The rapid adoption of AI is creating a significant skills gap, with 69% of business leaders acknowledging its impact, yet only 34% actively addressing it — a clear say-do gap, according to research from DeVry University. This disconnect leaves organizations vulnerable to inefficiencies and missed opportunities. Dave Barnett, chief administrative officer at DeVry University, emphasizes the importance of aligning leadership priorities with workforce development strategies. By overcoming barriers to upskilling, integrating training into workflows, and fostering a culture of adaptability, organizations can empower employees to thrive in an AI-driven future.
The rapid adoption of AI is creating a significant skills gap, with 69% of business leaders acknowledging its impact, yet only 34% actively addressing it — a clear say-do gap, according to research from DeVry University. This disconnect leaves organizations vulnerable to inefficiencies and missed opportunities. Dave Barnett, chief administrative officer at DeVry University, emphasizes the importance of aligning leadership priorities with workforce development strategies. By overcoming barriers to upskilling, integrating training into workflows, and fostering a culture of adaptability, organizations can empower employees to thrive in an AI-driven future.
Subscribe to The AI+HI Project to get the latest episodes, expert insights, and additional resources delivered straight to your inbox: https://shrm.co/voegyz
---
Explore SHRM’s all-new flagships. Content curated by experts. Created for you weekly. Each content journey features engaging podcasts, video, articles, and groundbreaking newsletters tailored to meet your unique needs in your organization and career. Learn More: https://shrm.co/coy63r
This episode is sponsored by Robert Half.
Ad Read: [00:00:00] HR leaders know hiring is getting harder.
A new survey from Robert Haff reveals companies are hiring for critical roles
Ad Read: making competition for top talent fiercer than ever.
So how do you stay ahead? With insights that matter?
The newly refreshed demand for skilled talent report from Robert half dives deep into hiring trends and the impact of AI on critical skills and staffing needs.
Get the data that drives smarter hiring. Head to Robert half.com/demand for skilled talent.
Nichol: Welcome to the A IHI project.
Nichol: I'm Nicole Bradford shr, M'S. Executive and Residence for AI plus. Hi. Thanks for joining us for this episode. We're tackling one of the most pressing challenges facing organizations today, the growing AI skills gap.
And its [00:01:00] impact on workforce readiness. As AI continues to transform industries, many employers are struggling to upskill and re-skill their teams to keep pace with innovation. Our guest is Dave Barnett, chief Administrator Officer at DeVry University, which recently released a report shedding light on the barriers employers face and actionable steps HR teams and other leaders can take to address these challenges.
Dave, welcome to the A IHI project.
Dave: Hi Nicole. Thank you so much for having me.
Nichol: so I am obsessed with research and I heard about your report, so I wanna dive in. Can you briefly summarize the key findings of Dev RISE's report and explain why addressing the AI skills gap is so urgent for organizations today.
Dave: Absolutely. So we've been conducting research for a few years, very specific to the topic of upskilling and reskilling, and there's a few things we've found [00:02:00] that are both exciting and alarming. They're exciting because we can address them, but they're alarming because if we don't, they can create real risk for organizations throughout the US and the globe.
Um, one, we found a pronounced say, do gap. And what that means is we see high affinity for upskilling and reskilling workers report very aggressively that they want upskilling, they want reskilling. But when we look at the percent of folks actually taking advantage of such things, that number is far, far smaller.
So there's a break between what people say they want and what they actually take action on. We found that ties to a number of barriers. When we get specific to ai, we've also learned some really interesting things. One interesting stat is employers say 32% of workers are novices in ai. However, when we ask workers that same question, only 3% report that they're in the novice category and nearly half say they're proficient or better at [00:03:00] ai.
So big gap in what we see employers experiencing.
Nichol: What
causes that? What, what is that gap?
Dave: That's a great question, right? I believe, um, we tend to overestimate our skills, right? And so as we start to play with something, as we see small successes, we tend to overestimate how effective we are at using something. We get a false confidence in the short run.
Um, but that may not mean true proficiency in terms of driving business impact or leveraging things within the right guardrails. Um, and so I think there's a perceptive gap between the two audiences for certain.
