. I can ask a question and I don't have to give it much background because it kind of has all of that. But if you haven't used any of these tools yet, I would recommend, and you're an educator, I'd recommend starting with Perplexity.
I do love how Perplexity gives you the links of where it's going. taking this information from. I think it does a much better job of sourcing, providing those sources to you. But yeah, those are excellent tools. You can start with the free version. There's no need to pay anything to use those tools. But there are many additional tools that are popping up and more every single day.
Some of the ones I've already mentioned, Gamma being a great one to start with. Some of the ones specific to college admissions, definitely CollegeVine has a number of quality tools. If you're looking for help writing a letter of recommendation, I recommend checking out their tool there.
If you're looking for some help with essay writing, I do like essay.aiesai.ai. They have some pretty good tools that help students with that process in an ethical way. There's also tools that are coming out that are mimicking that college counselor from start to finish, where it's nuts to bolts, everything that a student would need.
They could use these tools starting even as a freshman, sophomore, junior, and it would walk them through that process. Some of these tools are being used by counselors as well. Two of them that I talk to often is collegio.ai and Dabbl. D A B B L. Both of those are up and coming apps that I've seen a lot of really cool usages for them.
I'm also really hopeful that these can be tools that could be offered free or at a very, very low cost to students. So those are some that I recommend checking out, but definitely if you're just looking to start jumping into perplexity is a really fun one to start with.
Sebastian Brown: Yeah. And Eddie, I think I'll add that.
As Emily and I, like, got started with all this, it was sort of almost like the iPhone versus Android. And maybe there's that one weird person that still somehow has a BlackBerry. Folks are starting to tie in pieces of their identity with, like, which AI model they prefer. I like to use a tool called Claude, and I prefer it because Claude is a little bit more of a closed network.
It doesn't necessarily, like, pull as much or can't be influenced as much by things like Google, right? And kind of the open internet sources. So everyone has their own preferences. And I think it's important for folks to also shop around and stay current with the news. I know Emily and I get a lot of newsletters on just kind of what's happening in the AI sphere as it relates to education, but also you said just our day to day lives.
So there's a lot of great companies out there and I think they're all very niche right now. There's Element 451, and they have a great chat bot feature, and I'm sure we're going to build other things to compliment that, but it's important to keep an open mind and not necessarily just stick with one tool but see what's out there and what's developing.
Emily Pacheco: I also need to make a shout out for Notebook LM. That's an amazing, very interesting tool to play with. I took all of my presentations, so I took the PDF slides from the presentations I did. And then I took my notes and I put it up into Notebook LM and it created an eight-minute podcast. And it was insanely good.
I mean, it was an excellent thing that I could provide that to people and provide a pretty good idea of what I did at the conference. So I recommend definitely playing with that. It's a tool where you can upload documents, you can give it websites, and you can ask it to create you a guide. I also used it. I put all the HR stuff that I had from my employer, and I created a document where I can just ask a question and it finds the answer for me. And I used to go through, you know, ten websites trying to find answers to my HR questions. So definitely check out Notebook LM.
Eddie Pickett: As you're implementing these tools, what are some of the common challenges staff encounter?
Sebastian Brown: I think Emily's alluded to this quite a bit. We're in this newer phase. I think just getting started. Getting over that initial concern or what is this going to do with the data that you're feeding it. And kind of as we were talking earlier, you can start to implement it a little bit in your own personal life if that feels a bit safer.
I've used AI to help plan my brother's bachelor party and get creative with some ideas. Or the first time I remember being a regional with Oregon. We have something called the OPU, the Oregon Public University sort of tour. So we invited what was over a thousand California high school counselors to learn about all of the public schools in Oregon.
And I was in charge of sending out that email communication. So drafted everything out. I've sourced all these email addresses that go to click send. And I get that email from our IT department that says we've sort of been blocked out because it looks like you might've been, you know, compromised.
About 300 of those emails went out and 700 didn't. And I didn't know to which counselors had that actually gone out to. And so I remember getting the server email and it says these are all of the emails that it didn't go out to. And this was the error code. Realizing in that moment, it was like, “Oh my gosh, I have to cancel every meeting I have for the rest of the week to be able to essentially go through and copy and paste every individual email and put it into a different list.”
