The most neglected task in software architecture is not designing systems, but designing the business around them. Michael Stiefel's conversation with Joy Ebertz makes this plain: as AI reshapes how developers work, architects who focus only on technical delivery are building on sand. Long-term strategy is the load-bearing wall, and too many teams are treating it as a nice-to-have. That is a dangerous oversight, especially when AI is compressing the time between a new capability and its mainstream adoption. If you are an architect, your job title is no longer about structure diagrams. It is about foresight.
The shift is practical, not philosophical. AI agents can now generate code faster than anyone can review it, a reality that changes the economics of software development and demands a new kind of trust. As we have argued before, that trust has to be architected deliberately, not assumed: Why AI Needs a Trust Architecture Built for Human Control. When machines produce more of the work, the architect's role turns toward deciding what should be built, why it should exist, and how it will sustain the business over the next decade. That is not a technical decision. It is a strategic one, and it is the one most likely to be deferred because it is hard to measure and harder to defend in a sprint review. The consequence is that teams optimize for the current release while the organization drifts toward irrelevance.
This is also a moment to reconsider how work gets absorbed. The pace of AI-generated output means that code is outpacing our ability to read it, which should change your hiring and your review processes, not just your tooling: When Code Outpaces Reading, It's Time to Hire AI. Stiefel's point is that architects must think about the long-term health of the business, and that includes deciding when to let AI take over entire functions, not just assist with snippets. That requires a clear-eyed view of which parts of your system are strategic differentiators and which are commodity work. If you cannot answer that question, AI will answer it for you, and the answer will not favor your team's long-term relevance.
The practical takeaway is direct: schedule time for strategy the same way you schedule time for code review. Put a recurring, protected block on your calendar to map how AI changes your product's value proposition, your team's structure, and your technical debt. Ask what happens if the cost of building drops by half in two years, and whether your business model survives that. The architects who thrive will be the ones who treat business continuity as an architectural concern, not an executive distraction. Watch how your organization handles that question in the next quarter, because that will tell you more about your future than any roadmap you publish. The technology is moving fast; your strategy should move faster.