Your IBM i Data Is the AI Advantage Nobody’s Talking About
Rick Hurckes, director of services at mrc, explains the critical role of data quality in successful AI deployments, and why IBM i is so well positioned for this new era
Every IT leader is fielding the same question right now: “What’s our AI strategy?”
But if you’re running IBM i, it almost feels like every answer starts with a sales pitch for migration. Move your data to the cloud. Modernize first, AI later. When it comes to AI tools for IBM i, much of what’s out there starts with getting off the platform.
Recently, the conversation has started to shift. More voices in the IBM i community are talking about deploying AI in place (without migration). But, for most IT directors, the loudest message is still the same: Your platform is old. You need to modernize or buy something new before AI is possible.
I disagree. You’re already sitting on the thing that makes AI actually work: your data.
What Every Company Needs Before AI Can Work
Do you want to know the dirty secret of enterprise AI? Most enterprise AI projects fail because of data quality, not model quality.
One recent survey found that 44% of organizations identified data quality as the single biggest roadblock to AI project success.
The numbers get worse from there. According to the PEX Report 2025/26, data quality and availability are the single greatest barriers to AI adoption across industries. Informatica’s 2026 CDO research finds the same thing: 57% of organizations admit that data reliability is a top barrier to their AI initiatives.
Think about that for a second. Companies with modern cloud stacks, data platforms and dedicated data engineering teams are still struggling to get their data into a state where AI can use it. They’re spending months building data pipelines. They’re cleaning up data scattered across dozens of tools, spreadsheets and disconnected databases.
They must do all of this before they can build useful AI tools. This is where the conversation about AI tools for IBM i needs to start: It’s about the data.
What IBM i Shops Already Have
Now, look at what’s sitting in your environment.
If you’re running IBM i, your business data lives in DB2. It’s structured. Your applications have maintained and validated it for years (even decades for many of you). That’s the foundation that every AI initiative requires (and many companies don’t have).
IBM i environments provide the five foundational capabilities that AI deployment requires:
| What AI Requires | What IBM i Shops Already Have |
| Clean, structured data | DB2 with decades of validated transactional data |
| Single source of truth | One integrated database, not data stitched from dozens of sources |
| Data governance framework | Built-in security model, access controls and audit trails |
| Consistent data quality | Application-enforced data integrity rules running for years |
| Queryable data layer | SQL Services providing direct, standardized access to everything |
I can’t tell you how many times I’ve heard about companies on modern platforms spend six to 12 months just getting their data ready for AI. Meanwhile, IBM i shops have had that foundation in place the entire time.
The irony is very real. The platform everyone calls “legacy” has the data infrastructure that every AI project needs.
The Conversation Is Shifting
The good news is that the conversation is changing. IBM is investing in “zero copy” data access, letting you query data where it resides without moving or duplicating it. API-first modernization, where you expose IBM i logic through modern APIs rather than rewriting it, is becoming more common.
Now, don’t get me wrong. There’s nothing wrong with modernization. Sometimes it’s the right move. But modernization and AI deployment are two different conversations, yet … they’re getting conflated. You probably don’t need to modernize your IBM i environment before you can deploy AI. For most shops, the data foundation is already there.
The risk is assuming you need a 12-month readiness project before AI is possible. You probably don’t.
What Can You Build Over IBM i Data?
So, what can you build with AI tools for IBM i today, without moving your data anywhere?
AI assistants connected to your live business data. Think of a support rep who can instantly pull a customer’s order history, check pricing and draft a response … all from your DB2 data. Or a sales assistant that knows your inventory levels and can answer questions about product availability in real time. Now, I’m not talking about generic chatbots. These are task-oriented helpers that understand your data and your business rules.
AI-powered workflows that automate real processes. Suppose you have a ticketing process. A support ticket comes in. AI classifies it, detects sentiment, routes it to the right team and triggers the follow-up steps. It does all of this connected to your existing business systems, reading and writing to your DB2 data.
Chatbots embedded in your applications. Again, when I say “chatbot,” I’m not talking about a generic chat widget. I’m referring to a conversational interface built into your customer portal, internal dashboards, support pages, etc. … that can actually do things like look up order statuses, update tickets, check account balances and more.
AI that turns your documents into searchable knowledge. One of the most common uses of AI, I’ve seen many businesses turn their policy manuals, compliance docs, product specs and support articles into AI-searchable knowledge bases. Employees ask questions and get answers sourced from your documentation. This is one of the quickest AI “wins” you can implement.
The IBM i ecosystem is expanding to support this. IBM’s MCP (Model Context Protocol) servers for IBM i let AI agents interact directly with your system. SQL Services already provide a standardized interface to your data. The tools exist. But the most important piece is the clean, governed data that makes all of this work. Chances are, you’ve had it for years.
What This Looks Like in Practice
Here’s a good example.
An IBM i shop runs its entire operation on DB2. Their sales team needs answers fast. Right now, getting a custom report means submitting a request to IT, waiting in the backlog and getting results days later.
With an AI tool connected to their data, a sales rep types: “Show me our top 20 customers by revenue this quarter who haven’t placed an order in the last 30 days.” The AI queries live DB2 data and returns the answer in seconds.
The same approach works for operations, finance, customer service and any department that needs answers from your business data. The key is that the AI connects directly to DB2 on IBM i. Your security model stays intact and your governance framework doesn’t change.
The Bottom Line
The companies deploying AI successfully right now aren’t the ones with the biggest budgets or the newest platforms. They’re the ones with the best data. If you’re running IBM i, that’s probably you.
If you want to dig deeper into this, I’ll be at COMMON POWERUp 2026 in April, presenting on deploying AI in your IBM i environment. I’d love to talk through what this looks like for your shop. Come find me.