They just don't know it yet. This is the story of what happens when thirty years of encoded business knowledge meets a world where the translation layer is gone — and the person who knows why the rule has that one exception can finally build the system that runs it.
Not in sequence. Not on a roadmap. Simultaneously. Every IBM i organization is navigating all three whether they've named them or not.
The architecture of what's possible just changed. Agentic AI, MCP protocol, 500+ IBM i tools in 2026. The four-layer stack — Sovereign Core to Modern Surface — is buildable today. The platform isn't the barrier.
Every workflow in every organization is human-centric today. A human is the checkpoint across all process flows. That is about to change. The org designed for this world doesn't look like the org you're running.
The role of the practitioner is transforming — not disappearing. The bottleneck becomes the system author. The person who knew the business becomes the person who built the system that runs it.
In the pre-AI world, there was a translation layer between knowing and building. You needed a developer to convert your business knowledge into executable code. That layer created the bottleneck. That layer is now gone. The person who understands the problem can now build the system that solves it. The domain expert and the system author are becoming the same person.
This convergence has been building for years. The voices calling it are not from the IBM i community — they're from the center of the AI economy. And every single one of them is describing the person sitting in this room.
AI capability is advancing exponentially. Organizational readiness is moving linearly. The gap between them is the market.
Anthropic released Claude Mythos — an AI that found vulnerabilities in Linux, Windows, and OpenBSD that survived decades of human review and millions of automated tests. The architecture question just changed.
Found thousands of zero-day vulnerabilities across every major OS — Linux, Windows, OpenBSD — including a 27-year-old flaw in one of the world's most security-hardened systems that survived decades of human review and millions of automated tests.
Can analyze compiled binary code without source code. Legacy systems with lost source are now fully exposed — if they run standard architectures. The "nobody knows this old code" defense is gone.
Not built as a hacking tool. The same reasoning power that makes it an exceptional coder also makes it exceptional at finding and chaining vulnerabilities.
IBM i compiles to MI — Technology Independent Machine Interface — not standard binary. Not x86. Not ELF. Not the format Mythos is trained to analyze. The attack surface is structurally different.
Object-based, capability-based architecture. Authority checking and audit trail are built into the hardware and OS — not layered on top. Many standard vulnerability-chaining paths don't exist the same way they do in Linux or Windows.
The governance model the industry is now scrambling to retrofit — IBM i organizations have operated this way for 35 years. Mythos didn't threaten IBM i's architectural advantage. It proved it.
Six root causes. All organizational. None of them are about the platform.
No one knows exactly when the inflection point locks in. But multiple signals are converging from too many directions to ignore. The organizations that cross the chasm first compound their advantage permanently. The ones that wait face a gap that doesn't close.
The organizations scrambling to build agentic AI capabilities are starting from scratch — building the very foundation that IBM i organizations have had for decades. Transaction performance. Data integrity. Operational reliability. The foundation every modernization effort has to build on top of. You're already there.
The question was never whether IBM i could survive the AI era. The question is whether IBM i organizations will recognize their advantage before someone else tells them it's a liability.
"The future is here. IBM i is ready."
"Agents need tight integration: low-latency data access, transactional consistency, and secure execution. That's not a feature you add. That's the IBM i architecture."
"The hardest part of modernization is not the technology. It is the knowledge locked inside your existing systems and the small number of people who hold it."
The destination isn't an org without humans. It's an org where humans govern what matters most — and agents handle everything else.
Three tracks. Running simultaneously. The same logic as the transformation itself — you can't move them in sequence. The practitioner evolves as the technology modernizes as the organization learns to lead differently. Each one unlocks the others.
Imagine the person who spent 30 years watching good ideas die in the backlog. Defending a platform nobody understands. Carrying knowledge nobody captured. Squirming in their seat — full of ideas nobody asked for.
"Within a couple of years, AI will be integrated into the operations and applications of IBM i for all users."