BUV Connects Episode 6

Welcome and The Enterprise AI Shift

Wendy Colby: 0:11
This is Wendy Colby, Vice President and Associate Provost at Boston University and the host of BU Virtual Connects. I am pleased to welcome my colleague, Chris Sedore, Chief Information Officer and Chief Transformation Officer at Boston University. Chris brings extensive leadership experience at the intersection of technology, enterprise operations, organizational transformation, and innovation. Today, our topic is on enterprise AI and the future of work. As we all know, artificial intelligence is rapidly reshaping the way organizations operate, influencing everything from customer engagement and operational efficiency to cybersecurity, workforce planning, and enterprise decision making. Across industries, organizations are navigating how to responsibly adopt AI technologies while also preparing their workforce for a rapidly evolving future. Today’s conversation focuses on what enterprise AI transformation looks like at scale, and how business functions such as marketing, finance, supply chain management, customer operations, and risk management are evolving as a result. We’ll explore how AI is reshaping large enterprises, what it means for the future workforce, how organizations can approach upskilling and change management, and why leaders must begin preparing now for the next generation of AI-enabled work. Chris, thank you so much for joining us today.

Chris Sedore: 1:38
It’s a pleasure to be here, Wendy.

Wendy Colby:  1:39
Thank you. Well, let’s go ahead and get started.

AI’s Velocity Compared To Past Waves

Wendy Colby:  1:42

I want to really start with framing the state of enterprise AI. So over the last few years, I think we’ve all seen that AI has really shifted from being primarily an emerging technology to becoming a central business conversation, right, across nearly every industry. Firms like McKinsey, Deloitte are describing AI as a major driver of enterprise transformation, impacting operations, customer experience, productivity, decision making, as you know. From your perspective now, what feels fundamentally different about this moment in AI compared to perhaps previous waves of digital transformation?

Chris Sedore: 2:17
Yeah, it’s a great question, and I think really interesting. And, you know, in podcast format, you can’t see me, but if you could, you’d know I’ve been in this business a long time just by looking at me. And what’s interesting to say here is just the velocity, how quickly we’re moving in this phase of transformation. And I think interestingly, you know, I’ve been doing this for a little over three decades now, or maybe more than a little over three decades now. And I look at this and say, we’ve had, you know, several waves. We think about personal computing from the late 80s into the 90s. We had networks, then we had internet, then you know, cloud, mobile, social, all these things. We’re seeing AI move like a year to a decade compared to what we saw in prior digital transformation waves, both on the personal side and also on the enterprise side.

Wendy Colby:  3:05
Yeah, that’s awesome. As we think about maybe moving beyond experimentation today, many organizations, I think, we also see are experimenting a lot with AI tools, right? But enterprise implementation introduces a lot of complexity, a lot more complexity. You know, the governance, the infrastructure, security, workforce readiness, change management, operational integration.

What Makes Enterprise AI Different

Wendy Colby:  3:30
Can you talk about how you would define enterprise AI? You know, as someone who’s leading–who leads these large IT organizations that are really the underpinning of what drives organizations today.

Chris Sedore: 3:43
Yeah, I think that, you know, the you’re right about that. There’s experimentation, and I think the experimentation is easy on the personal side, right? We all use ChatGPT or Claude or a tool like this, and we’re getting real benefits. Like, I don’t write job descriptions anymore in the way I used to. I’m just like, hey, I need a job description for this, and then I tweak it from there, right? So there’s been a big uptake. On the enterprise side, as you suggested, like it’s a lot more work to do this with security, to get the right data integrations, to provide scale, to manage cost, all these different elements that are part of an enterprise launch. And I think as we look at this at the enterprise side, I can say we’re moving, really thinking about what we say in some cases, like what are the foundations? What are the–what’s the infrastructure that we have to get right? Data, absolutely critical to get that right. Security, of course, privacy, these sorts of elements of the sort of governance framework are really important. The other piece is really starting to work with the business side to explore those implications. They’re already seeing it in personal experimentation, but I don’t think that folks have fully grasped the difference of what it’s going to mean when we go enterprise with those those capabilities.

Wendy Colby:  4:55
And do you have any sort of vision for what this does look like as we think about and enterprise transformation more strategically, right? It’s not just placing a lot of these little bets anymore, it’s really looking at sort of system-wide, right? How do we empower and enable a new organization going forward?

