About this conversation
A CIO Talk Network conversation with Miro Humer, Vice President and Chief Information Officer, Case Western Reserve University, hosted by Sanjog Aul. Recorded October 2025.
If that is the bar, who owns it, and how will we measure it?
Everyone says they are getting people ready for an AI world. Programs launch. Skills are listed. Capability still feels uneven. Readiness is not a tool list. It is judgment, curiosity, and the habit of learning when answers keep changing. Classrooms start the process. Workplaces test it under pressure. The learning loop between them is still forming.
The way forward is a shared standard for what “AI ready” means and a way of learning that sticks. Teach people to frame problems, question outputs, and know when to pause. Measure this in daily practice, not just in slides.
What the conversation established
A university CIO and the host compared how academia and industry are preparing people for AI and found both drinking from the same firehose with no clear path forward. Miro Humer’s working definition of AI readiness has three parts: a critical AI literacy (which tool, for what), the adaptivity to change tools as they change week to week, and an ethical grounding for deciding whether AI belongs on a task at all. Case Western Reserve University has rolled AI into over 100 courses as a core skill in every discipline rather than a separate program, on the belief that AI gets you 80 percent of an answer and the graduate is hired for the remaining 20 percent of insight, judgment and analysis. That human share is turning out larger than industry expected: the picture has moved from one human in a loop of machines to one machine among many humans, and a company that replaced its help desk with AI has gone back to people. Industry does not yet know what it wants beyond ROI, which is measurable in administration but not in the classroom, and who judges the human judgment is still fuzzy: a senior person matched to the level of the work, held to account the way a manager is for an employee. The gap closes through proofs of concept done together on real corporate problems, lifelong learning rewarded in reviews, and leaders curious enough to move forward rather than refusing AI over cheating or trust fears.
Contributors
- Miro Humer, Vice President and Chief Information Officer, Case Western Reserve University
Host: Sanjog Aul, Founder and Host, CIO Talk Network
On this page: Key takeaways · Chapters · Explore more · Transcript
Key takeaways
- AI readiness has three parts: literacy, adaptivity and ethical grounding. Miro Humer (Case Western Reserve University) defines it as a critical AI literacy (which tool to use and what is appropriate), the adaptivity to change tools as they change from week to week, and an ethical grounding for deciding whether AI belongs on a task at all.
- Academia and industry define readiness differently. In the classroom the concern is ethics, how to teach with AI and when students may use it; in the corporate world it is efficiency and effectiveness in a role. Case Western’s aim is graduates who become the force that rolls AI out in the organizations they join, carrying the ethics they learned with them.
- Both sides are drinking from the firehose at once, and neither has the North Star. Corporate leaders tell Humer they must implement AI because graduates will expect it; universities say they must teach it because corporations will have it. Everyone agrees AI has to be understood; nobody has a clear path for what it looks like in practice.
- Make AI a core skill in every discipline, not a separate program. Case Western has rolled AI into over 100 courses, in finance, marketing, engineering and medicine. A finance professor sets a problem and asks students to solve it within three prompts and submit the prompts; the lesson is how to ask the question and how to analyze the answer.
- AI gets you 80 percent; the graduate is hired for the other 20. Anybody, including the boss, can get 80 percent of an answer from AI. The insight, judgment and analysis that complete it are what make a person worth employing, so the last mile stays human.
- The human share of the loop is bigger than industry expected. A year ago the picture was one human in a loop of machines; now it is one machine among many humans, needing more interaction than planned. A company that outsourced its help desk to AI and cut several hundred jobs has gone back to people answering phones because callers did not want the AI.
- Industry does not yet know what it wants beyond ROI. The common ask is payback on the AI spend. In administration the return is measurable (80 percent of an answer in 30 seconds, then 15 to 20 minutes to finish instead of four hours); in the classroom and in research it is hard to measure. Two years of proofs of concept on both sides are still building the formula for where AI works, and the tool keeps changing underneath it.
- Someone senior has to judge the human judgment, and that is still fuzzy. Time savings on basic tasks are easy to assess; an AI-drafted RFP response needs a senior person to evaluate it. Humer’s rule: the person applying judgment to AI output is the manager and the AI is the employee, so give them the rope you would give a manager and question their judgment the same way when a canned response gets passed on.
