Leadership in the AI Era: Beyond Hype to Human-Centered Strategy - Theresa Fesinstine

As AI transforms the workplace, senior leaders are grappling with fundamental questions: Where should AI fit into decision-making? How can it enhance rather than replace human connection? And who should be driving these changes within organizations?

Theresa Fesinstine brings a unique perspective to these challenges. With 25 years of HR experience spanning L&D, generalist roles, and C-suite positions at companies like News Corp, she made an unexpected pivot in 2022 when she discovered ChatGPT. What started as asking AI to write a haiku about company culture evolved into a complete career transformation. Today, she’s an adjunct professor, author, and consultant helping HR leaders navigate AI adoption while maintaining their people-first focus.She recently wrote a book, People Powered By AI: A Playbook for HR Leaders Ready to Shape the New World of Work..

In this conversation, we explore the practical applications of AI for leadership teams, the role of HR in driving organizational AI strategy, and why the future of work requires more human skills, not fewer.

From HR Generalist to AI Pioneer

Suzan: Hi, Theresa. It’s so good to have you because we met through Troop HR, which is a community I joined like a year and a half ago, and I’ve met really cool people like you. You do a whole bunch of stuff, and the reason I wanted to talk to you today was about AI and particularly around leadership, since I work in the leadership space with senior leaders. But before we get into what I know will be an exciting conversation, can you introduce yourself?

Theresa: First of all, thank you for having me. I’m so excited. Anytime I get a chance to talk to people that just make my soul purple and bring me joy is the best day ever. Troop is the GOAT as far as community is concerned, especially for the HR world.

I spent 25 years starting off in L&D for about seven years and then switching over to HR generalist. I worked for a division of News Corp for a very long time, then did some of my own consulting - I think now we would call that fractional. I worked with companies like Omnicom Media Group and Trinity Wall Street, then went back in-house for five years through the heavy COVID years leading to 2022 when I ended up really through a variety of burnout and trauma that HR experienced during that time, deciding to start my own business.

It was a culture consulting business that started off okay, but I didn’t really know a lot about starting my own business. Then in December of 22, I had a friend very serendipitously send me a text message and said, “Hey, have you tried this chat thing?” I asked ChatGPT to write a haiku about company culture, and I was laughing thinking it was so funny. Then I started asking it to help me build out the foundational structure for my business around sales processes and business plans. I was flabbergasted. It felt like I had truly an advisor.

2023 brought me through what I call my foundational education year - getting certified through MIT, going to Luxembourg for a month for academic summer school. 2024 is when other people started noticing. I became an adjunct professor, wrote a book, and here I am today spending all my working hours thinking about, talking about, and supporting HR leaders in their endeavor to lead the journey around AI for their businesses.

Suzan: I love that. I didn’t realize you have such an interesting journey from an HR perspective - the L&D and then the generalist and now it feels like you’re going back. What was the impetus to move from L&D to more of a generalist?

Theresa: I was working for News Corporation, traveling constantly - I had anywhere from 43 to 45 expense reports a year. I wanted upward mobility in the company, wanted to expand my income and scope. The best way to do that in what I considered a really fantastic company was to move into a generalist role. I wanted to be a manager.

AI Avatars and the Future of Leadership Meetings

Suzan: One of the things - you saw one of the polls I wrote about this CEO who created an AI avatar to attend meetings in his place. I was really curious what people thought about that. The responses were overwhelmingly “no, I wouldn’t trust it.” A few people said “fine by me in some settings,” but when I asked for which contexts, no one told me. What did you think about that question?

Theresa: I think I probably responded “in certain situations.” One of the things I’ve seen over the past three years - I started my AI journey December 5th, 2022.

Suzan: I love that you know that date.

Theresa: I literally know the date because it was five days after ChatGPT launched. I could envision scenarios where having my CEO’s avatar there would have provided space for others to share their voice in a way that maybe other times that CEO would overwhelm the conversation or prevent others from speaking.

Having worked in-house and been in the C-suite for quite a few years, there’s also a lot of conversation around the ROI of an executive team - your highest paid individuals sitting in a room for an hour, and half the time there’s a lot of bullshitting that goes on. I’ve been sitting on executive teams since I was 25 years old, and there’s a fair amount of that happening.

