Every few months someone else does a "Sam Altman" and announces that AI will make agencies obsolete. The case always sounds the same. AI can research faster. AI can ideate faster. AI can write and design faster. Therefore the agency model must collapse.
But a recent study published in Decision Analysis points to a different conclusion. The research suggests something that is at the heart of what I have been saying in my Rhetorical AI articles: yes, AI accelerates research and rhetorical formulation, but it does not improve strategic judgment. It does not decide what matters. It does not resolve contradictions, or clarify objectives, or turn a mass of plausible ideas into a coherent path forward.
AI speeds up the part of the process that was never the bottleneck. The bottleneck has always been judgment, alignment, and sense-making. That is the core of what agencies do. Agencies aren’t paid for typing. They’re paid for discerning the real problem, crafting meaning out of ambiguity, and guiding clients toward choices that hold up under pressure.
AI doesn’t remove that function...in fact, it increases demand for it. Which means the real story is not agency extinction. It is agency reformation.
AI floods the organization with more ideas, more inputs, more narratives, more “insights” than ever before. Without structural change, this surge overwhelms the slowest part of the system — the layers of prioritization, framing, coordination, and decision-making. Traditional hierarchies cannot digest the speed.
What to do?
First, agencies must reshape their structures. Pods, stable teams, and shorter decision loops raise the organization’s absorption velocity. They create clarity about who owns what. They give teams the ability to metabolize AI-accelerated inputs without drowning in them. Structure becomes the enabling technology.
Second, the judgment layer must grow faster. The old apprenticeship model asked people to wait years before being trusted with real strategic work. That model collapses under AI. When machines generate endless options, agencies need more humans who can evaluate, sequence, and choose. The path to becoming a strategist must shorten. People must reach strategic adulthood earlier in their careers.
Third, information and meaning-making must become collective. The lone strategist crafting the sacred deck was already fading. AI finishes that shift. Teams must develop shared cognition, shared context, and shared interpretation. Meaning-making becomes a group sport. The strategist still leads, but they lead a conversation, not a procession.
Fourth, clients must be brought inside the judgment process itself. Clients will bring AI outputs too. They will arrive with drafts, frameworks, and pseudo-insights of their own. The agency’s role shifts from delivering answers to helping clients separate signal from noise. Strategy becomes co-created, with the agency acting as the stabilizing partner in a world of rapidly generated possibility.
None of this is destruction. It is modernization. AI doesn’t erase the agency’s purpose. It calls for a new way of working that can harness the speed, not be crushed by it.
Agencies that cling to old structures will feel overwhelmed. Agencies that reform — structurally, culturally, and developmentally — will not only survive. They will thrive.
In an era where anyone can generate ideas at scale, the real value lies in judgment, discernment, alignment, and the ability to turn possibility into coherent action.
And that is very good news for agencies.