- Pressure exists in both worlds, but the classroom judges on more dimensions. Corporate judgment narrows to whether the project target was met; faculty also ask whether it was done ethically, how much judgment was applied and whether the student added to the output. Early-days problems such as legal briefs citing cases that do not exist should fade as people learn where AI can and cannot be trusted.
- Lifelong learning is the real readiness, and leaders have to be curious. Reward the behavior of learning, ask employees in the review what they learned this year, and expect the people who refuse to learn to suffer most. Collaboration means proofs of concept done together on real corporate problems; refusing AI over cheating fears (universities) or trust fears (corporations) is not an appropriate answer.
Chapters
- 0:00 Introduction: what AI readiness means and where it starts
- 1:28 A working definition: literacy, adaptivity and ethical grounding
- 5:03 Good citizens, profit motives and the values students carry into industry
- 8:21 Both sides drinking from the firehose: is there a shared North Star?
- 13:03 The last mile stays human: 80 percent from AI, 20 percent judgment
- 16:09 Does industry know what it wants? ROI and where to measure it
- 18:09 Where AI stops and the human takes over
- 21:37 Who judges the human judgment?
- 25:34 Pressure, timelines and the mindset that survives them
- 29:08 Chronic gaps and how academia can lead industry
- 33:55 Lifelong learning as the real AI readiness
- 36:33 What true collaboration looks like: proofs of concept together
- 39:24 What must change in the leader
- 41:49 Closing
Explore more
- The Metabolism of Knowledge in Agentic AI: Governing AI Before It Governs Us: Vince Kellen, Chief Information Officer, University of California, San Diego
- AI Agents in the Wild. Who Signs the Incident Report?: Aarti Singh, Enterprise CIO, Mahindra Group
- The CIO as the Conscience of AI: Earl Newsome (Cummins) and Viren Shah (AGCO)
- Reimagining the IT Team: People, Culture, Leadership: Dominic Redmond, Chief Information Officer, PageGroup
- The Future of Leadership: Is Traditional Hierarchy Holding Us Back?: Alexandre Kozlov (ABB), Andrea Heekelaar (Renault) and Mahendra Beharie (DHL Express)
- Agentic AI in Healthcare: seven healthcare leaders including Sanjay Gupta (Novant Health), Robbie Freeman (Mount Sinai) and Mouli Venkatesan (Blue Cross Blue Shield Association)
Transcript
Introduction: what AI readiness means and where it starts 0:00
Sanjog Aul [00:00:05]:
Hello and welcome to CTN. To learn more about the show, please visit ciotalknetwork.com, and the topic for today is Bridging the AI Readiness Gap. So what are we talking about here? AI readiness. You look at people at the leadership, or people who are starting out in the workforce, they are in middle management, young professionals. All that is on the industry side, but this starts much sooner. It starts with students who are getting into the academic areas, where they will start talking about AI and they will learn how to understand a problem, figure out how to deconstruct it, try to find ways to solve it, and in this new day and age, how to do that cheaper, faster, better with AI. And if this is being done in the right way, then we will build a great pipeline, and any new innovation that happens, outcomes, and which is coming at a very fast pace, academia and the industry would be able to collaborate and have this one well-oiled machine which will support the level of readiness that is required in the workforce.
Sanjog Aul [00:01:15]:
So to discuss this, I have Miro Humer, Vice President and Chief Information Officer with Case Western Reserve University.
A working definition: literacy, adaptivity and ethical grounding 1:28
Sanjog Aul [00:01:28]:
So now, when we talk about this term, AI ready or AI readiness, could you give your definition, or what you think should be the common universal definition?
Miro Humer [00:01:41]:
Well, I’ve been thinking about that a lot. I think it really requires a cognitive and an ethical agility in how we look at it and how we move through this area. I think we have to have a critical AI literacy. So if you’re using AI, which tool do you use, what’s appropriate? So it requires a certain level of literacy, that you know something about these tools, and because they’re changing so quickly, from week to week, from day to day, month to month, you have to have a certain adaptivity to it, be aware that the tool that’s really good today may not be really good tomorrow. So you have to be ready to adapt to what the industry is producing, what is available to you, what you can use, and then also I think it’s critical to have an ethical grounding. Should I use AI for this? Is AI appropriate for this task? Understand some of the ethical concerns around AI, some of the limitations of AI as a tool.
Miro Humer [00:02:48]:
So I would think those three things: having just a basic literacy of AI, having the adaptive thinking about what tools are there, what tools are appropriate today that weren’t appropriate yesterday, and have an ethical view of it. When should I be using AI, when should I not be using AI, what tasks are appropriate for it and what aren’t. I think those are three basic components of just being AI ready.