I think there could be more leveraging for the meat of the meeting. Maybe the CEO is trying not to be a part of every decision, trying to empower their C-suite to be decision makers, but be in the room to understand what conversation is happening. I can see space where it would make sense, though I think humans always need to be in the conversation.

The fear that brings about by people - because a lot of the hesitation around AI is really rooted in fear of the unknown and change - people might think, “Well, if my CEO doesn’t even need to be there, why do I need to be there?”

Suzan: I knew you would have a nuanced response. I really was curious when people said “sometimes” but no one except you has given me the “sometimes” response.

What role do you see as most important for senior leaders to be using AI? We can talk about HR leaders in particular too.

Practical AI Applications for Leadership

Theresa: My belief is that most organizations do a shitty job of preparing and guiding people in the C-suite or executive roles or even managerial roles. I remember early 2023, a guy in the back of the room at one of my first AI workshops asked, “Shouldn’t managers be trained by the company?” I said, “You show me 10 companies that do a great job of it and I’ll retract my statement about leveraging AI.” You can’t do it.

Even in companies where I led training and felt good about what we were doing - multi-level management training, mastermind programs, every person who became a manager went through a series - we still didn’t do enough. It’s hard to be there at all times.

Leaders are trying to juggle the day-to-day of their functional departments, people challenges, and their role as a leader within the organizational ecosystem. Where AI can enter that puzzle is helping with in-the-moment guidance on situations. Whether it’s a custom GPT in ChatGPT, a project in Claude, Perplexity, Gemini - just getting some “Hey, I’m in this messed up situation. What’s the best way to respond?” Sometimes eliminating the percolation on what to do and having an advisor to say “here are some options” can truncate the time it takes to get through some of that garbage.

Some very obvious things - if you don’t have transcribers working, I still talk to people sitting in meetings taking the same notes I took in 2001. There should not be a human tasked with doing those things anymore when you can get much more accurate transcription, clear next steps, assignments delegated from every meeting you have and use automation to summarize that at week’s end for planning the following week. All of a sudden the time you would have taken to do all those things manually, you get much greater clarity and outline of how to actually accomplish things and move things forward.

The other thing people aren’t leveraging enough is building custom projects around your OKRs or KPIs, whatever functional use you have, and building complex prompts around actually getting an action plan done. If you want to do a real meaningful offsite with your direct team, run a hackathon using AI and your annual objectives. Do it today, starting for 2026. You will be so much farther ahead of the curve.

I’m not here to say AI for everything. I’m very much about human interaction and centering people-first, AI-forward organizations. That being said, we need to stop feeling like we’re not putting in effort if we’re using tools to help us do things more efficiently with greater insight than we could ever have as singular humans.

Suzan: Yeah, I see a lot of binary thinking around it. All or nothing, right? Don’t use it at all or it’s everything. I love what you’re saying about there’s a middle way. Where do you use it and where don’t you? I think people want people connections. They will always need those things. In many ways, we will need more skills around people skills in the AI era. I love that you’re saying there’s a place to use it strategically and for what benefit.

Theresa: A great way to figure that out is to literally use any tool of choice and ask it to take a process - not a task, because people are using AI for very task-oriented things as opposed to process. Take a process or an objective, ask it to give you the 10 steps needed to successfully complete it, then ask it to give guidance on where your insight, your human capabilities need to be interjected in the process and create a process map with a back and forth of where does AI fit in and where does the human fit in.

You can direct it and say, “make this 80% human, 20% technology,” or “make this 50% human.” Or ask: “If you had your druthers as my advisor, what is the most optimal percentage of human to technology where the human is getting greater one-to-one interaction with other humans, and technology is taking on the things that it does its best at?”

HR is the only functional area within an organization that understands the impact that change has on every other functional area.

AI in Strategic Decision-Making

Suzan: I think that’s really well said. When we talk about senior leadership - organizational initiatives, business-wide initiatives, strategy - what role might AI play in decision-making, or should it?

Theresa: Well, what do you think? Do you think it should?

Suzan: I don’t know. Maybe? I love these questions. I’m really interested in the edges of all of this. I think it could provide a lot of context. Here’s a whole bunch of things we’re considering - give us the pros and cons, play it out for me. I think that could be interesting.

Theresa: Totally. What about if you took your last three months of executive leadership minutes, plugged them into a generative AI tool, and asked for gaps. What are we not seeing? What are we missing? Take that, redact what needs to be redacted, but put your financials in there. Take your business objectives and goals. Literally compile all this and say, “I need you to do a deep dive on what I’ve just shared. What gaps are we missing?”