Nichol: yeah, that lines up with some of the research that we've seen with SHRM, where we've noted that 51% of US workers prioritized enhanced training and upskilling for improving their AI outcomes. But, and nearly half of them also. Are insecure about their jobs without re-skilling. But what was interesting about what you said is, [00:04:00] you know, the people who, like, they feel like they're proficient, but they aren't actually proficient.
Dave: Yeah, I think that's right. And I think some of it comes down to context, right? I like to equate talking about upskilling and reskilling with something like AI of, at times we talk about teach ai, when that's sort of the equivalent of saying to someone, teach Hammer, right? We wanna teach you hammer.
You can't teach a tool.
Instead, what we teach is the skills and application that surround that tool that make it ultimately useful. And so I do think when someone says, I'm familiar with ai, I can put a prompt into chat, GPT, I can use an AI overview in Google by asking a question, we start to feel that we have pro because we've touched the hammer, but yet we don't know how to use it to do something meaningful and productive in the context of our work life.
Nichol: You know, I, I, I really truly [00:05:00] relate to that because my father was a plumber.
Okay.
And at one point this led me to believe that I could plumb. Yeah. So I tried something, it did not work out. So, but I, I truly like, I felt like I knew what to do because I'd seen it done so many times and it was not true. So, yeah.
I, I really relate to that. Like, that was me. Um, so Debra's report also reveals that 72% of employers aren't offering AI upskilling benefits to all workers. And so the question is, why and what challenges are professionals facing in addressing that gap? And, and is there something about, how do you, did you see anything about how organizations are deciding who gets it and who doesn't get it?
Dave: Yeah. Again, there's a few things we've seen. I think one is we do see organizations disproportionately giving it to those folks that show a higher career trajectory. And career [00:06:00] trajectory was one of the top predictors of someone being offered AI, upskilling and reskilling. Um, so we see folks tying it to high potential employees, right?
That we're going to give them AI, upskilling and reskilling. Um.
I think there's a struggle at times because we're still trying to define how AI shows up at work, and do I think we're better today than we were yesterday and better yesterday than the day before? Most certainly. But I still think we've got an interesting tool that we know creates a superpower, and we know that superpower lives best when it's matched with a bright, intelligent individual who can leverage that to unlock their potential.
But we're still looking for the direct business cases to apply it. And so. As we see AI as a separate initiative, as a separate
vertical, it's difficult to know who to train. It's difficult to democratize that because we've not yet woven it into the jobs. I think the next step for HR professionals is to go
beyond
AI training AI [00:07:00] upskilling, and instead move AI into the heart of the work and train people on those workflows, train them on how they conduct work in that new AI empowered manner.
Nichol: Oh wow, there. There's so much there too. Because I mean, when you were answering the question before, one of the things that crossed my mind is not only do we, are we learning how to use it in organizations, so we don't exactly know who to train, but do you feel like we have clarity on what success looks like?
How do we know when we've successfully, you know, upscaled. You know, a, a ai educated someone like how do we know if it's not in a project that's tied to ROI?
Dave: That's right. And again, in that separate world and that separatism, we can measure knowledge transfer, we can measure, can someone repeat a behavior. We can measure their level of knowledge output around ai. But until we integrate it into workflows, what we can't measure yet is those [00:08:00] impacts on the workflow.
Can we do things more effectively? Can we do things more efficiently? What are the unlocks of ways we can point human intelligence toward different challenges in higher order thinking? Because we're able to either automate tasks, we're able to uplevel thinking to a new level because data sources now are massive and able to be integrated through large models in different ways.
And so I think for us to measure effectively, again, it comes down to what are the objectives of the AI initiative. What are we trying to move and then how do we see those impacts come through?
Nichol: Could you share with our audience who might not be familiar with DeVry? Who are the types of people who are at the university that you're also considering when you're doing research?
Dave: Yeah. So DeVry University's been around since 1931.
Nichol: I remember the commercials when,
you know, when I was a kid even. Yep.
Dave: I'm a product of Chicago, and so Ry was a great Chicago brand. Um, [00:09:00] but since 1931, we've been creating access to education for people who should be able to participate in technology powered careers. And it looked very different in 1931, right?
We had oscilloscopes on benches then, but in today's world, it's still that same idea. We believe everyone should have access to education.
