Every business conversation that touches on marketing now seems to end up revolving around AI: how to use it, integrate it, generate content at scale, reach more people, and automate more of the labor intensive block-and-tackle communications work that humans have traditionally done.

The obsession is understandable… these tools are genuinely powerful and can create tremendous advantage for those organizations who leverage them well. This post is not an argument against using AI, but something very important is getting lost in the current tunnel vision. Much of the AI conversation has collapsed into a false choice: either you're moving at the fastest possible pace to adopt these tools, or you're falling woefully behind. What that framing misses is a third and more strategic position — one that doesn't reject AI's potential but insists on something more fundamental before your AI-driven machinery starts running.

// A FORCE MULTIPLIER NEEDS DIRECTION

AI is a force multiplier… it accelerates whatever you point it at. It doesn’t create direction. Which means the most important question in the room isn't “how.” It's “what.”

What is your organization actually trying to say? What does it stand for? What makes it worth choosing over competitors offering similar products or services? These are the questions that determine whether communication creates meaning or just adds unhelpful volume and noise.

AI is extraordinary at the “how.” How do you produce more content? How do you adapt messaging across formats and audiences? How do you analyze engagement and adjust distribution in real time? These are genuinely valuable capabilities. But they are entirely downstream of the “what.” And here is the limitation that tends to get glossed over in the breathless coverage of everything AI can do: the technology cannot reliably determine whether the message it is producing is actually worth producing. It does not reliably assess whether a brand feels genuinely authentic to its workforce and customers. It cannot intuit how messaging will land against a canvas in which real events are constantly changing human perception of what resonates and what misses the mark. AI operates on pattern, not purpose. It can replicate a voice and extend a narrative — but only if that voice and narrative have already been defined.

The “what” in your business and in your brand is irreducibly a human responsibility.

AI is a force multiplier… it accelerates whatever you point it at. It doesn’t create direction.

// THE GOVERNORS ARE GONE

In past generations, the consequences of leaving your “what” underdeveloped were contained by the nature of the communication environment itself. Content was created manually, reviewed by multiple people, and distributed through a limited number of channels. The system itself had natural governors. If messaging drifted or lacked clarity, the effects were generally contained and the propagation modest.

AI removes those governors entirely. Now the system no longer contains ambiguity, it amplifies it. Organizations that adopt AI without first defining what they are really meaning to say will find that the technology reproduces their strategic vagueness at remarkable speed and scale. Unclear thinking spreads, compounds, and teaches the market a version of the organization that may be incomplete, inconsistent, or outright wrong.

There is a compounding risk here that doesn't get enough attention. The rate at which AI capability is improving is not matched by a parallel improvement in reliability. The more organizations let AI do — because of what it is now capable of — the more consequential the mistakes become when it gets something wrong. At scale, these errors propagate quickly. And unlike a poorly conceived campaign that can be quietly retired, AI-generated content that has been distributed across myriad channels on blast can become institutionally baked into your brand narrative. Not having humans in control of what the machine replicates is no more strategic a workflow than “ready, fire, aim.” The work of defining the “what” cannot be bypassed, glossed over, or delegated to the technology itself.

// WHAT THE MACHINE CAN’T DECIDE

AI cannot determine meaning. It cannot choose direction. It cannot confirm what is worth saying versus what merely sounds like something worth saying. Defining those things requires the ability to understand an organization well enough to determine what it actually stands for, where it genuinely creates value, and what gap exists between how leadership understands the brand and how the market actually experiences it.

Those questions don't resolve through prompts. They resolve through a process that is deliberate, iterative, and guided by people who can read human signals — because the targets keep moving. Client expectations shift. Markets evolve. What differentiated an organization two years ago may be table stakes today... or irrelevant. AI works best when the destination is already clearly fixed. It doesn't navigate moving targets well, and it doesn't approximate human judgment well enough to be trusted with the work of setting direction.

The question, then, is not whether to use the amplifier. It's whether you've defined the signal.

The AI era does not reduce the need for brand clarity. It raises the stakes dramatically, as well as the level of precision required. A general sense of organizational values and legacy style guide are no longer sufficient. What AI requires to function well on an organization's behalf is a structured, detailed, and diligently maintained brand framework. It’s not decoration layered on top of operations, but infrastructure that actively shapes what people, systems, and machines communicate to the outside world.

This is the part of the AI conversation that most organizations are not having yet. They are focused, reasonably, on capability and implementation. The inherently human work — defining with real precision what the organization stands for and what it is trying to say — is being deferred in the prioritization of AI adoption. In some cases it has never been done with sufficient rigor to begin with.

That gap is already starting to become expensive.

The question, then, is not whether to use the amplifier. It's whether you've defined the signal.

// SEQUENCE BEFORE SCALE

None of this is an argument for moving slowly with AI. It is an argument for proper sequencing and the importance of checkpoints and governance.

Define the “what” first. Do it through the kind of deliberate, human-led process that actually produces clarity. Then build the frameworks that let your people and your systems execute against that clarity consistently. Test, verify, and remediate starting with a small sandbox… then gradually amplify, with quality controls.

The organizations that get this sequence right will find their AI more effective, their communication more coherent, and their existing clients more confident they are working with a firm that understands its own capabilities and is in control of itself. The ones that skip the first step will discover, at some point and at considerable cost, that a powerful amplifier pointed at an unclear signal produces one thing: noise at scale.