If you want to understand where your organization actually stands, and what it would take to make AI meaningful instead of mythical, explore Ondaro’s AI Visible Value™. It’s designed to show you where you are today and where AI can take you next. 
This article was originally published on LinkedIn by Eric Smith.
Introduction
I’ll be honest. My frustration with the state of AI in 2025 is the same frustration I hear from every enterprise leader I meet. The conversation has been hijacked by buzzwords, glossy marketing, and inflated promises. Meanwhile, the real organizational problems the industry claims to solve remain stubbornly untouched.
Too often, AI is presented as some kind of unicorn pixie dust sprinkled on old capabilities and repackaged as innovation.
If we’re going to make meaningful progress in 2026, we have to get past the hype and focus on verticalized, outcome-driven AI that actually moves the needle.
1. How AI Branding and Buzzwords Overshadow the Real Work
Let me be blunt: the industry has leaned far too heavily on marketing theatrics. I can’t count how many “new AI features” I’ve seen this year that aren’t new at all, just rebranded as AI.
And I’m over it.
The hype pulls people away from the questions that actually matter:
- What problem are you solving?
- What has already failed?
- How will we measure success?
Until organizations come back to fundamentals, AI will keep stalling out in proofs of concept instead of delivering real operational outcomes.
2. Meaningful vs. Meaningless AI: What Actually Matters
There’s a difference between AI that looks impressive in a demo and AI that actually makes someone’s job easier. I draw a pretty clear line between the two.
Meaningless AI
- Renaming existing automation or workflow capabilities without adding real value
- Chat-style “agents” that don’t understand an organization’s terminology, context, or user needs
- Generic AI features that aren’t connected to the systems or data that drive decisions
Meaningful AI
- Streamlining genuinely messy, resource-heavy work, for example, IT asset management being one of the strongest use cases I see today
- Simplifying the experience for overburdened employees and asset owners
- Integrating with systems in ways that deliver value instead of novelty
The question isn’t “Does it use AI?”
The real question is: “Does it remove friction and help someone do their job better or faster?”
3. Why Industry-Specific AI Is the Next Evolution
The future of AI isn’t generic, it’s verticalized.
I don’t want an agent that tries to be everything to everyone. I want an agent that’s built for the public sector. Or for telco. Or for higher ed. Domain matters.
Generic AI models don’t:
- Speak the way real agents in those industries speak
- Understand domain-specific terminology
- Reflect regulatory requirements, policies, or workflows
When your “agent” doesn’t understand the world it’s operating in, you can’t trust its decisions.
Verticalized AI flips that. It’s grounded in real industry outcomes, which means the logic behind every action (the “if this, then that”) is meaningful, accurate, and aligned to how work actually gets done.
4. How Workflow Data Fabric and Domain-Specific Models Change the Game
I was part of the teams that helped define Workflow Data Fabric, so I’ve seen the evolution up close. At first, it felt like more hype — a new wrapper on things ServiceNow had already been doing.
But WDF has matured into something real.
Here’s why it matters:
- It connects to external data sources in a way that’s truly AI-native
- It gives Agentic AI access to data without storing it — zero copy, without compromising security
- It provides the contextual foundation AI agents need to operate intelligently
- And it’s getting stronger because ServiceNow is acquiring companies with deep specialization
Let’s be clear: WDF isn’t magic. We’re still using APIs, IntegrationHub, web services — you know, the usual suspects. But the way the data is presented, governed, and utilized is finally at a level where AI can actually do something meaningful with it.
5. Why Organizations Must Start with “Why” and Measurable Outcomes
If AI is going to make an actual impact, we have to anchor every initiative in outcomes, not features.
Too many projects start with statements like:
“We just want to implement ITSM.”
“Just give it to us out-of-box.”
Instead, we need to start with:
- What does success look like?
- What outcome are we actually trying to achieve?
- What have we tried before, and why didn’t it work?
- Which KPIs prove we’ve made real progress?
When we elevate the conversation from what to why, we build solutions that:
- Solve real problems
- Drive adoption
- Produce measurable value
- Avoid the technical debt that comes from quick fixes or shallow AI deployments
Outcome-driven AI is the only thing that turns hype into impact.
The Future of AI
My prediction for 2026 is simple: the industry is going to move away from shiny branding and gravitate toward AI that is narrowly focused, vertically intelligent, and grounded in real outcomes. This will be the moment AI stops being hype and starts being transformative.
If you want to understand where your organization actually stands, and what it would take to make AI meaningful instead of mythical, explore Ondaro’s AI Visible Value™. It’s designed to show you where you are today and where AI can take you next. 