Videos
Venice Goodwine on Workforce, Urgency, and the Human Side of AI
Key insights
- AI adoption in defense starts with psychological safety. When personnel hear AI, they hear job replacement. Goodwine argues that CHCOs and HR leaders, not technology teams, own the responsibility of communicating how AI fits into each role and why the human remains in the loop.
- AI literacy is a national security issue. Domain expertise is what allows operators to catch hallucinations and drift; without it, the model cannot be trusted, regardless of the underlying architecture or security controls.
- Agentic AI belongs on the org chart. Goodwine’s argument is that when a non-human entity has a job description, a defined scope, and accountability, the workforce becomes more comfortable with it, and the organization gets a multiplier that does not need a lunch break.
Recorded at the Mission Ready AI event in Melbourne, Florida, this conversation between Jeff Kavanaugh, Head of the Infosys Knowledge Institute, and Venice Goodwine, CIO and AI Product Owner at Arlo Solutions and former CIO of the U.S. Air Force and Space Force, examines what it actually takes to move AI into production in one of the world’s most demanding operating environments.
Goodwine argues that the urgency now visible across the defense industrial base is not rhetorical: the adversary has access to the same technology, the pace-setting projects are public, and the funding is committed. What remains is execution, and the primary barrier to execution is not technical. She explains why workforce psychology and AI literacy have to come before any tool goes live, how cross-functional teams with domain expertise are the real safeguard against model drift and hallucination, and why she now tells organizations to put AI on the org chart with a job description and accountability. She also draws a direct line between the EU AI Act’s risk-tiered framework and what responsible deployment looks like across sovereign boundaries and addresses the next generation directly: the countries treating AI fluency as a mandatory skill are building a structural advantage, and curiosity is the starting point.
Three ideas from this conversation
Psychological safety before deployment
Goodwine’s argument is that workforce anxiety about AI is not irrational. It mirrors what happened during the RPA era. The answer is not reassurance but clarity: explain exactly how AI will operate in each specific role and make the human’s continued responsibility explicit.
Domain expertise as a trust mechanism
A generative model will hallucinate. The check on that is not the model. It is the operator who knows the domain well enough to catch the error. Building cross-functional teams with genuine subject matter expertise is Goodwine’s practical answer to the trust problem.
The NHI on the org chart
Non-human intelligence entities assigned specific roles, held to defined outputs, and subject to replacement when they underperform. Goodwine’s framing gives the human workforce a conceptual model for working alongside AI that does not require pretending the risk is not there.
Q: How is AI investment changing the defense industrial base?
The administration’s AI strategy, the Secretary of War’s pace-setting projects, and the funding behind both have given the defense industrial base a clear signal: deliver AI-enabled capabilities, not tools. Goodwine explains that the requirement is now for capabilities that help warfighters make faster, better decisions, and that industry cannot treat this as separate from the mission.
Q: What does AI urgency actually mean when the adversary has the same technology?
Goodwine is direct: everything built for good can be repurposed by an adversary. The answer is to outmaneuver, outmatch, and outsmart by moving with AI literacy, guardrails, and accountability built in from the start. She also notes that urgency has a different clock speed in the AI era than it did even five years ago.
Q: Why is workforce adoption an HR function, not a technology function?
The technology team can deploy the tool. Only HR can redesign job descriptions, communicate role changes, and build the organizational confidence that makes adoption stick. Goodwine points to SHRM’s growing body of work on AI implementation as evidence that this conversation is already happening outside the technology arena and argues it needs to move faster.
Jeff Kavanaugh:
As we look at the aerospace and defense or war industry, there's been a huge influx, especially in the US, of investment. What has that meant for the A&D traditional supply chain as well as the US economy overall?
Venice Goodwine:
The aerospace and defense industry makes the capabilities that we in the Department of War use to defend our nation, I think it's important that they understand the impact of AI to that. Because now, I am looking now for AI-enabled capabilities. And so when these defense companies or the defense industrial base now delivers these capabilities, I want them to think AI-enabled in mind, and what does that actually mean to help war fighters make decisions faster, better, and sometimes we say cheaper. And for what the administration has done, the commander in chief, our president, has made a concerted effort to issue an AI strategy that then trickles down to the community, the DIB specifically. But that AI strategy also comes with investment that's been made. I talked today on the stage about the Secretary of War and the pace-setting projects, where they have been very clear and very open about the expectation of, here is the areas that we want to focus on for AI, your war fighting, your intelligence, your enterprise. Because when I put those capabilities in the hands of the war fighters, we partner with industry. And I never want to forget the conversation that this is a whole of nation problem. Yes, we may have the requirements and the problem, but we cannot do it without industry. So going back to your question, it starts with the strategy, it starts with executive orders, it starts with funding, and then I think the secretary put a final point on it. And so now that is the goal that I think the defense industrial base need now to start to create these AI-enabled capabilities that'll help us soon compete and win against our nuclear adversaries. And you heard me say today, and in his words not mine, but I totally agree, that this is game changing for us. And we must do it right, and we have to do it faster than our adversary.