Sanjog Aul [00:03:18]:
And what you are thinking, and the way you describe in your words what’s AI ready. Do you think there are gaps and/or conflicts in the way people are defining AI today, sorry, let’s talk about AI readiness specifically. Do you think different people have different definitions, or are we converging to one common? At least that would be a great start.
Miro Humer [00:03:41]:
I do think there are different definitions of readiness depending on the industry, the field you’re in. I think it’s very different in academia as it is in the corporate world. I think in academia there’s a much bigger concern around the ethics of it, especially when you’re dealing with students in the classroom. How should we teach with it? When should they use it in the classroom? How should they use it? What do we need to teach them to prepare them for the outside world? So it’s a different perspective than, say, in the corporate world, where the focus might be on efficiency or effectiveness of people in their roles. So I do think we look at it differently. We are trying to prepare our students to learn it and go out in the workforce and be a force for rolling out AI, be a force for making it a tool in a corporate environment that would be helpful. So that’s our mindset, and we look at our students going through the university. What does the corporate culture, what does the corporate environment want from them, and how do we get them there?
Good citizens, profit motives and the values students carry into industry 5:03
Sanjog Aul [00:05:03]:
The way you explained about students. So anyone from academia, and we went through our schools as well, and we were always taught, keeping in mind that we want this individual to be a good student, grow up to be a good citizen, and then do good and make us proud and do the right thing when no one is looking. That’s the benchmark academia always had, and it’s a very noble industry for that matter, because it’s shaping the minds of people. Now let’s come to the industry, where industry invariably has a profit motivation. It has, and I’m not going to trash anyone, but there is a tendency to, as needed, bend the rules to make sure that the end objective of a business is met,
Sanjog Aul [00:05:52]:
and then the students who are going into the industry, they are sometimes up for a rude awakening. Other times they are kind of confused. So even if we kept AI aside for a second, and this has not happened just with AI, it’s been happening before also. Do you calibrate to making these people a good citizen and a good worker and be able to meet the company, the industry’s expectations, or the industry tries to mellow down and say, folks, the ones who we are going to bring from universities and academia, we need to make sure that whatever value system that was ingrained in them, that is not broken in the process of them breaking in into an organization. It’s not an easy question, but I would love for your responses.
Miro Humer [00:06:46]:
Well, the goal always is to produce those good citizens, the people that will go out and do good. That’s the goal. So I think to get there, we teach them the foundational aspects of whatever their area of study is, with a strong dose of ethical use of that knowledge, and they go out into the world and they will have pressures from various entities, various organizations, corporations, where the profit is motive, as you’ve said, but we hope they bring their education, the ethics they’ve learned, and push back on some of that culture. They may not be focused on doing good, but focused only on profit. We hope that they can push back on that, but as with education over the centuries, that’s always been a challenge.
Miro Humer [00:07:46]:
When you get into an environment where the focus is just profit, growing shareholder value, those types of things, how do you insert ethics and doing good and not doing any harm? I think at a university we do the best to ingrain those values, the values of doing good and being a good citizen and having the right knowledge, but everybody makes their own path once they get out into the real world and face those pressures.
Both sides drinking from the firehose: is there a shared North Star? 8:21
Sanjog Aul [00:08:21]:
Absolutely. So one was this ethics part, which you very well responded to. Second is the very skills and the competencies that are required by a person who’s going to enter the workforce and start delivering. Now, number one, there is a lot of fluidity in that, because the kind of skills that you needed yesterday are no longer today, and maybe it’s going to keep morphing. That’s one, but at the same time, even pre-AI era, we heard a lot from people to say what you learned in a school was not applicable in the real world, because perhaps there was a gap in the way how industry moved and how the academia moved. Luckily, in the AI space, both parties are learning and drinking from the firehose. Both of them have to build their workforce,
Sanjog Aul [00:09:08]:
if they’re in the industry, or your students in your academia, based on the curriculum or whatever practical training that you give. If you were to try to say these are two fast racing cars, are they somehow calibrating in such a way so that if they have to come and both have to win this race? I’m trying to draw a metaphor the best I can. Is there a convergence of North Star that’s happening, so that whatever is being taught, and the moment a person goes out for an internship or joins the workforce, those guys are not there in headlights, and the industry is not thinking, I do not know what to do with this individual.