Another great thing - I’m doing a Predictive Index workshop on Friday. If you’re using DISC or StrengthsFinders or whatever tools you leverage, I would literally take all of those, take all the minutes, all the objectives, your profiles, plug them in and say, “Where are we misaligned as leaders around what we’re trying to accomplish for the business? Does it make more sense for your financial person to be part of something completely outside their realm because they have a unique behavior or skill set?”

Beyond the fact that my CFO has a resume that speaks CFO, what are the behaviors and competencies and skills that this individual has unique to our business? Where do they align with our objectives and perhaps starting to look at roles a little bit differently?

Suzan: I love that. I do think there’s a role for AI in decision making. It’s about the interaction between people and technology. Where is that interaction? I love the idea of thinking about skills. I was thinking about information - what is the topic we keep talking about that we’re not making a decision on? If you see it and have AI saying “you keep talking about this,” it brings something forward. It can also help around how our decision-making process maybe isn’t the most efficient or effective. Where are we getting stuck?

Theresa: Here’s what you’re going to need to do to build that. Any company listening to this that wants to do that needs to build a prompt that isn’t a cheerleader, but is a challenger. I have a custom GPT I created called Tough Love. Part of the reason I created this was as a solopreneur, I’m pretty tough on myself. I always have been. I’ve always had bosses that said, “I could never push you harder than you push yourself.”

Suzan: Me too! I hate that! I’m like, stop agreeing with me! Stop telling me I’m amazing!

Theresa: It drove me crazy. So I created this custom GPT called Tough Love. It’s available - you can look it up under my name, Theresa Fesinstine in the marketplace. You got to be prepared for it. It really is hard truth - you need to get off your dead ass and move on this idea.

From a leadership standpoint, it needs to be able to tell the tough story. Do you realize that by spending two months at 20 minutes a week talking about a uniform policy because you’re not comfortable telling one guy to not wear a hat at work is costing the company this much money? There needs to be some tough love with the executive team.

Maybe that’s the place - speaking truth to power when nobody else in the room can. No matter how powerful executives feel, when they’re in that conference room together with other leaders, there’s a lot of ego and kowtowing, especially to CEOs. Maybe there’s a way AI starts to speak more truth to power and say, “Look, we’re wasting time on this.” Through data analytics that AI can provide, you start to guide your leadership team around meaningful forward action.

Building Custom AI Solutions for Teams

Suzan: I’ve been wondering about creating a custom GPT for leadership teams because of this whole “What are you missing when I can’t be in the room all the time?” I’m a facilitator working with leadership teams, but if I can’t be there, how can we continue to help them? I’ve literally been Googling how to create a custom GPT because I want to create one for them.

Theresa: I did a whole tutorial probably eight months ago. I was thinking about reloading it anyway. Do you have a paid ChatGPT account?

Suzan: No, I use Claude..

Theresa: In order to create custom GPTs in ChatGPT, you have to have a paid account. But what you could also do is say, “Hey, Theresa, here’s what I want this custom GPT to do.” I can create it and share it, or make it public.

One of the things organizations tend to do is think enterprise-wide - “Well, everybody’s not going to use this, so it might not be that relevant.” When it comes to AI tools - and I’m not just talking about generative tools, but very specific tools that assist HR teams, finance teams, marketing teams - you really need somebody doing vetting or a tool that’s going to do that vetting so you know which tools make the most sense.

If I were back in-house today leading a team of 15 or 20 people in HR, I would have at least two to three people with plus accounts. If my budget didn’t allow for a team’s account for just HR, I would pay for two or three people to have plus, and leverage those people to build out our custom GPTs and share that with the rest of the HR team. That gives people access to the custom GPT without having to have the functionality to create.

Here’s my current mission - no longer should HR be asking permission for stuff. We need to be providing direction on what we need and seeking partnership to help make it happen. If I have a CTO that says, “We can’t afford to get ChatGPT for everybody,” I would say I’m not asking you to get ChatGPT for everybody. What I’m going to get is ChatGPT for three members of my team - one in L&D, one in recruitment and myself. That’s going to cost us $60-75 a month. Everybody else is going to use the free version. I need you to help me understand what you’re concerned about with security so I can answer those questions and we can move forward. It’s not about “can I have it?” It’s “this is what I’m going to do, and I need your support to make it happen.”