Chris Sedore: 5:12
Yeah, I think it it’s going to be very interesting to watch this evolve. And it’s important to think about not maybe where you see the technology today, because as we’re doing the foundations work, we’re trying to project where do we think the AI capabilities are going to be in 24 months, 36 months, 48 months. Because as we build those foundations, we want to make sure, first of all, we don’t run out of runway in those directions, but also as we begin to work with our business side folks, understanding what those capabilities are gonna mean, right? So these are the things that we talk about quite a bit; automation of rote work. We see this already happening at the individual level, but we’re gonna see it at scale. Level one help desk tickets, not just IT help desk, but we have an enrollment help desk, we have an HR help desk, we have procurement help that happens. Many of those things are ripe for automation. And then even next scale things, contract review, there’s all sorts of things that we can automate and scale there, and we see pretty significant reduction in rote workloads as we pursue this automation.

Automation, Judgment, And Humans In Loop

Wendy Colby:  6:16
Yeah, no, I think that’s really helpful. And you may have answered this already, you know, getting into the changing business functions, right? Because we always talk about how we’re empowering new ways of doing business in an AI era. And we’re gonna get to this later in the conversation as we talk about, you know, more reskilling and upskilling and what’s needed now among our sort of employee populations, right, to foster that. And so I wonder also, you know, even outside of the higher ed environment, as you think about marketing and customer service, of course, we do these things as well. Finance, you’ve talked a lot about finance, supply chain management, you just mentioned this, right? We’re moving out of these automated, isolated workflows and into something much more sort of predictive when it comes to looking at, you know, what might enrollment mean? What will marketing look like, you know, in a future state? And are there areas that you see might have the most operational impact across functions? Do you think it will be more dispersed and diversified across functions? How do you see that?

Chris Sedore: 7:12
Well, I think it’s going to be quite dependent on where your organization is when you start this journey. I think we have the full spectrum. We have some areas where we still have processing that’s fairly manual, and we just see that getting automated very quickly. We’re already seeing that happen, even at an individual level, but the enterprise overlay, a lot of areas where we have folks just manually processing things through, whatever that may be, uh, we think that work is going to get automated pretty quickly. And so that part is happening everywhere, and that’s true in, as you reference, in marketing and finance, you know, all the–all those core business functions. I think the really interesting question said is what happens at the next phase of the work, right? Because that some of it, frankly, we could have potentially automated or at least partially automated with existing technology pre-AI. Now we actually have stopped some of those automation projects to say the old way of doing it wasn’t the right long-term approach. So we’re going back and redesigning it in an AI-enabled way. I think it’s what’s really interesting is when you get to the next level where we’re rather than just processing and handling, you know, documents or workflows, as the case may be, how can we start to think about automating some of the human judgment elements of what happens there? And where do we have humans in the loop? Where do we think that’s necessary? And you know, oftentimes when I talk with folks, the base assumption is well, we need to have zero errors. Like, can you point out a human organization that makes zero mistakes? We don’t have those, right? So the thresholding really needs to be when is it as good as or better than what we’re doing in our existing processes? And when we hit that, we should be comfortable moving to automation to drive those tasks.

Wendy Colby:  8:54
You talked a little bit, and I want to just probe on this for a minute, the human in the loop, right? And I’ve even heard some talk about HI, human, right, as opposed to artificial intelligence, human intelligence. And I think we all agree, right, that it’s still gonna remain essential as we move forward. But at the same time, there is a lot of, sort of fear in the system now that my job is going to change, I’m gonna have to do things very differently. Will AI replace my job? And so I’m wondering if you can talk a little bit about how you see that, how you see AI really now interacting in ways that humans still have a vital role to play.