Miro Humer [00:09:53]:
I think this is a very unique time, where both industry and academia are drinking from the firehose at the same time. I can’t remember a time when that’s happened like this. I was in a meeting several months ago, and it was with corporate leaders, so CIOs, CEOs of various industries, none of them higher ed, and the message they had was, we need to implement AI in our industries because we know graduates are coming out and they’re going to expect us to have AI, and they’re going to be trained in it. And in the higher ed world, when I talk to other universities, they’re saying, we have to train our students in AI because these corporations are going to have AI and I don’t expect them to know it. So we’re both trying to figure out what that North Star is. I think everybody, the general understanding is we need to understand AI, we need to know how to work with it, but there isn’t a clear path forward.
Miro Humer [00:11:00]:
Like, what does that look like? What does generative AI in the corporate world look like? Is it for efficiency? Is it for effectiveness? Is it for staff reductions? Where does it fit in? I think we’re trying to learn what our corporate partners are expecting of our graduates. So we meet with them regularly and we ask, what are you expecting? What kind of skill set do you expect our students to have in regards to AI? And one thing that we’ve done is roll AI into over 100 courses on campus. So not treating it as a separate area of study, but as a core skill in every area of study. So AI broadly, in the past, was its own separate program within computer science. Now it’s a piece of every discipline. So, for instance, we have a professor on our campus that teaches finance, and he presents the students with a finance problem and he says to them, solve this within three prompts and submit your prompts.
Miro Humer [00:12:16]:
So it’s in finance, it’s in marketing, it’s in everywhere. It’s how to ask the question, how to analyze the answer. The typical belief is that AI will give you 80% of an answer. What gets you the rest, what gets you the 20%, and what gets you the insight from that answer? And I think there is an understanding that anybody can get you 80% of an answer with AI, even your boss. If you get a job, you submit an answer, it’s 80%, your boss says, well, I could have done that.
Miro Humer [00:12:52]:
I don’t need you for that. It’s taking that and adding the other 20%, the insight, the judgment, the analysis to it.
The last mile stays human: 80 percent from AI, 20 percent judgment 13:03
Sanjog Aul [00:13:03]:
So in a way, what you are saying, what I’m trying to infer from this, is that the last mile challenge is still to be conquered by humans, versus hoping AI will somehow get you on it and you’ll sit on a recliner and let the whole thing happen. Are you seeing that happening over and over in industry and academia?
Miro Humer [00:13:28]:
I think in academia, yes. In industry, I think industry is struggling quite a bit with this, and the consensus is it’s an efficiency play or an effectiveness play. I think we used to say a year ago, the thought was there has to be a human in the loop, and the perception was there’s a bunch of machines and you put one human in there, so that the human is the minor part of that loop. I think what we’re getting to now is that there is a machine in the loop. There is more humans and one machine, or one AI, that’s working in there, and it needs more human interaction than I think was thought, say, a year ago or two years ago.
Sanjog Aul [00:14:23]:
So the acceleration and the creation problem, which they thought that, wow, that we can all sing kumbaya and problems will solve itself, you are coming back to saying, no, it’s a little farther away than what you think. So you can’t have one person sit and 200 machines work at snap of a finger, deliver everything. So the scalability dream, if we had. You’re saying that industry, and you as academia, representing academia, have realized that that’s not at least going to happen in the near future. So you’re preparing for that as a readiness, one of the readiness attributes, that you can tackle AI better.
Miro Humer [00:15:04]:
Yes. There’s an example. A corporation had said that they were going to take their whole, more or less their help desk, and outsource it to AI and eliminate several hundred jobs, and it would be totally AI conversational technology, and they made a big splash in the news. This happened over a year ago, and recently they’ve pulled all that back because the people who called in didn’t want the AI. So for whatever reason, I don’t know the details, but they’ve gone back to people answering phones and not AI. So I think the consensus view is it’s not a silver bullet.
Miro Humer [00:15:54]:
It’s great, it can be great, but you still have to consider where you’re going to use it and where you’re not going to use it, and maybe the human element of it is larger than we originally thought.
Does industry know what it wants? ROI and where to measure it 16:09
Sanjog Aul [00:16:09]:
Second thing which you brought up is you invite industry. You have a kind of industry consortium which you would invite in to say, tell me what you want. But do you think industry knows truly what they want, or you can play Steve Jobs and surprise them?