There are things you need to understand about large language models and how they work. Don’t just do a free-for-all with your organization and let them learn on their own and teach one another, because the last thing you want is for them to teach one another bad habits.

Why HR Leaders Must Lead the AI Transformation

Suzan: I would love to hear your thoughts about what role people leaders, CHROs, CPOs, heads of people should be playing in AI these days.

Theresa: The thing that comes to mind every time I’m asked this question is from March 2020 until the pandemic was “over”, HR leaders all of a sudden had a different level of authority. We were expected to know things we didn’t know, in the same way that IT today is expected to know AI and they often don’t. We rose to that occasion with fervor. We had to be much more agile and communicative about things that were new to us. That gave us this headiness of being the one giving the best way to move forward. I don’t want us to lose that momentum.

There are a few reasons why I truly think HR leaders should be leading the conversations, building the project plans, the communication cadence, the learning and development skill building. HR is the only functional area within an organization that understands the impact that change has on every other functional area. We understand that if AI or technology changes the function of a role, we’re going to have to hire differently for that role, and communicate to the organization differently. If processes change, we have to be part of how that process fits into other processes.

You can only really lead adoption if you’re familiar. It’s very difficult to say we need to do X if you’ve never done it yourself. This comes with an opportunity to gain clarity on how you understand your organizational dynamics and culture to integrate AI. You can only really do that if you are experienced yourself.

The other reason I think people leaders need to be the team leading this is when things start to shift - and we’re seeing it all over the place. Some companies are going 100% in. Every CEO is asking themselves, “How many functions do I need to retain? How much could I replace with AI?” If HR isn’t sitting in that room talking to them, we’re going to end up in the same situation I found myself in during COVID - being the voice of reason saying, “Why? Who’s going to see us as hypocritical? We have to be the voice of reason.”

We want to eliminate a full department. How is that going to affect the rest of the organization? What organization are we trying to build? How does this reflect our values? But more importantly, how are we upskilling, reskilling or new skilling the employees we have so that we’re ready for that trip down the runway?

What I spend a lot of my time thinking about is how do I help HR leaders get over that barrier of entry? That “okay, yes, I am the person to do this.” Doesn’t matter if your ego is saying you don’t know technology, your legal team is saying it’s not our job, it’s risky, let somebody else do it. We have to tamp those down, take a deep breath, sit in our opportunity and power and say, “No, I need to be a part of this. I need to lead the communication on this. In order to do that, I need to be a leader in the transition process.”

Practical Advice for AI Adoption

Suzan: Is there anything we haven’t talked about or something you’ve been thinking about around the role of AI and leadership that you think we should talk about? Anything else that comes to mind or that leaders need to know?

Theresa: The most important space to think about right now is just recognizing that everybody’s at a different place on their learning curve, but it doesn’t mean that everybody’s not learning. That person who says, “I use AI every day” - ask them what they’re using it for, how they’re using it. I’m guessing you’ll get fairly similar answers.

I use this analogy of a beach. In 2022, early 23, when I started learning about generative AI, a lot of people were in the seagrass, maybe not even aware of what it was. By 23, more people were coming down to the beach, dipping toes in the water. 24, we’ve got a lot more people on the beach, but even in 25, we still have people in the seagrass.

For your leadership teams, number one, communicate. If you don’t have guidelines - I don’t use the word policy anymore, I’ve given that up from my years of writing policies, but I believe in guidelines and roadmaps - communicate how you’re thinking about AI. If you’ve not put a guideline out there to your employees, know that they’re using it, they’re exploring it, they’re doing it on the slide. They may think they’re cheating. They may be very worried about their own jobs. Communication helps people understand and get rooted in what’s really happening.

Number two, don’t encourage people to go it on their own. I was at a conference a few months ago, and one practitioner gave advice that I would never give. It was very difficult for me not to stand up and course-correct her with 400 people nodding along. There are things you need to understand about large language models and how they work. Don’t just do a free-for-all with your organization and let them learn on their own and teach one another, because the last thing you want is for them to teach one another bad habits.

Invest in the training, invest in the learning. Now is the time you’re building your budgets - build money in and plan to talk to your team about it. Keep transparency as my guidance.