Everyone
should be able to find ways to participate in tech powered jobs to move our economy forward. And so that's who our learner is, our learners, that person who maybe wasn't well served the first time, they had tried higher education, they weren't well served by the traditional model.
They're working and raising a family and going to school, and they're looking for that next step for themselves and they wanna make sure they maintain relevance, they wanna be relevant to the workforce of tomorrow. And that's the student that we think of as we build AI into all of our curriculum.
Nichol: are you seeing a surge of people realizing or coming to you, saying, coming to the university, saying, oh my gosh, like I need these skills. Teach me, teach me.
Dave: We do, we see, we've seen great growth as an [00:10:00] institution. Um, we see a lot of students who truly. Want that tech centered education and they want it in different formats. So we have students who come to us because they want long form learning. We have students who wanna come to us for certificates or shorter form learning.
Um, and we even have a new product of Ry Pro that looks at AI learning. We have an AI for Leaders course and an AI for analysts course that is self-directed, taught by an AI coach. So actually, AI is doing the teaching
of
AI
Nichol: that's interesting. Yeah. I love democratized education and democratized access to these tools because AI is, you know, it, it's kind of, you know, you're right.
We can't say teach ai, it's like teach Hammer, but AI is like electricity and.
Dave: sure
Nichol: Everyone has to know how to use it. Um, and so I love what you're working on. One of the things that I've heard is that 42% of employers feel uncertain about their ability to train [00:11:00] employees on AI technologies. What are the factors driving this lack of confidence, and what steps can HR teams and leaders take to develop a stronger, more effective framework for training ai?
Dave: Yep. I think one of the biggest factors is that AI's moving at such a fast pace. Right. And so as organizations try to change it, they're, they're learning themselves and what they've learned. Tuesday may not be relevant by Thursday, and so there's such pace in AI right now that if an organization is doing what they should do, which is their core, they're driving the core of what they do.
They can't always be fully watching what's happening and developing on the AI front, or finding the right ways to integrate as they're still in a learning curve themself. I think one of the best strategies an organization can employ is find a good partner.
Find somebody who does this as their core function that can work with you and help in
all
of
those
baseline, direct things that are, how one [00:12:00] implies ai, both from a competency and skill perspective.
So things like how do you engineer a prompt? How do you effectively and clearly define a problem through design thinking. How do you effectively evaluate the output of AI through good, solid, critical thinking and refinement of the output so that you can make something better? All of those skills plus the hard skills of touching the ai, let an education partner bring that to you.
And as the
employer,
you work on two things. One, how do you make it relevant? How do you link that? Then directly back to the things you are doing each day, and bring that relevance for your learners. And two, how do you make it social? How do you make it part of who you are? How do you connect people through dialogue?
Get them thinking about AI and building new ideas and thoughts? Because the reality is we're all figuring it out. So drive that social learning within your organization, but let the heavy lifting be done by someone who
does it all day long
Nichol: a, a couple of things. One of the [00:13:00] things that I share with. Um, people often is that, you know, the, so like the very large language models and you know, the big foundation models.
So for our audience, those are the chat GPTs of the world, et cetera. Um, it doesn't start creating value in the organization until you've got your context, your culture, and your customer. And until you do those three things, it is generic. The company right down the street has the same thing. There's no differentiation until it gets into that.
And so, to your point, you know, just as a few years ago, we all learned to pandemic together. We're all learning to AI together. I was asked, I was doing a. A session a little bit earlier and someone was asking me for cases around a couple of things and I said, well, you know, there's some cases, but unlike previous changes, there aren't huge [00:14:00] cases of people who have really, you know, uh, of organizations, tons of organizations that have done many, many, many implementations because it's all new.
It's all new and it's changing nonstop. So in that context, when you have this rolling wave of change, what are the most pressing AI skills gap? Like, you know, we have to do some kind of planning. So
what are the most pressing AI skills gaps that you're seeing across industries in terms of both technical skills and durable skills, like critical thinking and adaptability.
Dave: yeah, I think, you know, one of the things we see, and you hit on it some from a technical skill perspective, really effective prompt engineering comes down to be repeatedly something that we see as a big opportunity. Effectively knowing what types of prompts and how different nuances and changes can impact the output of AI is a really [00:15:00] important skill set.