Chris Sedore: 9:30
Well, humans do have a vital role to play. And as we look at this, there are certain human functions that we don’t ever expect to automate and replace. And we talk about this, you know, in broad categories. We have to give adverse news to people sometimes. This is true in many dimensions of organizations. It’s true out of HR, it’s true in enrollment near and dear to me. And that’s a human function, not an AI function, and we want to preserve it in that fashion. I think the, you know, on the workforce side, we need to be honest. People’s jobs are going to change. And I don’t think that change is necessarily good or bad. It’s up to us how we how we think about it and how we evolve it. And I think as we look at this at Boston University, we’re saying, you know, we have values we live by here, and caring for our people is one of those values. And so we’re really trying to figure out how do we navigate forward in a way where these job changes are coming, they’re going to happen. Every transformation that we’ve had, we had this with the industrial age, with the digital or information age. Now, if we have an AI age or whatever we’re going to call this next feature, it is going to shift the nature of work and of jobs. So how do we navigate that with our values in mind and helping our workforce move on to that next sort of phase of running an organization and contributing to society, all the things that are important to us.

How Leaders Build Confidence With AI

Wendy Colby:  10:55
Yeah, really important. You know, a lot of our listeners here are both, you know, leaders inside of industry today. Often we are looking, as you know, as leaders inside of an university today, at how we pair those two areas, right? Industry and academia, and how does one service the other? And you just talked about the changes for HR leaders, for marketing leaders, right, for IT leaders today. And so, you know, again, how do we help enable, these new sets of leaders going forward, right? Who may not have the expertise you have, Chris, in this domain with all the years you have of leading, you know, business transformation for a lot of industries. How can we help these leaders going forward so that they are empowering their organizations and really instilling this sort of sense of optimism? I often like to think when I think about new fields like you know, computer science in AI or software engineering in AI, right? I like to think sometimes we’re like democratizing, you know, what it means to be a coder today, where it’s still very important to have the human element as part of that. And does that help to open really up these opportunities for education and advancement as well?

Chris Sedore: 12:05
I think it does. And I think the first thing I would say to leaders, and I’ll relay a conversation I had over dinner with a colleague a few weeks ago where he was describing a business problem that he was trying to solve and he wanted to use AI, so he reached out to some technical colleagues to talk with him about how to do it. And I said, well, explain to me what you’re doing. Essentially he was trying to reconcile some documents, but essentially, an invoice he was getting against some work he was expecting, and I said, well, have you tried just uploading the two sets of documents to Chat GPT and telling it what you need? He’s like, well, what would I do? I’m like, what you just told me, type it into Chat GPT and tell it that. It’s probably gonna do a much better job than you anticipate. He’s like, I don’t have to write code or do anything? No, you don’t need to write code. You can just talk to it in English and move this forward. So the biggest thing is you have to have an experimentalist attitude. There’s there’s really nothing to be afraid of here. And I’ll caveat that by saying, don’t put important institutional data into you know these public systems and whatever. Talk to your IT folks or your security folks or whatever, but run some experiments. Take things that are not sensitive, or you probably have enterprise AI if you’re in a large enterprise, use those tools and try them out. Make sure you verify what they say, as we always tell folks, but don’t be intimidated by it. It’s really very, very accessible. So I think that message is really important to folks. And the skills that are there are really in some ways just managing people kinds of skills, right? And if you talk to folks who use this every day, you’ll hear, and you’ll hear me say this in informal settings, sometimes I feel like I’m managing one of my teenage kids. Like I asked you to do this, but you did that. Why didn’t you do what I asked you? And it’ll say, Oh, I’m sorry, you’re right, I didn’t do what I asked you to do, right? So there’s this, there’s this willingness to experiment and iterate around using AI in those use cases. And what’s really helpful for leaders is it gives you a sense of where is it really strong and where is it less capable or weaker, at least at this point in time. And that begins to broaden your thinking about how you can apply AI in your business space. And you might be surprised if you haven’t asked it questions about your specific–whether it’s your industry–your specific function in your organization, how good answers you might get if you just say, hey, I’m worried about, you know, how we could transform what I do in HR or in finance or in manufacturing or whatever it is. If you give it some description of what you do, you can have a pretty good dialogue with it about what that might look like.

Governance, Hallucinations, And Cybersecurity Reality

Wendy Colby:  14:36
Yeah, I think it’s been really fascinating to watch how much it has evolved just even out over the past number of months here, right? You started to talk about, Chris, cybersecurity, risk, governance. So I just want to spend a couple of moments on that because I think that’s important for our audience too. You know, AI really introducing new questions around those areas and institutional trust, organizations trying to balance innovation and, as you said, experimentation and exploration along with responsible implementation. So, how should leaders think about governance and risk management as AI adoption accelerates?