Miro Humer [00:16:24]:
I don’t think they know what they want yet. A common message that we hear quite a bit is, we’re looking for ROI on our AI investment. Where are we getting the payback from investing in AI? And it’s a broad question. We were paying for this thing, and some organizations are paying a lot for access to AI. Where are we getting the payback for it? Agreed, and that’s a question with any significant purchase, and we’re asking that as well. We’re investing in AI, so there’s the academics of the classroom, and there really isn’t an ROI there, beyond students are expecting to learn it, so it’s hard to measure.
Miro Humer [00:17:13]:
We use it in research, so it helps support the whole research endeavor. We also have administration, and that’s where we actually can do some measurement of ROI. How much time does it save to do some tasks? Does it make our staff more effective in their job? Can they get to 80% of an answer in 30 seconds and then spend the next 15 minutes, 20 minutes getting to 100%, as opposed to four hours and getting to a good first draft? So that’s one area where I think it’s much easier to measure ROI, and I think industry is looking at that. How can they measure it? Where can they measure it? Where does it make sense? So I think, in general, many of us are looking at, is this investment paying off?
Where AI stops and the human takes over 18:09
Sanjog Aul [00:18:09]:
One is to confirm, if what AI produced, is this correct analysis, and shall we go ahead and put our shirt on the line for that? Second is to say that it’s not clear and we are not clear, so let me take ownership and bring some clarity. Like you said, you put three prompts. I know even after three prompts it’s not clear, it’s not giving me what it is. So I need to come up with 4th, 5th or 6th prompt, versus telling AI, give me the 4th, 5th and 6th prompt to help me get clarity on what is the outcome I should even seek before I start going after it. So where is the clarity lacking? Is it that you give it insufficient definition of what you want? I mean, that goes around in circles. Or we just say, okay, I want this outcome,
Sanjog Aul [00:19:04]:
like you said, industry says, I want my ROI. Well, that is an outcome-driven thing, which, as you said, with any purchase. But then how do I deconstruct it down to say, okay, so in what way would you like this outcome to be worthy of your investment? And then you go peel the onion and get to that level, and at what point would you say, AI, you take over? Take me to whichever percentage you can, but now from that point I will take over as human. Is there a methodology or a blueprint or a framework that is coming up across the board? Because it looks like you’re looking for a formula, or we are all looking for a formula, on where AI stops, or where AI stops taking control, human takes control, or they both go together for a little bit, et cetera. I hope you’re getting where I’m going with this. So is there a schematic you could have drawn to how what academia does to a student, what a worker is tackling in industry, and where the AI comes into play to produce the outcome which the business wants?
Miro Humer [00:20:03]:
I think we’re all struggling with that formula. What does that look like? In the classroom, the goal is different. The goal is to learn how to use it, and maybe there’s an aspect of saving time or getting a better answer or getting a better solution to a problem by getting a clear formula of where to use it. I think we’re all getting there, but we’re all struggling with that question. How do we use it? Where do we use it? How much do we spend on it? What’s an appropriate spend on token usage? So we’ve done two years of really science experiments, proof of concepts, and we’ve been doing it at the university, and I think industry has been doing the same thing. They’ll try major initiatives and do some level of measurement on the cost and the benefit, deciding, yes, this is worth it, or no, it isn’t. So we see a lot of projects that go on for a certain amount of time, and they either keep going or don’t, or get canceled,
Miro Humer [00:21:10]:
and I think based on those experiences, we’re trying to build that formula of where does it work, where doesn’t it work. But it’s also complicated by the ability of AI. So things that we’re contemplating today of where to use it, we wouldn’t have a year ago because it wasn’t good enough. So we have that whole thing that the tool is changing constantly.
Who judges the human judgment? 21:37
Sanjog Aul [00:21:37]:
So I would like to pick up on this judgment term. Humans very easily pass judgment on others, that’s a personality. We are also very good at passing judgments on AI’s results. But who is going to objectively judge a human judgment? The person who’s put at the last mile after AI has done its job, to see if, the way you’ve said, AI, you’ve done a decent job or not as good a job, and this is what you need to do, and eventually the outcome is coming out with a decent, or I would say a significant, dose of human judgment. But who is there to judge that individual’s judgment, who has been put at the last mile? Is there some sort of a benchmark, or is it still fuzzy?
