00:01:36 · Kevin B. Boyd
[00:01:36] S1: Absolutely. So, you know, I think one of the reasons I love my job is that there is no day that's the same. You know, we cover a lot of ground. So I think probably the most interesting part of the job is the work that we do to support research. University of Chicago is one of the top research universities in the world. And a lot of the time that we spend is thinking about how do we better enable and support the amazing research that goes on. And that could be biomedical sciences, it could be physics, computer science, social sciences. And, you know, I don't think it'll be any surprise to you that these days it's a lot of conversations about AI and how we can effectively use AI to support the research that's going on. One of the things that we did this summer is we rolled out Anthropics Claude with chat, co work, and code for all of our students, faculty, and staff. Excuse me. And that's been really exciting to see how that is being used. It's also led to many conversations across campus about what are appropriate ways to use AI, what are less appropriate ways to use AI, whether that's in classroom, in research and administration. And so that's been really interesting the last few months.
00:03:07 · Grace Gong
[00:03:07] S0: I I think like from our conversation, I feel like you guys have also are like also from like our research, you guys also have built out your own like internal system as well. Like, I think like we talk about, like, essentially, like, I remember when we're chatting something about quantum and as well as like the data center and stuff like that. Maybe we could talk about, like, you know, I know that, like, you know, rolling out like cloud or like a lot of the, like, I guess like AI native AI tools is extremely important. What are something that you felt like set University of Chicago apart from the rest of the universities from the AI perspective?
00:03:49 · Kevin B. Boyd
[00:03:49] S1: I think from the AI perspective, we've been taking a very thoughtful approach to it. So our president, Paul Alivisatos, wrote an opinion piece in, think it was the New York Times recently, and it was really talking about our philosophy of AI being with, without, and about. And what that means is there are many things that we want to do at the university with AI. At the same time, we've also been very intentional, particularly in the instructional side, of saying there are some places that we want to have teaching and learning happen without AI. And so the university has gotten a fair amount of press in the last couple of months about our humanities division, which has largely said, we want classes to be technology free now because we actually want people to learn. Know, something that's been key for us for many years is University of Chicago is not a place where you go to just memorize facts and regurgitate them. It's a place where you really learn how to think, learn how to challenge. And particularly as an undergraduate in the humanities and some of the other programs as well, the law school has been very public about like the first year law students. They want to be pretty much technology free. So it's this idea that we want you to learn. We want you to grow as a student before you start using AI to do everything that you're doing. And I think that's really useful. And then the about piece is we're gonna teach people about AI, how to use AI effectively, whether that's things like co working code, or it could be how to build agents and other tools like that.
00:05:52 · Grace Gong, Kevin B. Boyd
[00:05:52] S0: When... I guess, like, when you're thinking about designing on, like, a commercial AI... APIs, like, like, whether it's, a open source model or, a model that running your own, like, infrastructure, how do you typically make these kind of decisions? [00:06:07] S1: Sure. I think in many ways there's sort of the old way of doing things and then there's the new way of doing things. Universities are generally pretty conservative. So historically, they don't change all that fast. Technology wise, you know, our focus has been on how do we make sure that things are just rock solid, stable, and you know, delivering what our faculty, staff, and students expect from us. You know, in an AI world, we're really having to be more innovative, move faster, build tools that maybe in the past we would have focused more on buying things. Now we're actually building them just because it allows us to deliver efficiency and effectiveness faster. So I think when you talk about, you know, how do we make decisions about AI or about, you know, the APIs and things like that, It's really a question of how do we both get to market fast, deliver the business value that our business partners across campus are looking for, but at the same time make sure that it's going to deliver the same stability and reliability that they're looking for.
00:06:07 · Kevin B. Boyd, Grace Gong
[00:06:07] S1: Sure. I think in many ways there's sort of the old way of doing things and then there's the new way of doing things. Universities are generally pretty conservative. So historically, they don't change all that fast. Technology wise, you know, our focus has been on how do we make sure that things are just rock solid, stable, and you know, delivering what our faculty, staff, and students expect from us. You know, in an AI world, we're really having to be more innovative, move faster, build tools that maybe in the past we would have focused more on buying things. Now we're actually building them just because it allows us to deliver efficiency and effectiveness faster. So I think when you talk about, you know, how do we make decisions about AI or about, you know, the APIs and things like that, It's really a question of how do we both get to market fast, deliver the business value that our business partners across campus are looking for, but at the same time make sure that it's going to deliver the same stability and reliability that they're looking for. [00:07:31] S0: When you were thinking about, I guess, would you say is the most important skill like a CIO needs to deliver like a leading AI adoption across like different organizations and technical teams of like, you know, and like researchers and faculties and administrative. Seems like they all like have like different priorities. How do you kind of like think about the AI adoption from that lens?
00:08:02 · Kevin B. Boyd
[00:08:02] S1: You know, I think think that's an interesting question. I really think that a lot of it is it's leading with empathy. It's being a really good listener, and it's helping to translate across different groups. So much of what we're doing, it's almost less about the technology. Yes, the technology is really important. But in order to successfully drive change and drive the transformation, it's a lot of talking to people and trying to understand what are the problems that they have and how can we deliver solutions that are gonna help them solve those problems. So like an area where we've put a lot of focus at the university is in the finance and accounting space. So we implemented a new financial system a few years ago. Enormous transformation exercise, the largest IT project the university has ever done. And now we're a couple years down the road and we're in a position of saying, okay, how can we use AI to make things better, to automate things? Like one of the, you know, one of the examples, our CFO had come to us and said, you know, we're we're investing a lot of time and energy in what what they call like the triple matching with invoices. So you think about it, universities buy lots of things. We buy, you know, things that go in the research labs. Just... And something like 80% of our transactions are less than $5,000. So a lot of high volume, transaction value things. But one of the things you want to do at the back end of the purchasing process is you do this triple matching where you say, okay, the purchase order was this amount, and what did we actually receive? And then what did we get invoiced for? And do those three things match up? If they don't, then maybe, you know, send that to a human being to look at it. So it's a great use of an AI agent to say, you know, do those three things match? Is there anything else that has happened along the way that might cause them to be different? And if something looks out of order, then let's make sure that a human being looks at it and gets involved. And that's really high value for the finance and accounting team. It allows them to work on the bigger transactions, higher value things. And, you know, that happens because of good relationships and conversations with the the finance team about, you know, what were the challenges they're facing.