Turbo Brains or Slop?
A reflection on AI as a multiplier, not a substitute, and why selection has become the real creative work.
AI has turned creative work into a volume problem. The tools can generate endless options, drafts, and “good enough” versions of the thing. Which is exactly why the hard part hasn’t disappeared, it has just moved. In 2026, the scarce resource isn’t output, it’s judgment, the kind you only get by living inside a discipline long enough to know what to keep, what to kill, and what’s quietly average.
At Relish, questions about how to use AI come up constantly. Clients ask us about it. Prospects ask us about it. Internally, we’re watching new tools and features land every week, and trying to answer a surprisingly unglamorous question: what belongs in the workflow, and what doesn’t?
Underneath a lot of these conversations is the same uneasy assumption: that creative work can now be done cheaper and faster because AI can generate output.
We’re not “the AI agency.” We’re not trying to turn everything into a prompt. We’re doing what most teams are doing right now, poking at the tools, stress-testing them, and keeping what earns its keep.
When AI is useful for us, it’s upstream. It helps us get to more starting points faster, explore alternatives we might not have explored, and compress research into something we can react to. Then the human work begins: selection, edits, restraint, and the final calls that make the work feel owned.
The question is placement: where it helps, and where it starts quietly lowering the bar.
A framework we keep coming back to is a simple 2×2. One axis is AI usage (not using it vs using it). The other is judgment (weak vs strong). The only tweak we make is that judgment isn’t a personality trait. It’s local. It lives inside a discipline. You earn it through reps.
When you combine high AI usage with weak judgment, you get Slop Cannons, high volume, low signal. When you combine high AI usage with strong judgment, you get Turbo Brains, speed without losing taste. Low AI usage with strong judgment is Steady Hands. Low AI usage with weak judgment is Dead Weight.
A marketer can have sharp judgment in paid acquisition and still be a beginner in brand systems. A designer can be deadly in typography and composition and still have no feel for what makes someone click and convert. The point is that judgment is local. AI can amplify judgment where it already exists, it doesn’t transplant it into a different lane.
That’s what the 2×2 clarifies. AI isn’t a blanket upgrade. It’s amplification. In the right lane, it multiplies skill and speed. Out of lane, it multiplies noise. And because output is cheap now, selection becomes the work.
This is where we keep seeing confusion. A lot of brands, especially under budget pressure, try to use AI to shortcut disciplines, generating everything from brand story to campaign creative, packaging, and ads. From a distance, it looks like momentum, because a lot of files get created. But competitive markets don’t reward “something exists.” In CPG, the bar is high, and consumers have a ruthless slop detector. They can feel when something is generic, unowned, and stitched together.
What holds up is work with a point of view. Decisions that feel earned, not generated. Cohesion you only get when someone with depth in the lane is steering.
We don’t know exactly what creative workflows will look like two years from now, but we’re confident about the constraint: AI can speed you up, it can’t give you depth. If anything, it’s making that depth easier to spot.
