AI-Powered Coaching: The Future and the Present

Dr. Anna Tavis — Clinical Professor and Chair, Human Capital Management Program, NYU Steinhardt. Anna has written the book on the future of digital coaching and brings 15 years of executive experience in technology and financial services, including serving as a global head of talent development. Her research focuses on the intersection of coaching, technology, and adult learning.

Matt Dreyer — Head of Talent Management, Prudential Financial. Matt leads the deployment of AI coaching across Prudential's 40,000-person organization. He has been one of the most vocal enterprise voices on the value of getting AI coaching tools into employees' hands now rather than waiting for a more settled technology landscape.

Parker Mitchell — Co-Founder and CEO, Valence. Moderator.

The science of how adults learn has a clear answer: not in classrooms, and not once a year at a performance review. Adults learn by doing — in the moment, in the flow of work, when a real situation is actually unfolding. In this webinar, NYU's Dr. Anna Tavis and Prudential's Matt Dreyer explore why AI coaching is the first technology that can deliver on that insight at scale: bringing personalized, on-demand coaching to thousands of employees, helping them practice before high-stakes moments, and building familiarity with AI itself in the process. Their shared message on timing: this technology is as bad as it will ever be right now. The gap between early movers and the rest will only widen.

Key Takeaways

• Adults learn by doing — and AI coaching is the first tool that can deliver that at scale. The science of adult learning is unambiguous: people learn experientially, in the flow of work, not in classrooms. AI coaching is the first mechanism that can deliver in-the-moment, personalized support to thousands of employees simultaneously — at exactly the moment they need it, not at the end of a performance cycle.
• The power of AI coaching is employee-led demand, not HR push. When employees can bring what is most important to them in a given moment — rather than being assigned a topic — their engagement and commitment to the coaching is qualitatively higher. Matt Dreyer at Prudential is betting on that pull dynamic to drive adoption far more effectively than any mandated program could.
• This technology is as bad as it will ever be right now. The NYU Coaching and Technology Conference produced a line that Matt cites directly: the AI coaching tools available today represent the floor, not the ceiling. They will continue to improve exponentially. Organizations that wait for the technology to mature are misreading the moment — the cost of delay is a compounding gap between early movers and everyone else.
• AI coaching also builds organizational AI fluency from the inside out. Prudential sees its AI coaching pilot as serving two purposes simultaneously: delivering development value to employees and getting AI directly into the hands of users in a tangible, positive way. That direct experience helps demystify the technology and reduce fear of AI across the broader workforce.
• The future of AI coaching is a connected development ecosystem, not a single tool. Matt Dreyer describes a near-term roadmap where AI coaching connects to the broader talent marketplace — nudging employees toward relevant learning programs, gigs, projects, and mentors based on their development goals. The coaching conversation becomes the entry point to a much richer network of development support.

Questions This Session Answers

Why is learning in the flow of work more effective than traditional training programs?

Adult learning science is clear: people learn experientially, by doing, not by sitting in a classroom. Traditional L&D programs — even well-designed ones — suffer from a transfer problem: learning happens in one context and then has to be applied in a different one, often weeks later. AI coaching solves this by delivering support in the actual moment a challenge is occurring — when a leader is working through a real project, navigating a real relationship, or preparing for a real conversation. That is when insight converts to lasting behavior change.

Why is Prudential deploying AI coaching now rather than waiting for the technology to mature?

Matt Dreyer is direct about the reasoning: the technology is as bad as it will ever be right now. AI coaching tools will continue to improve exponentially, which means the practical question is not whether to adopt but when to start building the organizational learning and muscle memory that comes from real use. Prudential evaluated the options and concluded the technology is more than good enough to provide genuine coaching support — and that waiting is itself a form of falling behind.

How does AI coaching help employees become more comfortable with AI generally?

One of Prudential's secondary goals for its AI coaching pilot is organizational AI fluency. Many AI tools benefit employees indirectly — through process improvements or backend efficiency gains they never see. AI coaching is different because it puts the technology directly in the hands of the user in a personally valuable context. Employees who have a positive experience using AI coaching are more likely to approach other AI tools with curiosity rather than fear — making the coaching pilot a change management lever as much as a development one.

What does the future of AI coaching look like as the technology develops?

Matt Dreyer describes a near-term evolution where AI coaching becomes the connective tissue of a broader talent ecosystem. Rather than being a standalone conversation tool, the AI coach would nudge employees toward relevant learning programs, surface open gig opportunities or projects matched to their development goals, and even facilitate introductions to mentors with complementary skills. The coaching relationship becomes the context through which the full range of organizational development resources becomes accessible — personalized and proactive rather than self-directed and passive.

Why is it important for HR leaders to start experimenting with AI now rather than waiting for best practice?

