The AI Learning Curve

Why the journey to measurable ROI still matters

In conversations with colleagues, friends and stakeholders across industries, one thing is clear: AI is here! It’s exciting, and it’s also daunting. We’re seeing breakthroughs, but we’re also still learning…navigating ambiguity, early wins, and uneven returns.

Matthew Arnheiter said it well: “We’re in the learning curve.” That’s exactly the mindset we need right now.

We’ve Been Here Before

Every wave of innovation comes with disruption. This one is no different.

It’s easy to forget how uneasy the early days of transformation can feel.

Consider the past:

• Water power and textiles
• Steam and railroads
• Electricity and internal combustion
• Aviation and electronics
• Software and the Internet

Now we’re living through the sixth wave — one shaped by AI, clean tech, and robotics. The tools change, but the learning curve remains.

Where It’s Working

Even amidst uncertainty, there are real signs of progress. In our own portfolio, we’re seeing clinical uplift, operational gains, and financial benchmarks moving in the right direction.

Each use case is different. The tools vary. But one thing holds true: the willingness to disrupt the status quo is paying off. It is not just about reducing administrative burden — it is about creating new capability.

And it is happening everywhere. From summer interns to senior engineers, from IT to operations, our teams are not just participating — they’re shaping the curve.

From the Field (All Sectors): What the Research Shows

Many are working to measure the value of AI initiatives, but the results vary widely. The data is still early, and at times even contradictory. This reflects both the promise and the complexity of this moment.

75% of businesses say they are not yet seeing measurable ROI from their AI efforts

53% of organizations report some early benefits from AI, mostly around time savings and automation

50% of companies expect to begin a transformation journey within 12 months, yet few have operationalized AI at scale

Across the board, we see signals: early pilots, scattered success stories, and foundational work still underway. But sustained, measurable outcomes remain elusive for most.

What Our Healthcare Clients Are Experiencing

Across our client base using AI-powered clinical documentation, these outcomes are showing up every day.

• 1.5 days faster reimbursement submission
• 67% improvement in average note writing speed
• 57% reduction in time from session to signature
• 11% increase in average claims per Direct Support Provider (DSP)
• 5.2 hours saved per staff member per week

More than just stats, these gains are improving consistency, reducing errors, increasing job satisfaction, and accelerating new-hire productivity.

From My Own Experience

Early in my career, I worked for a manufacturing company. That’s where I was introduced to the concept of value engineering. It was a full-team effort — engineers, product leads, support staff, sales — all focused on making the product more competitive without compromising on quality.

In many ways, that is what this era of AI feels like. It’s not about the flashiest tech. It’s about durable improvement.

I also think about my own family — generations of ironworkers and tradespeople. Their mindset was always the same: do it better, smarter, more efficiently. They weren’t afraid of tools. They embraced them. That same mindset belongs in our AI conversations today.

This Summer’s Intern Program

This summer, we leveled up our intern program. Our interns have always worked alongside our teams on real-world projects, but this year we made AI a core part of their experience. Instead of just observing, they’re embedded in active workstreams—building, testing, and improving solutions that leverage automation and AI. They’re not just learning the tools, they’re using them with intent and impact.

Every conversation I’ve had lately reinforces the same point: this isn’t a pilot. It’s a shift. And what’s emerging doesn’t feel like extra work or tech for tech’s sake. It feels like uplift.

What Must We Keep Doing?

No matter your role, your industry, or your toolset here are five essentials:

1. Ask hard questions: Every initiative needs a why, a how, and a how much.
2. Prioritize meaningful use cases: ROI starts where value exists: workflows, risk, forecasting, decision support.
3. Build from the inside out: Data, governance, measurement, and teams that work across silos.
4. Walk the floor: Skip the “open door.” Talk to people. Listen. Ask: What’s working? What’s not?
5. Share what you’re learning: Network. Collaborate. Turn insights into practice. There’s never been a better time for shared learning.

One Last Thought

I see this firsthand. Not just in the solutions we’re bringing to market, but in how our teams are working internally. Every department. Every office. Every role. Our associates are energized, using these tools to challenge what got us here…because they know it won’t get us where we need to go.

That’s why now, more than ever, walkaround conversations matter. Not the “open door” kind. Not another group meeting or status update. I’m talking about hallway chats, coffee moments, and genuine exchanges about what people are seeing and doing.

I’ve had the opportunity to connect with many of my colleagues going deep — Mike Brand , Erica G. , Paul Snider , Matthew Arnheiter , Ryan Behan , Lalitha JS , Erik Ankrom and many more, as we actively challenge how to disrupt the status quo for the greater good. These conversations remind me that some of the most important work happens in real-time collaboration. In the pace of execution and busyness, it’s easy to forget that. But shared thought, when focused on progress, is where real change begins.

And if you’re interested in talking or collaborating, feel free to send me a note. There is power in shared experiences and perspectives. Good things are happening, it’s going to quite a ride for the next few years.