And so we see that from a technical perspective, um, as an absolutely key area, um, from the durable skill perspective, those skills around it. Um, one design thinking. And so being able to thoroughly flesh out a problem and get to a clear. Crisp problem statement that we're working toward, we think is a really important skill that will become even more important in a world of ai, um, critical thinking.
And so, it's unfortunate that some of the early signs we're seeing out of MIT around AI is that people are thinking less in some ways when they get an AI output that the brain is firing in fewer regions when they get that output. 'cause
it's taken at face
Nichol: yeah, that's out of Patty May's lab.
Dave: Fantastic. And it's, it's important research because '
I think
critical thinking is so important as we get an AI output to say, how does this make me think about the next question? How does this make me want to refine this to the next [00:16:00] best level of information and evaluate it before I bring it forward?
And so we, we see critical thinking is a critical skill. Um, and then third communication. Um, becomes even more important because someone needs to be able to wrap around context and storytelling behind what the AI output is. And so those are the places where we're focusing and see a lot of, a lot of need.
Nichol: Yeah. You know, I, so I use AI every day and I am constantly learning and I will come across the way someone else or you know, has prompted something. And I will think, even though I use it every day on everything, and I will think
I
prompt like a baby,
you
know,
like,
because.
You know, the, the evolution of it.
And I, and I think also one of the other benefits of really learning how to prompt effectively, whether it's written or vocally, um, is that I think it helps [00:17:00] people really understand the importance of their domain expertise. Because if you, if you're. You know, if you, if you aren't leveraging your domain expertise, when you're looking at these outputs, it, it easily can be slop.
And if you don't know, you know, without your domain expertise, you wouldn't know it. So when, when I'm working with ai, I'm constantly adjusting the prompts as I'm moving through, working through a problem
Dave: I do the same, and I'm not great by any stretch, right? But to continue asking a question differently, shifting the prompts, or allowing the output to lead me into actually saying, actually my thinking's now evolved. My question is really something different. You know, I love those moments when it refines my own thinking and creates a genuine insight that helps me think about something different.
And leads me to the next question.
Nichol: Yeah. Um, so how can HR teams and organizations [00:18:00] understand the gaps within their own organization so they know. Which gaps to fill? Like what's the process that works for
that?
Dave: Yeah, I think some of what works best in a case of ai, and it's interesting to say this 'cause AI is so new and exciting. But it's the things that have worked incredibly well in org effectiveness and l and d for an incredibly long time. Right? So I think it comes down to having clarity around objectives from an AI implementation or an AI evolution within an organization, and getting crisp and clear on the behaviors you need to see individuals demonstrate to show mastery, to show success, and then measure those.
Measure those through surveying, measure those through observation, measure those through performance data review. But those same things we've done before can unlock where we see the actual gaps at a more micro level, right? Beyond hammer, we're getting to what portion of the work is it that we [00:19:00] see that we need to upskill and train.
And so I really do think it's about asking the question and going
to
the
organization
for data.
Nichol: Something I'm curious about and, and because like you just, you have a wonderful vantage point being at DeVry
and
also the way that you partner with other organizations. One of the things that HR leaders are facing is that. They need to do workforce planning today for jobs that they know the task and skill composition will be different in three to five years, and so they have to be in the present and the future at the same time.
How do they do that? What? What would be your advice be to someone who's in that situation?
Dave: We have that conversation a lot, and I think it's something that everyone is grappling with right now. I think it's a few things. I think one, [00:20:00] first and foremost, we have to shore up the work of today. As an enterprise, we have to be highly effective at what we do right now that earns us the right to do what we do tomorrow.
Right? And so that's step one. Step two is we have to be
bold.
Envision the
future. Clearly that's hard. That's really hard work. But we have to vision a future model. That's an aggressive view of how the work may look in three years.
I wouldn't go
further,
but can
we
get to
three years? And then we've got to walk that back to the requisite skills that get into that and start to develop that baseline so that we can drive that metamorphosis.
So you've got to fortify today and then upskill and start to re-skill people toward tomorrow in small ways and invite so they can start to employ and use those skills. If we teach a bunch of everything right now and say in three years, this will be really important. Learn it all. It's just like a workout.