Anna Tavis argues that the traditional HR posture — waiting for best practice to emerge before investing — is no longer viable. The gap between organizations that are actively experimenting with AI and those watching from the sidelines is already widening, and it is compounding. The advantage of AI coaching is not just the tool itself but the organizational learning that comes from genuine use: understanding what works in your specific culture, your specific workforce, your specific leadership context. That knowledge cannot be borrowed from another company's case study. It has to be built through real pilots.

Full Webinar Transcript

What is Prudential's approach to rolling out AI coaching — and why now?

Parker: Good morning, everyone. Good afternoon if you're joining from Europe. I am very excited to welcome people to today's webinar.

Anna: Thank you so much, Parker. Excited to be here. Welcome, everyone. I'm the department chair for human capital management at New York University.

Matt: My name is Matt Dreyer. I'm the head of talent management at Prudential. This technology shift over the past few years is enabling us to do things we have struggled with in the past. For the broader group of the population that might not have access to that kind of coach early in their career, what we want to make sure is that we get this in their hands. We're going out to a couple thousand of our leaders of people initially — globally, different countries where we operate, different levels — and providing them access to engage with Nadia and bring what is on their mind at that time. That, to me, is where the power is going to be most prevalent.

We have confidence in the power of coaching largely because a good coach allows the coachee to set the agenda, bring what's most important to them in that moment, and help them identify a solution that is tailored to them because they're creating it themselves through that coaching. To provide that to thousands of people as they need it — the opportunity to do a role play, a practice conversation, get some guidance around how to think about something differently, or try something new in the next couple of weeks — that pull from individuals will drive much greater commitment than an HR push. This is also one of our AI pilots that gets the AI tool right into the hands of the user. There are some pilots where employees benefit indirectly, but this is one where they benefit directly and get to use it themselves. In addition to benefiting from the coaching, they're going to become more comfortable with this technology. It will hopefully allay any fears they have about AI.

How does AI coaching connect to what we know about how adults actually learn?

Parker: Can you share more about how you've seen digital tools — and coaching specifically as knowledge building or skill acquisition — evolving? What could this look like in two, three, four years?

Anna: What is important to understand here is how adults actually learn. The science supports it clearly: adults learn by doing. Adults learn experientially, not in traditional classrooms. The new capability around learning in organizations — connected to change management around leadership models and everything else — is learning in the flow of work.

When people can get feedback not once a year at the end of the performance cycle, but when they are actually working on projects, encountering issues, and can get support right in the moment a particular situation occurs — that is when learning happens. That is when real transformation in how people think takes place.

Matt: Looking ahead, if we think about how quickly this technology can continue to develop — some of the things on our wish list for our talent marketplace go beyond just coaching. If I have a person at Prudential who wants to develop in a particular area where they feel they have a flat side, something like this could help connect the dots in a variety of ways. Depending on the data we give it access to, it could say: "There's a new learning program that might be of interest to you." Or: "There's a job, gig, or project going on that could help you develop that — here's how to connect with it." It could make introductions: someone has just joined the company and identified a skill they are willing to mentor others on — "Would you like me to put 15 minutes on your calendar to connect with that person?" But even before it gets to that level of integration, the ability to provide nudges day-to-day — the opportunity to reflect on how to practice a skill today — is already powerful.

Why is it important to put AI coaching in employees' hands now rather than wait for the technology to settle?

Parker: Why is it important to put something in the hands of your users now rather than wait 12 months when things might have settled down?

Matt: We were not going to do this before we felt the technology was ready to provide a really good experience. When we looked at options a year or two ago, we didn't think it was there yet. But I'm struck by something said a couple of times at the NYU Coaching and Technology conference: this technology is as bad as it's ever going to be right now. It's going to continue to get better, and it's going to get better at an exponential rate. Based on what we've seen, it is more than good enough to provide genuine coaching support to our employees and deliver a really positive experience for those who want to engage with it.

Parker: Anna, you've talked about HR leadership experimenting in an agile way. I imagine that involves a get-started-now component to it. Can you share why you think that's important?

Anna: As a profession, we have always wrestled with innovation. We were more risk-averse in making investments into something new. We always wanted to see best practice first. The injection of AI into the bloodstream of the HR function is changing that whole dynamic. We cannot wait — that would be at an extreme loss to our organization. We need to understand what innovation means to us now more than ever. Because, to Matt's point, it is going to accelerate the gap between those who get on board with AI — even in an incremental way, starting with a small safe pilot — and those who do not. If you are behind, you are behind.

And if you're going to be adopting somebody else's practice, the magic of this technology is its personalization. I can understand what Prudential is doing and use it as inspiration — but I have to think about how my particular culture and my particular organization is going to move ahead with adoption. We cannot be sitting on the sidelines, going to webinars, and doing nothing. Start small. Practice your agility by participating in these conversations — absolutely. But also go back and try it.