How Companies Successfully Invest in AI: Augmentation to Differentiation

One year ago, in my blog, How Do We Reimagine an Entire Business with Artificial Intelligence?, I argued that AI shouldn’t just be bolted onto existing systems—it should be treated as a foundational design element. It’s not just about adding AI to existing workflows or products, it’s about building entirely new business models, experiences, and sources of value with AI at the core. That perspective still holds and today, it’s more urgent than ever.

If you’ve started to view AI through the ARC framework—Augmentation (or Acceleration), Replacement, and Creation—you understand that it forces a choice. You can use AI to cut costs, automate routine tasks, and optimize what already exists. And while you likely need to, because your competitors will, this approach is ultimately a cost-cutting exercise that improves the bottom line but offers little opportunity for differentiation. Or you can take the transformative path: reimagine your business, create entirely new forms of value, and set yourself apart in ways that were never possible before.

This choice is now staring every enterprise in the face. As Sebastian Heinz, CEO of Statworx, recently put it in a LinkedIn post: “AI isn’t failing. Companies are. They chased hype… without laying the foundation: No data management. No operating model. No enablement. No strategy.”

This may be the beginning of what Gartner defines in its famous Hype Cycle as the Trough of Disillusionment—a phase where overinflated expectations give way to reality. Some might even call it the onset of an AI Winter. But that’s not a bad thing and that’s not necessarily a setback. It’s a turning point. It draws a clear line between those chasing buzzwords and those rolling up their sleeves to deliver real outcomes and value.

ARC was always a compass

When my colleague Nabil Bukhari wrote about the ARC theory—a well-established framework used to describe the progression of transformative technologies—he also outlined how the concept applies to AI and business strategy. Augmentation is the first phase where AI helps improve existing processes, boost productivity, and streamline tasks. Then comes replacement, when AI begins to outperform or eliminate functions that were once human led. Finally, there's creation, where AI unlocks entirely new offerings, experiences, and business models. Some companies may even choose to leap directly into the creation phase, bypassing the earlier stages altogether. Remaining in the early stages for too long can leave you trailing behind competitors who are quicker to innovate and differentiate.

The mistake too many companies made

There was a rush to experiment with AI after ChatGPT launched. Boards demanded pilots. Every vendor promised AI-enhanced tools. Prompt engineers were suddenly in demand. Demos were flashy. And yes, some even launched their own "CompanyGPTs."

But many skipped the foundational work. They rushed to showcase AI without investing in the essentials—like a robust data infrastructure, a scalable operating model, organizational enablement, or a clear strategy tied to defined outcomes and value. Instead, they placed bets on surface-level wins, hoping momentum alone would carry them forward.

And now the questions are starting to bubble up. Why is our AI pilot stuck? Why is adoption so slow? Where is the return on investment (ROI) in AI?

Automation is table stakes

Let me be clear: using AI to improve customer service, automate support tickets, or streamline workflows is essential—but it won’t differentiate you. Nearly every organization is already doing it or will be soon. These are baseline capabilities, necessary for survival, but not enough to set you apart.

The real leaders are pushing beyond automation. They’re asking a more transformative question: How can AI help us deliver something fundamentally new—something we couldn’t offer before? That might mean reinventing a product, reimagining how customers engage with your brand, or unlocking new business models altogether.

That’s where the outsized returns are. That’s where durable competitive advantages live. And that’s the difference between incremental improvement—and industry leadership.

AI value is real—but only if you build for it

While AI’s impact at the macroeconomic level has yet to spark widescale productivity bursts, a more compelling picture emerges at the enterprise level, especially in financial services. Evident Insights reported “monster earnings” from AI-led banks, with Goldman Sachs beating EPS forecasts by 23.4% and Morgan Stanley by 18.2%, both significantly outperforming their peers. Their data showed the top half of AI-focused banks delivering twice the EPS beats and three times the revenue surprises compared to their industry counterparts.

We see the same trend in other industries as well

Mondelez International—the global snacks company behind Oreo—has used AI not just to optimize operations but to reimagine product innovation. By applying machine learning to design new recipes based on flavor profiles, cost, and nutrition constraints, they’ve accelerated time to market and launched over 70 new products, including the Gluten-Free Golden Oreo. The result? A four- to five-fold boost in R&D speed and a 5.4% quarter-over-quarter sales lift in a soft consumer market.

These aren’t incremental improvements or AI layered onto legacy systems. What these companies are doing signals a full entry into the creation phase of the ARC framework—where AI fuels not just efficiency but strategic reinvention. Whether it’s accelerating product development, redefining customer experiences, or outpacing peers in earnings performance, these organizations are using AI to build what others haven’t even imagined yet. This is where sustainable growth—and true competitive differentiation—takes shape.

How we think about AI at Extreme Networks

That’s why we created Extreme Platform ONE™: a modern, cloud-native architecture with AI at its core. It wasn’t retrofitted; it was purpose-built to embed intelligence into every workflow, interaction, and decision. We’re redefining how users experience and manage networks—not through static dashboards, but via adaptive, real-time intelligence that predicts issues, suggests optimizations, and simplifies decision-making.

And it’s not just for customers. We’re designing for partners, vendors, and every stakeholder in the ecosystem. With Extreme Platform ONE, AI becomes the connective tissue that unifies people, platforms, and processes.

That’s the future. And we’re building it now.

How to invest in AI the right way

If you’re leading an AI initiative, here’s my advice. Use ARC as your compass. Identify whether your investment is augmenting, replacing, or creating. You need all three. But you need to aim for differentiation. Don’t just automate—innovate. Use AI to create experiences others can’t. Demand measurable ROI. If your project isn’t delivering results, move on. Time-box it. Test fast. Learn faster. Lay the groundwork with strong data, governance, and enablement. And think holistically: AI isn’t a feature. It’s a foundational capability that should span your business.