Jeff Kavanaugh:
I echo what you're saying, and I think as a former CIO of a major branch of the military, you've got a unique perspective because you were in those rooms where it happened. In fact, you made it happen. And now as you take a step back, you can talk a little more about those things with the tens of thousands of people and millions out there you've influenced. The thing that got me the most when you spoke was the word urgency. Because what do you think about government and military in the past? Not actually the war fighters on the field, but in general, slow moving, methodical. What's coming out now is urgency, speed, an MVP or minimum viable product. How is that, together with AI, changing the face of the military complex?
Venice Goodwine:
Yes. When I served as a CIO and I worked at that time for the secretary at the time, we had an initiative called the Great Power Competition. And in it, he was very clear on, we're already in a competition, and so we need to be prepared for that. That is what heightened the sense of urgency that we needed to take one as-a-service, but we needed to also ensure that our partners had that same sense of urgency. And so we had to be clear-eyed in the conversations we would have. Every time I spoke, or every time any senior leader spoke from Department of the Air Force, we had the same message about this urgency, because this capability is not just in our hands, it's also in the hand of our adversary. And so when you think about that, yes, everything that's intended for good, someone will find other ways to use it. And so we need to outmaneuver, outmatch, and outsmart our adversary with this technology. And I can't say it enough, this is why we need industry to help with this. The frontier model developers, we need their help. The DIB, we need their help. Because when you create these partnerships together, it helps to keep us on the cutting edge and leading in this space. And so I think that's what's important, and that's the sense of urgency that we need.
Jeff Kavanaugh:
And urgency before might've meant a certain clock speed. But with AI, it's moving so fast that it's hard for our minds to get around it. And yet you've got to push companies, push organizations, and manufacturing faster.
Venice Goodwine:
Right. I don't want to miss the idea of, we have to make sure there is a level of AI literacy in this space when we do that. Not just put the tool in the hands of individuals and say, "Go." Let's put the tools in their hands and make sure they understand the capability of the tool, but not be limited in their imagination or the ingenuity on how to use the tool, but I want them to understand it. What actually is AI? Artificial intelligence isn't new. We've been using it for a while. What we see new is now this new generative or the chatbot.
Jeff Kavanaugh:
The democratization of it as well.
Venice Goodwine:
It's being pushed toward individuals now. And so yes, because it is widely available, I want to make sure that the AI literacy piece is there, that the training is there. But also, I want them to understand the guardrails. I want them to understand that yes, AI, these generative models, they hallucinate. This is why domain expertise is important. So if I ask AI a question about cybersecurity, being a former C-suite CISO, I could identify if it was providing me information that was not so true, right? But if you don't have domain expertise, how do you do that? And so having cross-functional teams with that domain expertise helps us keep the trust in the model as well. But also, having the literacy, understanding the guardrails, and then accountability is key, and transparency. We have to be transparent about these models and what they can and what they cannot do. And then also with that transparency is understanding and being clear on what it trained on. How was it trained? And understanding the architecture piece of where the data comes from, how it enters your network, how it leaves your network, and what happens when you use it. All of those things.
Jeff Kavanaugh:
So as a former CIO, you're not telling me how great the technology is.
Venice Goodwine:
Yes.
Jeff Kavanaugh:
Hmm. I heard the word human there somewhere.
Venice Goodwine:
Yes, because you heard me say this before. The psychological safety of our workforce in the age of AI is important. What do I mean? There is angst, there is apprehension, there is nervousness in our workforce because when they hear AI, they hear job replacement, even though those words may not have been spoken. It was the same thing in the robotic process automation era. So now we have to be very clear-eyed on how do we explain to them how we intend to use AI in their particular responsibility, their work, and how they are actually helpers because there is still a human in the loop for AI based on the criticality of the decision that's made. So I want them to know that you still have a role because you get to make sure that the AI didn't drift and that it didn't hallucinate because you have the expertise. And so I think we need to spend more time making our workforce a little more comfortable with the technology. You asked a great question earlier. What do I see in the next 10, 12, 18 months? That when I say the word AI to a workforce, they don't say, "I don't trust it." I want them to say, "Oh yeah, I have my agent buddy," because we now put AI on the org chart. The AI has a job description because it is now a multiplier for my workforce and for my organization. So I need to also hold it responsible as well.