Miro Humer [00:22:22]:
I think there has to be somebody. It’s easy to assess based on time savings. So let’s say you have some really basic function, you can do it in 30 seconds, or it took an hour before. It’s easy to assess where it’s analyzing a spreadsheet or something like that. Doing something more complicated, let’s say writing a response to a request for a proposal. So you’re a business, you’re responding to an RFP, and you use AI to do that. It’ll, of course, look at the opportunity, it’ll create a response to it, but somebody with experience, and somebody with senior level experience, would look at that and evaluate it. Is this appropriate? Does it get to the type of response we want to give? So I think you have to have a senior level person at that level.
Miro Humer [00:23:24]:
It’s not always the same. I think it varies quite a bit on what type of work the AI is doing. If it’s entry level work, it’s easier to get somebody to assess that and say, yes, it’s good or bad, or it needs more improvement. The higher level you go, the more senior person you need involved in the evaluation of it. But it is fuzzy. It is fuzzy. I think it varies by situation about who’s the person who’s going to evaluate the response and say it’s good enough or it’s not.
Sanjog Aul [00:23:58]:
When we are trying to assess someone’s judgment, and/or the results for that matter, if they do not do well, as you mentioned, like when you try something obvious, great, but if this was little more complicated, and what they eventually end up producing, whether AI or human, we start losing trust in them. So how much rope should we give to AI, and to the corresponding human counterpart who are part in the same loop, decision making loop or whatever mechanism that you’ve created, before you say we’ve got to replace either or both?
Miro Humer [00:24:34]:
The human that’s, say, applying the judgment to a solution, that’s the equivalent of a manager, and the AI is equivalent to an employee. It would be the same amount of rope you give to a manager in regards to an employee that reports to them. If that employee is producing subpar work, that reflects on the manager. If you’re using an AI and somebody who’s charged with judging the results keeps passing it on, and it’s really just a canned response, you question the judgment of that person the same way you would a manager and how they react to one of their subordinates’ work. So I think it’s a similar arrangement. How much rope do you give your managers would be the same as how much rope do you give to somebody who’s judging the work of an AI and evaluating it to be appropriate or not.
Pressure, timelines and the mindset that survives them 25:34
Sanjog Aul [00:25:34]:
So when it comes to the students, I’m sure you give them assignments. They could be time tested, or it could be complexity, et cetera, like case studies. But then when the rubber meets the road, and when they get to the actual corporate environment or actual industry, that’s when the realism kicks in, the real context kicks in, and I’m sure you try to do the best you can in academia to have them get ready for it. But the real pressure, when their job is on the line and their next paycheck is on the line, the pressure gets real. How much of that, besides of course coaching them, could you simulate in your academic environment? Because now we’re talking not about tools and skills, but the mindset of an individual who’s been expected to produce miracles, and they are given a certain amount of time, and there is a lot riding on it, including their own job, in the context of AI, where things are not fully cooked yet, benchmarks are not fully set yet.
Sanjog Aul [00:26:40]:
Yet there is expectation from a worker who comes into the workforce and start producing, because they say 80% of the work is done by AI, you’re supposed to only do 20% of the work. So the expectation, or the patience, is also running low in that regard, because they say fire small, fire fast and just move on. That’s becoming the mantra in today’s day and age. So what do you do?
Miro Humer [00:27:07]:
Well, definitely there’s pressures in the corporate world, and there’s timelines, and quite a bit of pressure passed down to somebody working with AI trying to get to a response in regards to a specific situation. But don’t forget, in higher ed there’s also pressures. There’s also timelines for projects and assignments. There’s also assessments, tests, things like that. So there is that element of pressure in the classroom as well, and I think they’re more harshly judged. Are they just taking the easy way out with AI? So there’s a higher expectation of applying some of those ethics and judgment to how it’s used, how it’s presented to the faculty member, and the faculty member will be judging it on many more aspects than somebody in the corporate world would.
Miro Humer [00:28:03]:
I think the judgment in the corporate world would be, are you meeting the target of the project? Is this going to generate more income or make me look better, or whatever? So I think the range of judgment is narrower in the corporate world, and ultimately will lead to more profit. In higher ed, there’s pressure there, but the focus, the judgment, is on more elements. Did you meet the targets, whatever the goal was? Did you do it ethically? Did you apply a good amount of judgment to it? Did you add to it and show us how you did it? So hopefully that experience will teach them how to deal with the pressure and how to do it appropriately and effectively without giving in to that pressure to just do a quick and easy and move on.
