AI isn't salt ... more like ketchup.

Lately, it feels like AI is being treated like salt: sprinkle a little on anything and you’ve made it better.

But in reality, it’s probably more like ketchup.

Put it on french fries? Fantastic. Slather it on a farm-fresh peach? You’ve just ruined the peach and the ketchup.

Last week at the AMA Central PA conference, I hosted a panel on AI in marketing. The conversation reinforced something I’ve said before: while AI is a powerful tool, it’s not a cure-all. And using it just because it’s “available” can be a recipe for disappointment.

Here are a few key takeaways from the discussion:

It’s all about the data. Garbage in, garbage out. If your inputs are wrong, your outputs will be too—only faster and at scale.

Start with the problem. Don’t hunt for shiny AI use cases just to use AI. Look for the biggest pain points, often the mundane, repetitive work no one wants to do.

Ask not just “Can we?” but “Should we?” AI can mimic a therapist, but it’s not trained to keep you safe or healthy. Capability isn’t the same as responsibility.

Your work will change. Just as farmers can harvest more with modern equipment, AI will reduce the number of people needed for certain tasks. This isn’t necessarily bad—it means the nature of our work will shift.

Human experience matters. AI will never feel the warmth of the sun on its skin, never watch a sunset and feel something. That perspective is uniquely yours—and it’s irreplaceable.

The bottom line? AI can be an incredible accelerator, but it’s not seasoning you throw on everything. Use it thoughtfully, and it will enhance your work. Use it indiscriminately, and you might ruin the peach