And that was one of the first times is said, “Wait, maybe I have this amazing assistant that can do that for me.” And went to an AI tool and said, “Take this list. And pull out either all the emails and or delete the codes that said this is the error code like remove this sentence,” and it was able to do that in less than a minute. And now all of a sudden, I have all that time the rest of the week to continue doing much more important things as an admissions counselor, and reaching out and connecting with my students. I think those are some of the things. Just giving it a chance and trying it is really important Emily and I have talked a lot about I think prompt engineering is sort of the next phase of this. Being very intentional with what you're giving the AI tool so that it meets your expectations.
Like Emily said earlier, a lot of folks go to it with that one sentence, Google sort of question, and that's not fully harnessing the power of the tools. And it could get rather complex. Jeff Neal, who's an amazing resource, has this amazing prompt but it's a couple paragraphs. He's put in all the work to create that.
And now folks just can copy and paste that into an AI tool, and now they're getting some list building recommendations from an AI model.
Emily Pacheco: I would say the biggest problem with implementation right now is that there is a lack of implementation. I think really it's a lot of people just kind of playing with these tools on their own, and a lot of people are lacking assistance in learning how to use these in the appropriate ways.
And so it's kind of a scary place because when you have everybody with a computer and you can just upload whatever you want into chat GPT, you can imagine that there's sensitive information that could easily get shared there. And when you have faculty and staff that haven't received any training on this, I think we all know that's happening and that shouldn't be happening.
So what really needs to happen is that people receive more help in that implementation stage. There are a few universities that are doing this really well. Georgia Tech has done really great things. They’re talking about what they're doing as well and sharing some resources about that on how they've implemented it.
Yesterday I attended an incredible webinar by the University of Texas at Austin. They put out an AI guide for their faculty and I posted it yesterday on our LinkedIn to share that with counselors and other educators because I see a lot of places where even though that was for faculty, that could be really useful for other educators as well.
So I think what we need is more help with that implementation stage. I think that it just generally is just not happening yet. We're just not at the place where institutions have put that as a priority. The majority of institutions just aren't providing much assistance. A lot of schools, it's kind of do what you will, be very careful, don't use these tools to plagiarize.
Their guidance to students is look to your faculty, make sure that your faculty is okay if you're using it. But beyond that, they're not really saying how it should or should not be used. And so we're hopeful. We're actually looking to provide guidance ourselves through the SIG group to help universities who want to do more to support their staff and faculty in that implementation. I am hopeful that we're going to see schools prioritize that in a different way moving forward.
Eddie Pickett: These are all great tools, great suggestions, great things. But one of the big pieces you also talked about earlier were just the ethics of AI. So what are some of the ethical concerns admissions officers should consider when using AI tools?
Sebastian Brown: Yeah, Eddie, that's a great and really important question. With a lot of power comes a lot of responsibility. Everyone can have their Spider Man moment. But I think one of the big ones is the student privacy. Anything that you are feeding an AI tool could be used to help train the model. And so any of that personal identifiable information, PII, really needs to be securely removed before you're throwing it into any sort of training data to protect students.
I think to also protect institutions. I talk about the excitement and how amazing it can be to help institutions and enrollment offices. To go through large data sets, but if that is being collected on the back end, it could also be used elsewhere. So institutions need to be careful of where they're putting some of their enrollment data.
And then I think the other important one is the environmental impact of running AI. Yes, there's free options or they're using things like credits, but it is a pretty significant environmental task to be able to run AI servers and the amount of electricity that goes into being able to process incredible amounts of data is certainly no light task.
And so I think folks need to also keep that in mind as they're continuing to run all their different queries and using different AI tools.
Emily Pacheco: Yeah, this is such a big topic. We are actually going to be doing a webinar focused on this topic in early next year. I've presented at a couple of sessions and recently at the end of one, I had a counselor raise their hand and they brought up that issue of AI and how much energy it uses, the vast energy consumption that goes into using artificial intelligence.
And at that point, I realized he had not heard anything else that I had presented because he was blinded by that issue. And I understood it. It's a concern I have as well, but because we hadn't addressed that and I hadn't talked about that, the person was unable to hear anything else that I had presented today.
He basically said, I will never talk about AI until I understand this is not having something such an impact on US climate wise. I think that's a dangerous place for a counselor to be because that counselor's students are all using artificial intelligence and just refusing to look at it or talk about it isn't going to make it go away.