Chronic gaps and how academia can lead industry 29:08
Sanjog Aul [00:29:08]:
Are there any specific areas you would think are still chronic in nature, and/or happening more often than usual, that you might have worked very hard, your professors, your whole academic system, to put somebody out in the workforce, but before that they have grilled them, they have taught them, they have to try to work on their mindset, but then invariably, when they go in the workforce, and especially now in the context of AI readiness, there are some deltas that keep showing up, and either they could be chronic issues or they could be something which could be solved?
Miro Humer [00:29:43]:
I think because these are relatively early days with a broad use of AI, I think there are some chronic issues around its use, its ethical use and appropriate use, and not applying maybe as much judgment as there should be. But I do think that’s driven just because it is early days. As we learn more, as we see the cases of legal briefs that are written by AI that cite cases that don’t exist, we learn from that, and we learn that there has to be more judgment involved, that there has to be more evaluation of the response. I hope that those situations, where there may be too much pressure on using AI to get to the easy answer, I hope that starts fading away as we learn more about it, and where it can do well, where it can’t, where the human takes a bigger role or a smaller role. I think as time goes on, we’ll learn more about how to do that better.
Sanjog Aul [00:30:51]:
One was that industry was trying to guide you, as I’d asked question earlier about you go ahead and ask the industry what to do. Are there areas in which the academia has responsibility and an opportunity to lead the industry in AI, so that the final outcome of AI readiness across the board improves overall?
Miro Humer [00:31:13]:
Our hope and expectation is, as our students come out of the university having had over 100 courses that have AI as a significant element in them, in various disciplines, that when they go to industry, they become the leaders in this effort, that they become the force that pushes us forward. That’s the hope. I think every university has that opportunity right now, to produce the graduates that have had experience in their discipline with AI, how to use it, how to be effective with it, and how to push the organization forward when they get into the workplace.
Sanjog Aul [00:31:57]:
Are there specific areas, though? So what I was trying to get to is, suppose you’re preparing students and they go into the workforce, 2025, 2026, and you are doing some things proactively, whether a secret sauce or openly made available, that this is how we would prepare these students, so in specific skills and competencies and experiences and scenario-based thinking and solutioning, that they could ace, or rather go even above and beyond a regular worker who might be learning about the same thing in an industry setting. Do you have an advantage somewhere where, if you sent a student in, they could right away, not take over, but do way beyond an average person in the industry? Can you help industry versus only supporting it? Can you lead the industry?
Miro Humer [00:32:53]:
Yes, absolutely. I think the way we’re teaching AI in the different disciplines, it’s on the leading edge of how it’s being used, in finance and marketing and all the business courses, as well as engineering, as well as medicine. So in all these areas, the direction is, how do you ask the question, how do you use AI to get to an answer based on the discipline you’re in? So they’ll have experience in those areas in how to solve problems, how to apply judgment, and get out into the workforce having had more experience with more difficult problems. I think absolutely our students would be leaders out into the working world, having worked on case studies and different problems that they’ve been presented by their faculty.
Lifelong learning as the real AI readiness 33:55
Sanjog Aul [00:33:55]:
So if you were to look at the type of people you want to put out in the workforce, and knowing that AI is just getting started with it, and one of the reasons, people in many industries, except IT, they say, okay, I’ve learned this, now I’m going to apply it for the rest of my life. Not everybody is a learner by design. But if someone who you are trying to teach AI to adopt it, that means you are in a way saying, there’s an unsaid rule that you will never stop learning, because the very skill you were asked to apply, that itself, the foundation, it might keep getting destroyed creatively with a new version coming in, and you’ll have to learn all over again. So you can teach people the skills, but what about that attitude of becoming a lifelong learner?
Miro Humer [00:34:46]:
Absolutely, I think that’s critical. Especially, as you mentioned, in IT. Today it’s generative AI. What’s it going to be in five years? I think that we have to instill the lifelong learning aspect of it, and I think in the corporate world we have to reward the behavior of learning, not necessarily the behavior of gaining something, like you’ve gotten a degree, so you get more money. It should be about the process of learning constantly, and rewarding that. I think it’s important to ask employees, what did you learn this year, in the review process? What are you learning? What are you working on? What are you doing to get better? I think it’s a difficult concept to get across in many areas, but it’s critical, and AI is pushing everybody to at least consider that, because it’s touching so many different aspects of the working world. And I think the people that don’t accept that, I can learn this and be better, if they say, okay, I’m fine, I’ve been doing it this way, I don’t need to learn anything new,
Miro Humer [00:36:18]:
they’re the ones that are going to suffer the most in this environment. I’m sure you’ve heard the saying that AI is not going to take your job. A person who knows AI is going to take your job. And absolutely, I think that’s true.