We wind up with atrophy.
'cause you're not regularly using those [00:21:00] muscles. So you've gotta start to build just in time training and feed that in as the organization goes. Equally important is creating a culture and environment that rewards, experimentation and risk. You've got to create space for people to try to do and do differently within ai because if they are learning something but they don't feel safe stepping out and doing it, they're learning something and they're not rewarded for demonstrating that behavior.
Again, it all goes away and so it's nuanced. I wish I had a clear magic answer, but it's clear picture of the future. Paint that compelling vision and then gradually walk people through that journey through progressive upskilling and reinforcement of risk taking behavior, and moving into the
new world.
Nichol: Yeah, and also probably a iteration and adjustment too.
Like if it's not, you know, giving things long enough to know and see if you're meeting outcomes and being [00:22:00] very di diligent on being data oriented. And also if, if it, if this particular thing isn't working, trying something adjacent. You know, in baby steps, like you said, because, you know, one of the things too, so I, I have, I have, um, I have one major pet peeve with, um, the way that many technologists think about technology and, and I'm, I'm from the technology side, so this is with my own people, is that I feel like there's a great deal of ambition for tech.
And so if you read the headlines, there's so much ambition for AI is gonna do this, AI is gonna do that. But there is so little ambition for people. And so what I love about what you said is, you know, earn the right today, fortify today, but you know, have a compelling vision. And I think that compelling vision also has to be, we have to get ambitious for
people, [00:23:00]
Dave: We do. They deserve that. I mean, people deserve the right to, to aspire to something better. That's motivating and exciting. And I love your point about adjacencies. The risk in building out a three year vision is that the only thing you can truly know about that vision at the time you write it, is it's wrong.
Right, and so you have to be able to pivot and move as you continuously
learn
Nichol: Yeah. And, and also before we talk about budgets, 'cause that's my next question, it's also the. You know, it's, it's really clear that many of today's tasks are going to be done by ai. We can see that, and the big question is where are the new jobs? What are the new jobs? And what it is, is that the new jobs are in this space where humans and machines or humans and AI do things that they couldn't do before.
Without that, we couldn't do without it. It's not actually possible for people to [00:24:00] jump ahead to that. Because if you look at the list, often when you look at the list of, you know, these are the, when people say these are the next jobs, the number of times I look at that list and think, well, I don't wanna do any of that, you know, it's like, it's not compelling.
What it is, is that we're actually gonna have to iterate our way to that space of
things that only humans and and AI can do together. And it's going to be, um, taken care of today, experimenting, failing, adjusting, iterating, and so we're going to have to iterate out to that new space where the new jobs are.
Dave: I totally Yeah.
Nichol: Yeah. a hundred percent. You're great.
So, okay,
so budgets, you know, lots of people are listening and, you know, some organizations have mega budgets, but a lot of people, they, they don't. So an organization with limited budgets, what are, what's the cost effective [00:25:00] strategy or, and resources that they can use to do this?
Because everybody has to use electricity. So if you have a limited
amount,
how do you do the
transition?
Dave: Yeah, I think it's a great question. Um, 'cause certainly no one was planning for a massive revolution in the workforce, um, you know, just a few years ago. Yeah. Um, but it's upon us. And, and that's the reality. I think there's a few ways that organizations can effectively manage the change.
Um,
one
is super, users are a phenomenal tool, right? And so as we talk about how do we reach the masses. I said, working with partners is really important. I still believe that, but how do you go deep with certain individuals that are influential and well positioned within your organization to help you lead a grassroots charge to spread those skills and those changes throughout the organization?
So I think one is organizations can and should look at a super user type model. Um, two, there's a continuum [00:26:00] of learning options, right? And so individuals can pursue degrees. Um, that will bring AI to the center and forefront all the way through things like I mentioned, our DeVry Pro Pro platform, which is a simple five week platform where individuals can learn targeted AI skills in an asynchronous format.
That's a big range, right? And so being thoughtful about what individuals need to learn at what levels of their learning ecosystem can be a really smart way to conserve dollars as well by putting people through the right places of the training program. Um, and the third answer that we bring to all of our partners that we talk about is organizations have tuition assistance budgets.