Chris Sedore: 15:09
Yeah, it’s a great and difficult question, I think, because you know, there’s a risk of locking things down too much and scaring people away from experimentation. But at the same time, for enterprises, it’s really important that we secure our data, we make sure that we give people safe paths to do the work that they need to do. And I think, you know, in enterprises, in large enterprises for sure, you’re gonna have a security team that’s working on this. You should have already access to approved AI tools that you can use and and just follow those instructions if you’re not in the IT space. I think on the enterprise side, what I would say is we look at this through an IT lens, the judgments here are complicated, right? And as is the case in cloud, for example, and we went through this when cloud first started emerging maybe 20 years ago, looking at, well, how much do you trust your providers? Because really, at the end of the day, all you have is a contract that tells you what they promise to do or not do. And so you need to be really honest with yourself about okay, this is a contractual guarantee, it’s not ironclad in that sense. And of course, you know, there are lots of reasons to be concerned, and we have a lot of movements now about where does your data live? What legal jurisdiction does it live? So we’re very thoughtful about what things have to stay in the United States, what things we’re willing to look at being in other legal jurisdictions where their protections and rules may be different. So all those things come into play. But generally, your enterprise teams, a combination of IT and your counsel’s office, is gonna do that in enterprise scale. If you’re doing this in a smaller scale operation, you should probably get some outside advice. Make sure you’re talking with folks about what those risks are so that you’re protecting your intellectual property and also the information about your customers or constituents if you’re processing that using.

Wendy Colby:  17:03
And maybe just a quick follow-on to that question, you know, are you seeing teams? Because you mentioned, you know, leaders should look at their IT or governance or risk teams. Some of this is still very new even for those teams. So do you feel like the teams today have the skills needed to do this, or do you find yourself as a leader of a large organization also looking at what it takes to really keep pace and keep pulse on what’s happening in this space?

Chris Sedore: 17:25
Well, certainly it’s interesting. What I would say, in some ways there’s nothing new under the sun, right? So these issues of using outside providers, we’ve had those for a long time. And in fact, when AI first emerged on the scene, everybody said, oh, we have to update our policies. Like, actually, our policies already say things like don’t put institutional data into unapproved systems. So if you followed the rules, you were probably already safe. We did update them to clarify that that included AI so people could read it there in those in those cases. I think that, so if you’re following good practice, you’re generally going to be pretty safe in terms of where your data goes. I think the greater complexity, and we see this play out often in the business side, is are you validating the outputs, right? We’ve seen lawyers sanctioned by courts because they’ve submitted briefs that have hallucinated citations, right? So you really need to look at the output side and make sure you’re responsible and you’re owning that. There’s been some really interesting research by Professor Wiles and some other folks here in Questrom covering how people see AI outputs, and it’s worth thinking through how your workforce is using those AI tools, how much responsibility they feel for that work product. And it’s interesting, depending on how you position it, people perceive those outputs differently. So I think that’s important. The other piece I’ll say about cybersecurity is really kind of on the other side of the fence here. And if you’ve followed this recently with Project Classwing and Mythos from Anthropic, where we’re seeing AI expose security vulnerabilities in ways that humans really haven’t been able to pursue. And I won’t go deep down the technical side of this, but I think you need to really be paying attention to how you’re managing your systems, that you’re investing appropriately in your cyber infrastructure because AI is accelerating the ability to find vulnerabilities in those systems. And so if you’re not invested there as you need to, you really need to be paying more attention to that.

Skills That Raise The Workforce Bar

Wendy Colby:  19:24
You know, Chris, based on what you just shared, I was thinking a lot about, you know, we’re coming off commencement weekend, and I had the opportunity last weekend to also meet with a number of our recent graduates. And in our world, as you know, of the online space, we have a lot of working adults, right, who actually flew in from all over the world to accept their degrees here. And what was very interesting to me is many of them are now focused on what does it mean for me in business, in my healthcare operation, as a marketing executive, as a leader, to now work within an environment influenced so heavily by AI. And so I’m now seeing these leaders say, you know, if I am in IT, great, I got this MBA, right? But now I need either an AI for business degree or an enterprise AI degree online to really help support where I need to go next. And so I’m just wondering, you know, if you can comment on that a little bit, right? Because again, I think I’m seeing a hunger for how do I show to my own workplace, you know, the value of what I can bring, the ability to pivot quickly, the agility that’s needed to do that. But I I need some core skills as well to do that.

