Fast and Slow at the Same Time: What Enterprise AI Adoption Actually Looks Like

Lessons from Discover's Keith Toney at the 1871 Emerging Tech Summit in Chicago

Most people talk about AI like it's moving too fast to keep up.

But for those trying to implement it inside a Fortune 250 enterprise, it often feels painfully slow.

That paradox was at the heart of Keith Toney's opening keynote at 1871’s Emerging Tech Summit in Chicago. As EVP and President of Credit & Decision Management at Discover, Keith leads a business unit responsible for enterprise-wide adoption of advanced decision science, giving him a front-row seat to both the velocity of AI innovation and the inertia of large, regulated organizations.

It feels like we're going to get speeding tickets and parking tickets at the same time. - Keith Toney, EVP and President of Credit & Decision Management, Discover

The Tools Are Ready. The Organizations Aren't.

The pace of AI innovation is staggering. We're seeing advances in multimodal reasoning, code generation, agentic systems, and orchestration capabilities. These breakthroughs are coming faster than most teams can process, let alone deploy.

But inside the enterprise, most of that potential is still stuck in pilot purgatory.

It has never been easier to experiment with AI tools. It has also never been harder to integrate them into complex, regulated, risk-sensitive systems.

That tension is where most initiatives stall.

Buying Tools Isn't the Hard Part. Changing Behavior Is.

One of the most powerful ideas Keith shared is this: tools create linear progress, but behavior change drives exponential outcomes.

You can buy access to cutting-edge models. You can license platforms. You can run proofs of concept.

But changing how executives think, how risk is managed, how decisions are made, and how outcomes are evaluated? That’s the hard part. And that's what separates experiments from impact.

The biggest unlock is not better tech. It's better leadership.

The Complexity Behind Real Business Impact

Keith walked through Discover's end-to-end customer journey. From media spend and acquisition through underwriting, servicing, and collections, every touchpoint is powered by nested decisioning systems.

At this scale, even a one percent improvement in performance can unlock massive value. But the consequences of getting it wrong are just as significant.

In financial services, explainability, consistency, and fairness are not features. They are requirements.

You don't get to move fast and break things when you are deciding who qualifies for a loan.

The Hidden Barrier: The Change Ceiling

Keith named something we all see but don't always talk about clearly: the change ceiling.

It is the invisible layer where most enterprise AI initiatives hit resistance and lose momentum. It shows up in governance delays, integration failures, internal politics, and the inability to scale beyond the pilot phase.

Everyone is running pilots. Few are transforming their operating model.

That is because adopting a new tool is very different from absorbing it into how the business works. The real work starts when you try to make it real across teams, processes, and systems that were not designed with this in mind.

My Takeaways: The Mindset Required to Actually Impact the Business

Being in the room for Keith's keynote reinforced a few big truths for me:

• This isn't about experimentation, it's about integration. The companies that win won’t be the ones with the most AI pilots. They'll be the ones who figure out how to plug AI into real processes for significant business impact and make it stick.

• You can't shortcut systems-level change. Pilots are often scoped in isolation. But real impact requires understanding the full value chain, what has to change upstream, downstream, and within the current operating model for your AI solution to work, scale, and sustain.

• It takes a new kind of leadership. AI won't transform your business unless leaders are ready to change how decisions get made, how value gets defined, and how risk gets managed. The work is as much human as it is technical.

• The speed paradox is real, and strategic. AI is moving fast. Large companies move slow. But if you know how to move fast in the right places, and methodically everywhere else, you can create real, lasting impact.

This is the work we’re focused on every day at HauerX Holdings across our portfolio, helping leaders move from AI curiosity to capability, from isolated pilots to integrated platforms, and from promising tools to real business transformation.

Final Thought: You Don't Just Adopt AI, You Absorb It

If you're serious about bringing AI into your business, you need to think beyond the surface.

It's not just about what the model can do. It's about what your organization can absorb, technically, culturally, and structurally.

AI adoption is going really fast. And it’s going really slow. - Keith Toney, EVP and President of Credit & Decision Management, Discover

That's not a contradiction. That's the leadership imperative.

And the opportunity is there for the ones who get it right.

About the Author

Jason Hauer is CEO of HauerX Holdings and an Inc. 500 honoree. He partners with commercial AI tech and solutions companies to turn ambition into market leadership.