Those aren't budgets that always work as hard as they can for an organization. Many organizations have tuition assistance programs because it's what you do. It's the right thing to do. Those dollars. Can be strategically deployed to higher education institutions to partner with you, to help you effectively drive these skill sets.
Um, and so that's another way to [00:27:00] leverage perhaps unlocked pieces of the budget that may not be working as hard for you. Um, or those dollars can work harder.
Nichol: So what are some of the pitfalls
that
you've seen organizations face when they're upskilling around ai and how do they also ensure that they're being inclusive and accessible to all?
Dave: Sure. You know, I think one of the biggest pitfalls that I've seen is organizations who have a bit of a
cultural
identity
issue
around ai. And so I think setting clear guardrails and clear expectations for AI use is really important as any organization goes on an AI upskilling journey. And so. Being clear of what does good look like, what's out of bounds, what's squarely inbounds?
Because I've seen organizations say, we need to use AI more. I want us to be an AI shop. I want everybody [00:28:00] leaning
into
this
revolution. And then the moment they take a breath and speak again, it's, that's ai. I can tell how dare they submit a piece of work that was AI work, right? And so that duality is really confusing for an organization.
And so we have to be clear. About what does it mean to effectively use ai and what are places that are out of bounds, right? When we say, I can't believe they used ai, what we really mean is they applied no critical thinking to that AI output to make it their own right. And so words matter, guardrails matter.
And so I think organizations hit a pitfall where their employees just don't know how to move forward.
' cause they want to participate, but then they feel that there's risk or negative consequence to participating. So I think words matter and absolute clarity is really important from a cultural perspective.
Um, the other piece is being clear about how we train and skill people. We can't assume that all AI is intuitive. Um, we have to train people with context. How do you use [00:29:00] the hammer in this situation when we're trying to build a house? Or a shed or break a window, right? Very different. Things all have uses and so we have to apply context and be really clear in the skills that we're bringing forth to folks.
Nichol: in an increasingly automated workplace?
How can HR teams and leaders help employees embrace AI as a collaboration tool? A lot of what's going on is people are using AI independently, but when you look at the research on productivity gains, it's about collaboration. So how can we help people
use it as a collaboration tool?
Dave: Yeah, I think it really comes down to something that I said earlier and that's that we should put social learning around ai. Right. And so when we teach people skills, oftentimes it is go off and learn a skill, and we teach people to do that on their own. It's rote, it becomes mechanical. I think the biggest way to make AI collaborative is to put it into social learning opportunities and use it to [00:30:00] spur conversation and thought, not just be a standalone output.
Nichol: And then for HR teams and leaders who want to address the, the, like they've heard this conversation and you know, the audience has heard this conversation and they're going back Monday, what do they do? To address the skills gap with ai,
Dave: Yeah, I think step one is get crisp and clear around how you intend to use AI today. What are the use cases? What are the places you want that to show up? Um, step two is get clear about that bold vision. What does three years out look like? And be willing to move from that. Um, the next step, understand the skills
of your workforce
in the context of both of those,
and
then as you define that skills profile, lean into those individual upskilling places so that you could skill in a targeted fashion, the place where your workers need different skills to [00:31:00] use this phenomenal, new, powerful tool.
And then finally find the right partners. Find partners to help you do this, who are spending their time thinking about this all day. Don't feel as an HR professional that you have to have all the answers to what is truly
a workforce
revolution that we're in right now,
Nichol: Dave, I could talk to you for a very long time. You are so insightful about this key issue that our HR leaders are facing right now. This is the most important thing, and so thank you so much, and that's it for this week's episode. A big thank you today for sharing his experiences and insights with us. Did you know our podcast is only one part of the AI Plus HI project. We also host an annual event featuring visionaries from the worlds of business tech and hr. Plus, we publish a weekly newsletter that includes topical articles and the latest research from SHR RM all. To help [00:32:00] you stay ahead of the trends.
To sign up, just make your way to SHRM.org/aihi. And you can also follow SH RM on social media for even more EC clips and stories like share and tell us how you are innovating towards a brighter future.
Ad Read: Hiring is evolving is your business.
Explore the newly refreshed demand for skilled talent report from Robert Half to see how AI is shaping the future of hiring. Visit robert half.com/demand for skilled talent.