Chris Sedore: 20:35
Yeah, I think it’s a great question, Wendy. And you know, when I talk to folks about hiring and even advancement, right, I have three things I generally look for: folks’ attitude, aptitude, and motivation, right? Because I can’t give you those things, but if you have them, I can let you go as far as your motivation and aptitude will take you. The skills piece I think is is really interesting and important here. And in AI, especially I would say in enterprise AI, we’re seeing really emergence of architecture in the way we think about AI. So we talk about things like ontology, which you can think of really as the sort of roadmap for how your business operates. It’s some combination of policies and procedures and practice, and even I would say culture, organization, how you think about things and do those things. The data infrastructure, I mentioned this earlier, absolutely critical. But what we’ve built for the last 30 years is data infrastructure for humans. It turns out that data infrastructure for AI looks a little bit different. And so, understanding, you know, how do you think about a semantic layer that supports your AI enterprise? And, it’s funny maybe to talk about it this way, but one of the things we like to do with AI is reduce its cognitive load. So the harder you make your AI system think about how to access the data, the less sort of cognitive capacity there is left to hold a conversation with the user. And so you wind up loading it up with more instruction so that it knows how to use the data as opposed to giving it better instruction on how to solve a business problem or chat with a person or whatever that that application may look like. I think this is really where the interesting educational opportunity is.

And really part of what I say to folks about AI is it’s raising the bar on where you need to be to enter the workforce. And, I think it’s been true in my experience doing hiring in the last 20 or 25 years. If you go back to the 80s, many people would start a new role, and the expectation was that the the entity that hired you would spend six months or 12 months training you, bringing you up to speed. That really went away, honestly, in the late 90s and the aughts. And we only hired people who are walk-ons. You couldn’t walk in and know how to use the tools, and this drove change in higher ed. We started making sure that our students walked out knowing how to use the tools that they were gonna walk on and use in industry when they got their first roles. Now with AI, unfortunately, that bar’s getting raised a little bit for folks in the sense that, you know, you referenced coding, just being able to write code now, that’s not something that’s a differentiator because we can produce that with AI. I want to gloss over that. AI is not a replacement for developers in that sense. And if you look at vibe coding, you find out that pretty quickly you run into a need for a developer. In fact, there’s a job category emerging of senior developer who cleans up vibe coding because it made an architectural mess, right? So, but that leveling up is really at the data level, understanding what this looks like, architecturally understanding how to build enterprise AI systems, software development, it’s understanding system architecture, not just how to write code. And so I think those are really where the skilling up opportunities come in, the value generation for you as a staff member or even a leader in your organization.

The other thing I’ll say is interesting is that the nature of what one person can take on is getting bigger. This has really interesting implications. So several weekends ago, I had an absolutely glorious weekend. I worked about 33 hours over two days, alright? It’s a great time to be alive. Yeah, it was just a great weekend. But I was running six agents in parallel the entire 33 hours that I worked over those two days. It was like having six junior developers cranking out code for me. And what I was doing was providing architectural commentary, guidance, like, no, no, no, I don’t want to do it this way, I want to do it that way, right? Well, what did that mean? Over those two days, I produced about 20 or 30,000 lines of code. That would have been for me six to eight months worth of regular effort in two days, right? But I was able to solve a much larger problem that I was pursuing in that case. Now, though, what I’m doing is I’m talking to the business side and thinking about what the business problem is. I’m designing the solution, I’m overseeing essentially the coding, doing some coding review, overseeing the testing. We of course have the AI write the test. Job magnifying, right? But here’s what’s interesting about it what it does, is it collapses the work. So now, and I think it makes it more interesting, right, because now I’m not just working on code in the background. Yes, I get to do some of that, but I actually get to go work with the business side on what the problem looks like, walk that all the way through, take it back to the business as the first, you know, version of the solution, maybe MVP on the solution, get that feedback, run it back through a loop, and we’re seeing that really supercharge the rate that we can roll out solutions. And for my teams, this is energizing because we’re here to actually solve business problems. We really enjoy doing this, we enjoy enabling the business, we get intrinsic rewards from doing this. So that ability to move and go faster. But sometimes when we talk about this, we think about like, well, I’m a developer and I create code. This is my box, right? Now it’s more like, no, no, actually, I want you to go out and meet with the client or the customer, however you like to phrase that, understand what their business problem is, essentially develop the specs, perhaps collaborating with AI, you can check in with the business on what that is, but then you go end-to-end and you actually own that product. And that’s really empowering for individuals to do that.