What true collaboration looks like: proofs of concept together 36:33
Sanjog Aul [00:36:33]:
You are in a very interesting position. You’re a CIO, so you are, while not in industry, but the business side of your university, plus you get to see the student side of that university, and you are your own customer, if you wanted to. You can absorb the students coming out of your university. So what should each side do to help each other? So both of them are very anxious, both at the management level, at the academic, or people who are teaching or learning. So we say, help me help you. What should that look like? What does that true collaboration look like?
Miro Humer [00:37:16]:
I think there’s a lot to be said for doing proof of concepts together. Having industry come into a university and say, hey, we have this problem. Put your students together into a team, see if they can address this issue and do a proof of concept for a corporation. I think they can learn, we can learn what their problems are, and they can learn from students who are in a learning environment and have been focusing on that industry and technology in that industry. I would like to see a much better, much closer connection between industry and universities. I think there’s a lot to be learned one from the other, and I think it would help us produce the kind of graduates they need, they want. Not too long ago, my son graduated with an IT degree from a large university, and in their IT program they were still teaching COBOL.
Miro Humer [00:38:31]:
Now, for non-IT people out there, COBOL was considered old back in the 80s. I don’t know how that program aligns with the needs of industry today. So I think it’s important that we keep aligned with what industry needs, and I would like to see industry coming in with a problem, saying, how would your students address this? How can faculty get involved? Bring students together and put together a solution that we can try. Does it work? Students get experience in real world problems, or corporation might get a solution that works, or at least ideas that could lead to a solution. I would like to see a much closer connection. We’re lucky, we have quite a few corporate partners that come in, but I think it would be nicer to have more come in, with different industries, different problems.
What must change in the leader 39:24
Sanjog Aul [00:39:24]:
When you look at the leadership, because this looks like change management, if you notice, and that means the responsibility lies with the leaders, and if something is not happening, the fingers still have to be pointed back at the leaders. So if this AI readiness gap needs to be reduced, or hopefully eliminated, that both parties are working along well, and if you were having lunches with your peers in the industry, you are the industry, for example, other academic folks and the industry folks were to have lunches together, and/or lead their respective team, or lead each other, or lead themselves, what should change in the leader today? Because that’s where the buck stops, and that’s where actually everything starts.
Miro Humer [00:40:16]:
I think what would be a good step forward is more curiosity and more openness to looking at different areas and different ways of getting to a solution. If you’re looking at AI readiness, if I’m running a $500 billion industry, or a company in such an industry, how do I accelerate that AI readiness? How can I get from just playing around with it to getting a real return on investment with it? I think in this environment, with what’s going on now, you have to be a little bit more creative, open to different solutions, and be curious. Reach out to universities, and universities reaching out to corporate partners. I think it requires just that curiosity. I’ve heard from some universities that they don’t want to use AI at all because they’re afraid of the whole cheating aspect. I’ve heard of corporations saying we’re not going to use it because we don’t trust it, and neither of those answers are appropriate. You have to move forward, you have to be curious, you have to learn about it,
Miro Humer [00:41:32]:
and being willing to look at different methods of solutions. I think it’s a different world, it’s a different environment, it requires different, broader thinking.
Closing 41:49
Sanjog Aul [00:41:49]:
So thanks so much, Miro, for sharing the thoughts here, because the AI readiness is actually very, very foundational and very critical, as you and I both know, and we’ve discussed about it. The goal here, and which was very well achieved, is to talk about those nooks and corners where the academia and the industry may be having some gaps in alignment and/or focus and/or effort. But what we discussed here today had a lot of nuggets, and I just hope that our audience, wherever they are, any part of the world, and if they are working on AI readiness gaps, they get those nuggets and use them, and as a result we have, starting a student all the way to a worker, to a leader, they all come in alignment and create the most value, rather harness the most value from AI. So thank you so much, Miro.
Sanjog Aul [00:42:46]:
And our audience, I hope you got a lot of value here. I enjoyed it, and so please subscribe to our channel, social and content channels, and this is Sanjog Aul, your host, signing off. Till next time, take care and God bless.


