AI-powered learning and the human edge: Lessons from Julie Stone, CLO at TTEC
Table of contents
Skilled Podcast

AI-powered learning and the human edge: Lessons from Julie Stone, CLO at TTEC

By
GROWTHSPACE
Madeline Miles
May 18, 2026
-
- READ

TTEC is a global customer experience company serving clients in financial services, healthcare, retail, and government.

With 50,000 employees across 22 countries, it's a complex learning environment by any measure. But what makes it unusual is this: the training TTEC provides isn't just for internal development. It's part of what clients are paying for.

That external accountability changes everything. Speed to proficiency, consistency, compliance, skill quality are all client deliverables, not internal HR metrics.

"You can't hide behind internal goodwill," Stone says. "The learning has to work."

Most L&D leaders spend their careers proving ROI to internal stakeholders. Stone has to prove it to the market. That pressure has sharpened her thinking on what AI-powered learning actually requires to succeed.

The people problem

Stone's core belief is simple: the speed of AI will always be constrained by the people.

"Even with a great change management plan, people can only change so fast," she says. "There was almost this sense that you don't need change management anymore because the AI will handle it. But even with a great plan, you are still constrained by people's ability to actually implement it."

The skill gap nobody is naming

Here's what Stone finds most worrying about where L&D is heading: organizations are racing to build AI fluency while quietly neglecting the human skills that make AI actually useful.

AI now produces outputs that look exactly right. Perfect resumes. Detailed change management plans. Well-structured analysis. But go two questions in and the substance often isn't there. AI gives you what it thinks you asked for. It doesn't always know whether the answer is real.

"You have to dig for the substance," Stone says. "I've seen it in my own hiring. The resume is perfect. But when you start asking how they made decisions, it's not there."

The skill she's most concerned about is discernment: the ability to challenge what AI produces, probe for what's real, and know when to trust the output. It's also the skill most organizations have spent decades de-emphasizing in favor of speed and efficiency.

"How come we start life as question marks and end as periods?" she asks. "Children ask why constantly. Adults have learned to stop."

Building that curiosity back into organizations is, in her view, one of the most important things L&D can do right now.

The human in the loop

At TTEC, the answer to all of this is what Stone calls the flipped classroom model, applied across the entire enterprise.

AI handles content delivery, initial practice, and feedback. The human handles what AI can't: coaching in the moment, reading the room, bringing cultural context, and knowing whether a learner actually understood something or just clicked through it.

"There is no facilitator guide anymore," she says. "You have to take data about how your learners are doing and coach them on the spot. That's a very different skill."

TTEC's RealSkill platform puts this into practice for frontline associates. They practice customer interactions with AI-powered simulated customers, inside replicas of actual client systems. Every session is scored and followed by coaching. It's consistent, safe, and personalized — what Stone calls embarrassment-free practice.

AI makes it scalable. The human makes it stick.

What to take away

For L&D leaders navigating their own AI transformation, Stone's experience points to a few clear priorities:

Keep the human in the loop. AI enables scale. Humans enable substance. The temptation to sacrifice one for the other is the most common mistake in AI-powered learning.

Build discernment, not just AI fluency. The most important skill in an AI-augmented workplace is the ability to challenge what AI gives you. Most organizations aren't developing it.

Treat your innovation team like a software team. Pilot, stabilize, then release. Constant change without a stabilization phase creates cognitive overload, not capability.

Tie learning to business outcomes. At TTEC, learning quality is a client deliverable. That standard is one every L&D team should apply.

Everyone can buy the tools. The organizations that win will be the ones who figured out what the human is for.

Resources

For a deeper dive into the conversation, listen to the full episode:

Ready to turn insights into impact?

Discover how Growthspace can help your team apply what matters with expert-led development tied to real business outcomes.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Book a demo
We saw measurable skill growth in weeks, not months.
L&D Manager at PayPal