Wendy Colby:  26:01
I love the way you’ve talked about that because sometimes I think we skinny down sort of the, you need critical thinking, you need leadership, you need, you know, all of those kind of soft skills as we used to call them, right? And this is really thinking much bigger and broader. There’s a reason you have, you know, information and transformation, right? And your broader remit. You know, as you also think, I mean, I love that story of 33 hours on a weekend and having all your agent people working for you.

Universities, Industry, And The Next Steps

Wendy Colby:  26:28
Often we talk a lot about, you know, academia and industry, and I talked about this a little bit earlier, but are you seeing greater opportunities for collaboration between the university side, the higher education side, given all the changes going on in higher ed and industry?

Chris Sedore: 26:44
Absolutely. I think there’s a significant opportunity for collaboration there in several domains. I think one, we are still in the early stages of what does architecture of enterprise systems look like for AI, right? And many of the principles we had before, they still hold. But AI is a new actor in this space, and it is driving us to think differently about how we structure things. And we’re already talking with some of our industry partners about collaborations around how to develop new ways of working to understand that. Many of our industry partners are seeing the same thing. They have their own RD operations, which is of course a place like Boston University has been in the RD business for a long time as a research university. So lots of productive collaborations there. I think the skills development piece, we’re all concerned about what is the future of employment, how do we think about jobs and skills and where do we really need humans on the other side of this transformation? And the good news is we’re gonna still need humans. That’s not going to stop or change. But the skills, the roles, they’re going to be different, and I think largely better. You know, I look and say, if you go back and I remember having an opportunity to look at a picture of a construction site at a university in the early 1900s and the work that those folks were doing, a lot of it’s by hand, right? A lot of it’s much more dangerous than it is today. Now, thankfully in most enterprises, we don’t have those sorts of things. Some certainly do. But what we’re going to see, I think, is a similar kind of evolution where much of that work that we do by hand today, we’re going to be able to automate and scale in ways we couldn’t before. And humans are going to be more orchestrators of those processes. It is an extended skill set. I don’t want to say different. There are some different skills, but it’s really an extended skill set. You still have to know, like in enterprise IT, you still have to know what enterprise architecture looks like, how systems work, how networks work, all those skills there. But you’re doing less of the individual hand manipulation. This is why I say it’s such a great time to be alive. I talk to my colleagues, none of us want to go to bed anymore, right? Because there’s–I was like, oh, 15 more minutes. Imagine how much I could get done, right? So that empowerment is one, and that’s what I want everyone to see from this. It’s like, rather than being afraid of this, you should be energized by it because your ability to make a difference, to have impact is so amplified by this capability. And yeah, there’s gonna be change and it’s gonna be scary. But I’ll say if you’ve worked in IT, we’ve had change for at least 50 years in IT, and it’s only really been accelerating. So if you’re in this field, you’re already used to that. So don’t run away from it, run towards it and figure it out. And if you’re committed and you work hard, you’re gonna make a positive difference on the other side.

Closing And How To Stay Connected

Wendy Colby:  29:25
Thank you, Chris. This is a perfect place to bring our conversation to a close. Remember everyone, attitude, aptitude, motivation. It’s such a great time to seize opportunities, learn new skills, and imagine all that may be possible in this evolving AI era. Chris, you’ve given us a lot to think about. Thank you so much.

Thank you for joining us for this BU Virtual Connects podcast. Special thanks to my colleagues at BU Virtual and to our media team who produces this podcast under the leadership of our studio director, George Vago. To keep up with our BU Virtual Connects series, be sure to subscribe wherever you listen to your favorite podcasts. You can also learn more about our portfolio of online programs at BU Virtual by visiting BU.edu forward slash virtual.