Most organizations are sprinkling agentic AI on top of a legacy, heterogeneous stack — and calling it a strategy. That works until the low-hanging fruit runs out, because it rests on a flawed assumption: that the business will keep operating exactly as it always has, only faster.

In this episode, host Andreas Welsch sits down with Grant Ecker, Chief Architect at Ecolab and founder of the Chief Architect Network — a 682-member nonprofit collaborative of Fortune 500 architecture executives. Drawing on architecture leadership roles at Danaher, Walgreens Boots Alliance, Medtronic, Lowe's, and General Mills, Grant lays out what it actually takes to make an enterprise architecture — and an architecture organization — ready for agentic AI.

Key insights from this conversation include:

• Start with capability-based planning, not use cases. Decompose your strategy into the capabilities you have to be great at, map the gaps, and build a heat map of transformation opportunities. Some will be exciting (product innovation); some won't (order management). Both determine whether the strategy lands.
• Earn the seat at the table through IT first. Architects get invited into strategy conversations because IT leadership trusts them — not in spite of it. Go to the business first and you risk undermining the partners who own that relationship, and losing the invitation entirely.
• Move from T-shaped to pie-shaped. Technical depth alone no longer sustains an architecture career. Add a second point of depth in business understanding and communication, and become the connector and strategist for your part of the ecosystem.
• Rethink "if it ain't broke, don't fix it." If something is fixed and sitting in the middle of a transformation, that's precisely the moment to challenge it — when a business outcome justifies the disruption. The prize is refactoring decisions so agents can make them with the right guardrails, governance, trust, and permissions.
• Treat AI FinOps as a budgeting discipline, not an afterthought. Meter consumption by department, forecast token spend before the sticker shock arrives, and tie the next wave of investment to demonstrated ROI. Consumption is not a proxy for value.
• Centralize, then federate. Pilot and learn centrally with safe off-the-shelf tools, then embed capabilities into the organization as maturity grows — the same curve cloud followed, and the one quantum will follow next.
• Peer communities beat published best practice. By the time an approach is written up, it's six to nine months old. A trusted relationship with a peer in a non-competing industry is the fastest way to stay ahead of a market moving this quickly.

Whether you're a chief architect building the agentic reference architecture, a CIO deciding who holds the keys to AI adoption, or a technology leader working to be in the room where strategy gets set, this conversation offers a grounded view of the organizational work that makes agentic AI possible.

Tune in now to learn how to prepare your architecture — and your architects — for a shift Grant frames as an age-like change on par with the printing press and the internet.

Questions or suggestions? Send me a Text Message.

Support the show

***********
Disclaimer: Views are the participants’ own and do not represent those of any participant’s past, present, or future employers. Participation in this event is independent of any potential business relationship (past, present, or future) between the participants or between their employers.

Level up your AI Leadership game with the AI Leadership Handbook (https://www.aileadershiphandbook.com) and shape the next generation of AI-ready teams with The HUMAN Agentic AI Edge (https://www.humanagenticaiedge.com).

More details:
https://www.intelligence-briefing.com
All episodes:
https://www.intelligence-briefing.com/podcast
Get a weekly thought-provoking post in your inbox:
https://www.intelligence-briefing.com/newsletter