So what I really encourage is for people who have ethical concerns, it's most important that you understand how these systems work and understand the ethical concerns. I think what often happens is when people, educators, start learning about this, some of their ethical concerns quickly go away. But then there's other concerns that they weren't previously aware of that they learn about, and new concerns grow.
So I think that the best thing that we can do is make sure we have people in our spaces who are experts on these ethical areas and that we make sure we are engaging in conversations about these concerns and working towards solutions. I mean, there are people out there, many, many people working on the energy problem right now.
And these systems, each one that comes out, they actually use less energy. They're finding ways that they can use less energy in the way that these systems are created. But we don't have a solution yet. And people need to continue talking about this and pushing towards making sure we have those solutions.
At the university level itself, definitely it's a huge concern that I could literally just put an Excel document of all of my students data into ChatGPT, upload that, and at most universities, nobody would know that had taken place. And that is very, very scary when you have thousands of people with computers being able to make these kinds of decisions in terms of what goes into ChatGPT or whatever AI system that you're using.
So, we need to educate people. I mean, universities who have not talked to their employees about how they should be using this and how they should not be using this, that's a really dangerous place to be in right now. I recently heard a story of a university person who was trying to make a decision on a candidate.
They had five finalists for a role on the campus. And what they did was just uploaded all five applicants into ChatGPT and asked ChatGPT to come up with the best candidate. This person was really, really proud of themselves in terms of this usage, this case usage. What an amazing way of using these tools. But they clearly had not thought about the privacy implication, the implications of sharing that much private data with these systems.
And that was because they didn't know any better. If we were not educating our staff, our faculty, they will be using these systems in unethical ways. So many universities have implemented closed systems. A lot of schools are using Copilot and the only Copilot where you can just use that tool on campus.
You can't use any other of these tools when you're uploading data to it. So there are guardrails that a lot of universities are putting into place to kind of help make sure that their employees can't make those huge ethical mistakes there. But silence is dangerous, and that's really where we are right now, for the most part, is most schools are not talking about this, and that's a scary place to be.
Eddie Pickett: All right. And for our last question, we're going to think about our crystal ball, and you have to give your elevator pitch. So you've got 30 seconds to answer this question. And that question is, how do you think AI is going to shape the work of college admission professionals in the future?
Sebastian Brown: Yeah, I can take a swing at this.
So I think one of the first ones that we're starting to see it already is that chatbot, the AI assistants that can help handle more of those routine inquiries from prospective students and really allows the student facing admissions counselors or on the high school side to really spend more of that human interaction face to face having those deep conversations.
I think the other important one is more advanced algorithms that can help colleges predict which students will be successful. Or I think an area outside of admissions that we didn't talk much about is schools are starting to use AI to analyze over 100 And 20 different factors for student retention and success and being able to identify some of those things once students are actually on a college campus to make sure that they persist and graduate and become alumni and future donors and all those kinds of things.
So I think that's another important area is more of that student retention piece in the future.
Emily Pacheco: For me, I think it can be summed up in two words, and that is meaningful work. That is where I believe this is going to take us. I have two college degrees, and the amount of work that I do daily that does not require one of those college degrees is very, very high right now.
And I think that's the nature of the work we do. There's a lot of the work that we do right now that can be automated. And there's a lot of the meaningful work that we need to be doing right now, that's not taking place. And so what I hope is that moving forward, I can use more of my brain because of AI.
This allows me to be a more impactful employee, a more impactful counselor, admission counselor. I want to do more of that work, that impactful work. So that's what I believe where this is going to take us.
Eddie Pickett: I'm afraid that's all the time we have today. A huge thank you to Emily and Sebastian for an authentic conversation on artificial intelligence.
And thanks to you, my friends in the audience, for joining us for another great episode. College Admissions Decoded is a podcast from NACAC, the National Association for College Admission Counseling. It is produced by Resonate Recordings.
If you'd like to learn more about NACAC's mission and the college admissions process, visit our website at www.nacacnet. org. That's N A C A C N E T dot O R G. If you enjoyed this episode, please leave a review and rate us on your favorite podcast app. And don't forget to subscribe. See you next time on College Admissions Decoded.
CITATION: “AI in College Admission: How is artificial intelligence impacting students and professionals?”, NACAC College Admissions Decoded, National Association for College Admissions Counseling, January 15, 2025.