The 6 Issues With AI Nobody Is Being Honest About

Every edition of this newsletter exists to say the thing that most marketing leaders are thinking but nobody in the room will say out loud. 

Let’s start with the truth nobody at the AI conference is putting in their keynote:

AI didn’t make marketing harder. It made average marketing free.

Which means if your value as a marketing leader was in the production (the copy, the content, the campaign setup, the reporting) that value just got commoditized. Not in five years. Now.

The marketing leaders who are panicking right now are panicking for the right reason. They just haven’t named it correctly yet. It’s not that AI is replacing them. It’s that AI just replaced the part of their job they were hiding behind.

Here’s what actually changes when you adapt, and what doesn’t.

1. Stop Using AI and Start Training 

Every marketing team in your competitive set has access to the same AI tools you do. ChatGPT, Claude, Gemini, Perplexity: the playing field is flat, and it costs $20 a month.

Which means the tool is no longer the differentiator. What you put into it is.

The marketers who are winning right now aren’t the ones who found a better prompt. They’re the ones who built a documented brand Voice, a five-layer customer profile, and a set of creative principles and then trained their AI on all of it.

The output looks completely different. Not because they have better AI. Because they gave it better inputs.

The question to ask yourself this week:

“What has my AI actually been trained on?” 

• If the answer is “the same generic marketing knowledge as everyone else,” that’s your problem, and it’s fixable.

2. The Vision Has to Exist Before the Prompt

Here’s what AI-generated content actually reveals: whether or not your brand has a documented point of view.

When the Vision is clear (your purpose, your differentiators, the specific language your customer uses to describe their own pain) AI produces content that sounds like you. When the Vision isn’t documented, AI fills the gaps with the most average possible version of your industry’s language.

Your AI content doesn’t sound generic because your AI is bad. It sounds generic because your brand strategy is living in someone’s head instead of in a document.

The fix isn’t a better tool. It’s doing the upstream work that makes every tool perform better. Vision first. Then prompts.

This is exactly why the Disruptive VSET framework starts where it does, because without a documented Vision, everything downstream (strategy, content, team direction, AI training) is built on a guess.

3. Your Customer’s Language Is Your Competitive Moat

Most marketers don’t know it well enough.

AI is very good at producing marketing language. It is very bad at producing your customer’s language: the specific words they use when they’re frustrated, the exact phrase they type into Google at 11pm, the sentence they say to their colleague the day before they make a buying decision.

That language only comes from deep, documented customer intelligence. Interviews. Sales call transcripts. Support tickets. Reviews. The five-layer ICP profile that goes far beyond demographics into fears, desires, and the precise vocabulary of their pain.

When that intelligence exists and is fed into your AI, the output stops sounding like marketing and starts sounding like the thing your customer was already thinking.

When it doesn’t, your AI produces content that could have been written about any brand in your category. Because it was. Your competitors are running the same prompts.

The marketers who will be indispensable in an AI world are the ones who become obsessive students of their customer, not because AI can’t write, but because AI can only write as specifically as the intelligence you give it.

4. Execution Discipline Is Now the Unlock—Not Execution Speed

AI made content production fast. Which means speed is no longer an advantage. Everyone is fast now.

The new unlock is testing discipline, the ability to treat every piece of content as a hypothesis, capture the learning when it succeeds or fails, and make the next piece sharper because of what you know.

Most marketing teams are producing more content than ever and learning less than they ever have. Because volume without a testing framework is noise generation.

The marketers who pull ahead in an AI world will be the ones who slow down the analysis just enough to know what’s actually working, because they’re learning from everything they produce.

One question to pressure-test your current execution: 

“Can you name the three biggest content learnings from last quarter and show how they changed what you’re doing now?”

• If not, you have a testing problem, not a content problem.

5. The Human Things Are Now the Valuable Things

Here’s the uncomfortable flip side of AI commoditizing production: it has raised the value of everything AI can’t do.

AI cannot build genuine trust with a client who is skeptical of agencies. AI cannot look a frustrated CEO in the eye and say, “The math doesn’t support that goal, here’s what would.” AI cannot walk into a room where everyone disagrees and leave with alignment. AI cannot invest in a brand it doesn’t believe in.

These are human things. And in a world where every deliverable can be generated in seconds, the marketers who can do the human things with skill (the strategic conversation, the honest pushback, the genuine investment in a client’s outcome) are the ones whose value becomes harder to replace, not easier.

This is not a soft argument. It’s a structural one. As AI compresses the value of production, it expands the value of judgment. The marketing leaders who lean into their judgment and build the systems that let their teams exercise theirs are the ones who will be indispensable.

Everyone else will be a prompt engineer with a shrinking budget.

6. The Team That Can Train AI Beats the Team That Uses AI

There is a meaningful difference between a marketer who opens ChatGPT and types a prompt and a marketer who understands what makes a prompt produce a specific result, who can look at an output and diagnose exactly why it’s wrong, how to fix it, and what the customer actually needs to hear.

That second marketer is rare. And they are becoming exponentially more valuable.

The teams that will win in an AI world are not the ones with the most tools. They are the ones who have invested in developing marketers who understand marketing deeply enough to direct AI with precision, who bring the strategic judgment, the customer intelligence, and the creative instinct that no prompt library can replace.

At Disruptive, we move on from 96% of marketing candidates because the work we do requires people who can do the thing that matters, which is increasingly less about execution and increasingly more about the quality of thinking behind the execution.

The question for every marketing leader reading this: 

“Is your team being developed to use AI, or to direct it?” 

• Those are not the same job. And one of them has a future that AI can’t touch.

The Bottom Line

AI didn’t change what great marketing requires. It just made it impossible to fake.

The marketers who will be indispensable in this environment are the ones who:

• Have a documented Vision that makes their AI sound like them
• Know their customer’s language well enough to make AI speak it
• Test with enough discipline to actually learn from what they ship
• Lead with judgment, not just production
• Build teams that can direct AI, not just use it

These aren’t new skills. They’re the skills that always separated great marketers from average ones. AI just removed the option of hiding behind busy work and calling it strategy.

The good news: if you’re reading this, you already know something is shifting. You wouldn’t be here otherwise.

That instinct is the thing worth developing.