Jeff Kavanaugh:
When you think about putting it in the workforce like that, what does that mean for the very structure and operating model for a company? Any thoughts come to mind on how that's going to shift?
Venice Goodwine:
Have you ever had all the manpower you need to do the work that you needed to do?
Jeff Kavanaugh:
No.
Venice Goodwine:
Absolutely not. But maybe we will know. But AI is a multiplier because it doesn't need a lunch break, it doesn't get tired. It can work longer than I can. I can give it something to do, walk away, or I can give it something to do and go do something else, or I can assign it to a specific role and task, and that's all that it does. But this is why agentic is important. And so now I think our organizations will shift because now on the org chart, you have a combination of human and non-person entities or non-human entities, as we're calling them today. You have now both of those. And I think that's important that when I as a human, look on an org chart and see that I have an NHI on the box next to me, that that NHI is also being held accountable and responsible. And oh, by the way, can get let go as well. I think that gives that level of comfortability that I think we need in the workplace. So there is a shift, but change management is hard. Organizational change is hard, and it needs to be communicated properly. So to all my CHCOs out there, my human capital officers, all of my HR, it is important that they own the responsibility of implementing and the adoption of AI and how the role is going to play in an organization. That's clearly an HR function. That's not a technology function.
Jeff Kavanaugh:
You were eloquent about your comments about the workforce and trust and literacy. Let's rewind a little bit. What about the people who are in high school and college? What are you saying to the next generation of aspiring workers who probably, not probably, have fear and angst.
Venice Goodwine:
Yes.
Jeff Kavanaugh:
And yet there's so much opportunity as well. What do you tell them to get ready?
Venice Goodwine:
I've had the pleasure of being an adjunct professor at a college for about the past 10 years now. And when AI started, right, in the education industry, it was, "Thou shalt not use generative AI for..." And then you'll see the shift, though. And so I now watch this shift in the academic community where they say, well, let's tell them how to properly use AI to help them with the critical thinking. Don't tell AI to write for you. Help it ideate with you, and then ask it to check. When you think of tools, Grammarly and all those tools, they're AI-enabled. And so it's not that I automatically accept the recommendations. Go back to your basics of writing and mechanics of writing before you accept. And so I think what we've seen is that shift. And why that's important is because my elementary, middle school, and high schoolers, if we do not start to inculcate AI into their mindset, into their learning, the next generation is going to be lost. Think two generations next. They're going to be lost because our adversaries, they do this. It's mandated. It is the same if you remember the conversation around having learning second languages? With other nations, they have three and four languages. We struggle to have two. And so I don't want to see that for us in AI. So the next generation that I mentor, when I speak at schools or I talk to them, I tell them, be curious. Get an app, whichever one you like, and play with it. Learn what it can and cannot do. Build agents on your own. Find use cases for yourself. Because the more comfortable you are, the more confidence you have, and then you start to bring that into everything that you do. This is how we're going to spread AI throughout the generations and in different verticals.
Jeff Kavanaugh:
What is the lesson that non-defense industry companies and sectors can learn from this new sense of urgency to adopt AI with these guardrails?
Venice Goodwine:
When America's families send their sons and daughters to join the military to defend and protect our freedoms, when something goes awry and something happens, you feel that differently. Right? Because these are sons and daughters, and you know this for yourself. Less than 1% of the eligible population serve in the military. So when something happens, we feel it. And so we spend a lot of time on things like airworthiness, making sure that aircraft should fly and their safety, and we have safety boards and all those things around our weapon systems. I think it's important to have the same on the commercial side. The same type of standard. And we do, I mean we have the NTSB when something happens commercially. But from a societal perspective, I think we need to have that same sense of urgency and confidence in safety and security of technology. So when I say AI-enabled for A-747, I mean the same as AI-enabled if it was a C-5 or a C-130. Right? Different missions. Probably the same airframe. And so my point there is making sure that when you think about the system, regardless of how it's intended or how it's used, let's apply the same standards. I think that is how we get the commercial and the defense to come to the same playing field.
Jeff Kavanaugh:
And maybe on that uplifting note, that's a good place to end. Thank you so much.
Venice Goodwine:
Thank you. Really appreciate it.
Jeff Kavanaugh:
I appreciate the conversation.
