DRAWING THE HARD LINE: AI DOESN'T NEED YOUR TRUST, AND DOESN'T CARE IF YOU'RE LAZY.

Sure, we all know artificial intelligence doesn't really have a soul, yes? We all know it doesn't harbor we humans any ill will, correct? I mean, IT'S JUST A MACHINE.

However.......

AI is getting very good at sounding like it knows what it's talking about, and humans are getting very comfortable believing it.

That is where the danger lies. Not because AI is evil or is going to suddenly wake up and take over the planet.

The more immediate problem is considerably more ordinary: People are becoming dependent on something they don't understand. In some cases, they don't want to understand.

Let's start with a simple question: What does “human oversight of AI” actually mean?

It doesn't mean having a human somewhere in the building, somebody occasionally glances at the screen, and it certainly doesn't mean: “Well, AI produced it, so it must be pretty good.”

Human oversight means the human has enough knowledge and authority to challenge what the system produces. That means understanding the assignment and the limitations. Understanding the sources. Recognizing possible bias.

Checking the facts.Questioning the assumptions. Knowing what happens if the answer is wrong.

That's a much bigger job than clicking ENTER.

Here's where age enters the conversation. Pew's latest research shows that 66% of Americans ages 18 to 29 have used AI chatbots. Among people 65 and older, the number is 23%. Daily use shows an even wider gap.

Hang on. Here's what I don't want anyone to do with that statistic: Don't conclude that young people understand AI and older people don't, because that's too easy.

Using AI is not understanding AI.

You can use something every day and still have no idea what's happening underneath the hood. In fact, that's part of the problem.

YouGov found that trust in AI increases with usage. Among avid users, 82% trust AI at least somewhat. Yet accuracy remains the biggest concern.

Think about that. The more familiar people become with the machine, the more comfortable they can become trusting it. Even while knowing it can be wrong.

That can be chalked up to human nature. If something repeatedly gives you useful answers, you start lowering your guard. That's how dependency develops. Then something even more troubling can happen.

You stop doing the thinking yourself.

Researchers are now examining what's called cognitive offloading—essentially letting technology carry mental work that you once performed yourself.

Microsoft researchers found that higher confidence in AI was associated with less critical thinking among the knowledge workers they studied. Again, that doesn't prove AI makes people stupid but it should make us ask a very uncomfortable question: What happens to a skill when we stop practicing it?

If AI writes every first draft, do we become better writers? If AI summarizes every report, do we become better readers? If AI answers every question, do we become better thinkers?

Maybe, but only if we're still doing the intellectual work of checking, challenging and understanding what came back. Otherwise, we're not using AI as a tool. We're using it as a substitute.

This is where the corporate world has a serious problem.Companies are throwing AI into workplaces at enormous speed. McKinsey found 88% of organizations reported using AI in at least one business function in 2025, but only 7% said AI was fully scaled across the organization.

Therein lies a big problem.

KPMG found that 66% of employees in its research said they relied on AI output without critically evaluating it, and more than half reported mistakes from AI. 48% reported putting company information into public AI tools.

So let's stop pretending that giving everybody access to AI automatically makes an organization smarter, because it doesn't.

Sure, it can make an organization faster, more efficient, more productive. But it can also make an organization wrong faster.

That's why oversight matters. You see, oversight isn't a department. It's a skill. You need people who can look at an AI answer and say: Wait. Stop.

Is this actually true? What's the source? What's missing? What's the context? What assumptions are buried in here? What happens if this is wrong?

The most important question: Who benefits if I believe this?

That's where Control the Narrative comes in. Because AI isn't just writing emails. It's increasingly shaping what people see, what they search for, what they believe and what they repeat.

If you're going to control the narrative in an AI-driven world, you have to understand the machine helping create it. You don't need to become a computer scientist.

You do need to become AI literate.

You need to know enough to question the answer, recognize the limitations, protect your judgment. Because eventually the danger isn't that AI replaces your thinking.

The danger is that you stop thinking because AI is doing it for you.

That's not progress. That's dependency.

Once you become dependent on something you don't understand, you don't control it.

It controls you.

REMEMBER: Control the narrative, or someone else will control it for you.