In the AI Era, Curiosity Beats Credentials

At Scaled we’ve been watching the same pattern repeat across leadership teams: lots of reading, lots of ‘AI strategy’ talk, and a creeping sense that the pace has changed — but not always a clear sense of what to do about it. The answer, we think, starts with people. Not roles. Not org charts. People.

The pace has changed – and the old signals are getting noisier

For years, business planning enjoyed a kind of comfort blanket: steady improvement, predictable cycles, incremental change. We could debate tools, wait for the market to settle, and still make good decisions.

That era is fading.

Now we’re in compounding change. Capability leaps arrive faster than organisational learning. The gap between “what’s possible” and “what we believe is possible” widens quickly – and that gap shows up everywhere: in productivity, customer expectations, delivery quality, and the speed at which competitors start doing things that used to feel unrealistic.

The practical consequence is simple: learning velocity becomes a competitive advantage.

Curiosity is the real differentiator

When you strip away the hype, the people who stay valuable in fast-moving eras tend to share one trait: they’re genuinely interested in the world around them.

They notice shifts early. They follow the thread. They ask better questions. They test things without needing permission. They don’t treat change as an interruption — they treat it as the job.

You can often spot them by where they spend their time. Not because any platform is “the truth”, but because curiosity has a gravity to it. People who are experimenting tend to show up in the messy places where new ideas collide: X, Reddit, niche communities, practitioner threads, half-built demos, early product releases.

Curiosity is not a vibe. It’s a behaviour. And in this market, behaviour wins.

The fastest way to learn AI is to touch it

Most AI content is abstract. Models, roadmaps, predictions, “disruption” narratives. Some of it’s useful, but it doesn’t build capability.

The people who are getting better, faster are doing something more direct: they’re installing tools, prodding them, breaking workflows, rebuilding them, and learning what’s possible through friction.

This matters because most teams still hold an outdated mental model of AI as “a smarter chatbot”. What’s emerging now is more agentic behaviour – systems that can reduce the mental load of day-to-day work: meeting preparation, summaries, drafting, reminders, admin coordination.

A good example is experimenting with something like Claude. Not because any one product is “the answer”, but because touching capability resets expectations. It stops AI being an idea and turns it into a set of practical, testable behaviours.

And once you’ve done that, the conversation changes. Less debate. More iteration.

This is a language shift, not a software rollout

A lot of AI adoption fails because leaders approach it like a standard rollout:

Pick a platform. Run training. Write a policy. Expect consistency.

But this moment behaves more like learning a language (or learning music). You don’t become fluent by reading about it. You become fluent by practising badly, repeatedly, until it clicks.

That’s why the common instincts are now liabilities:

• “We need the perfect use case first.”
• “We’ll wait until the tools settle.”
• “I’m not technical, so this isn’t for me.”

They all sound reasonable. They all slow you down.

The organisations that win won’t be the ones with the most AI theatre in their board packs. They’ll be the ones that normalise experimentation, shorten learning loops, and build internal confidence through doing.

The hiring implication: look for builders, not spectators

“AI literacy” as a CV line is already becoming close to meaningless. It’s too easy to claim. Too easy to parrot.

What you actually want is evidence of learning velocity. People who can describe:

What they tried. What broke. What surprised them. What they changed next.

In fast-moving eras, the most important capability isn’t expertise. It’s adaptability with energy.

A simple next step for this quarter

If you do one thing off the back of this, keep it practical:

Pick one workflow and rebuild it by doing, not debating.

Choose something ordinary and high-frequency – meeting prep, weekly summaries, internal updates, diary planning, follow-ups. Run a small experiment. Share what worked. Share what didn’t. Tighten the loop. Repeat.

You don’t need a perfect stack. You need a learning system.

Because the winners won’t be the companies with the fanciest tooling.

They’ll be the companies who learned how to learn, faster than everyone else.