What AI Needs From Us

Why storytelling is the real skill behind AI

Here’s something worth knowing: when researchers at Stanford asked students to remember information from a one-minute pitch, 63% could recall a story. Only 5% could remember a statistic.

In another study, psychologists Gordon Bower and Michal Clark found that students who wove a list of words into a narrative remembered 93% of them. Those who just tried to memorize the list? 13%.

We’ve always known this intuitively. Stories stick. Facts fade. But we rarely stop to ask why — or what that means for the most powerful technology we’ve ever built.

Stories Are How We’ve Always Worked

Long before writing, before books, before the internet, human beings passed everything that mattered through stories. How to survive. Where to find food. Who to trust. What to fear. Stories weren’t entertainment — they were survival technology. They were the original operating system for the human brain.

Think of stories as memory bubbles. They package up complex information — context, emotion, meaning, consequence — into something the brain can carry. A fact is loose change rattling around in your pocket. A story is a photograph in your wallet. One falls out. The other stays.

This is why every culture on earth, no matter how isolated, developed storytelling independently. It’s why the oldest known narrative cave painting — found in Indonesia — dates back more than 51,000 years. Stories aren’t a nice-to-have. They’re the currency human beings have always used to exchange information, build shared understanding, and make sense of a complicated world.

We don’t think in data. We think in stories. Always have.

Now Along Comes AI

So here’s what’s interesting. We’ve built this extraordinary technology — artificial intelligence — and the entire conversation around it has been about what it can do. How fast it processes. How much it knows. How many parameters it has. Benchmarks. Capabilities. Performance scores. We’re still living in the age of ego — measuring worth by what we know. And we’re measuring AI the same way.

But almost nobody is talking about what AI actually needs to be useful.

And what it needs is the thing humans have been doing since we first sat around a fire: telling stories.

AI doesn’t work well when you give it instructions. It works well when you give it context. A narrative. A point of view. A framework to think inside of.

When you tell AI “write me an email,” you get something generic. When you tell it “I’m a manager who needs to deliver difficult feedback to someone I respect and want to keep on my team, and I tend to avoid conflict” — now it has a story. And the output changes dramatically.

This isn’t a minor distinction. It’s the whole game. And the research backs it up. Researchers at the University of Bonn recently developed what they call a “Story of Thought” approach — using narrative structures to guide AI reasoning. When they tested it, narrative-based prompting consistently outperformed traditional methods across multiple AI models. In one case, it improved accuracy by more than 14 percentage points over standard prompting. The same AI, given the same task, produces wildly different results depending on the narrative you wrap around it.

The richer the story you give it, the better it performs. The more it understands about who you are, what you’re trying to accomplish, and why it matters to you, the more useful it becomes. AI draws conclusions from narrative themes. It reasons through context. It gets better when you give it the same thing humans have always needed to function: a shared understanding of what’s going on and why it matters.

The Thing Nobody’s Teaching

And this is where the entire AI industry is getting it wrong.

Every company is focused on teaching people how to use AI. Here’s how to write a prompt. Here’s how to upload a document. Here’s which button to click.

But AI isn’t really something you use. It’s something you talk to. Whether you’re typing or speaking out loud, the interaction is fundamentally a conversation. And conversations aren’t built on commands. They’re built on narratives.

The companies building AI should be teaching people how to talk to it — not how to use it. “Using” something implies pushing buttons and following steps. “Talking” to something implies sharing context, building understanding. One is mechanical. The other is human.

This Is What We’re Good At

Here’s the irony of the whole AI fear conversation. Everyone’s worried about what AI will replace. But the single most important skill for getting value out of AI is the most human skill there is: the ability to tell a story.

To provide context. To articulate what you’re feeling, what you’re dealing with, what you need.

The people who will thrive with AI aren’t the most technical. They’re the ones who can best articulate their narrative — who they are, what they’re trying to do, and why it matters. That’s not a prompt engineering skill. That’s a human skill. It’s the same skill that’s made us effective communicators for thousands of years.

Remember the stories we tell ourselves — those personal narratives that shape how we see the world? They’re not just psychological patterns. They’re the raw material AI needs to help us. The more honestly you can articulate your story, the better AI can work with it.

AI doesn’t diminish the value of human narrative. It amplifies it. The better your story, the better AI performs. The clearer your context, the more useful the output. The more you bring of yourself to the conversation, the more you get back.

The Real Instruction Manual

So if the AI companies actually want to reduce fear and increase adoption, they need to rethink the story they’re telling. As I wrote in “The AI Resistance Is Growing,” these companies have been talking about what AI does — pitching capabilities to investors and letting that language leak into public messaging. But research consistently shows that narrative-driven communication increases technology adoption. McKinsey found that brands communicating value through storytelling can increase adoption rates by as much as 30%. People don’t adopt technology because they understand its specifications. They adopt it when they can see themselves in the story of how it helps.

Which means the answer isn’t better tutorials on how to click through an interface. The answer is helping people understand something they already know how to do.

You already know how to tell a story. You already know how to explain your situation to another person. You already know how to share context, describe what you need, and articulate why it matters.

That’s all AI needs from you.

The technology that everyone thinks requires technical expertise actually requires the most ancient human skill there is. The thing we’ve been doing since cave paintings and campfires. The thing our brains are literally built for.

So don’t think of AI as something that demands technical skills. Think of it as something that demands the one thing you’ve been practicing your entire life. Give it context. Tell it why. Lay out the problem you’re facing, the challenges ahead, and your goals.

Then see what happens.