Lately, I’ve been noticing how quickly people have decided what they want AI to be.

Faster answers… the best recommendations… confident summaries…
My bigger thought is that for most people, speed feels like progress.
And in a world that already feels like there should be “more”, speed becomes intoxicating.
But while editing my conversation with Roger Martin (who happens to be on my personal Mount Rushmore of business thinkers) for this week’s Thinking With Mitch Joel, I was reminded that we’re not really using AI for speed at all.
We use it to provoke our thinking… and, ultimately, to sharpen our judgment.

We use AI to expose thinking gaps.

Roger and me (see what I did there?) kept returning to a distinction that matters more than most people who use it for work care to admit.
Most people are using it as an answer engine.
A much smaller subset are using it as a sparring partner for thinking.

They are not the same thing.

And choosing the wrong one feels productive… right up until it isn’t.
An answer engine gets the results done fast.
It hands you something that looks finished (and often is).
But strategy doesn’t suffer from a lack of answers.
It almost always suffers from poorly framed questions.
Roger also reminded me…
Most analytics assume the future will look like the past.
Most strategy depends on the future not doing that.

So when leaders ask AI for answers, what they’re often doing is accelerating a flawed premise.

They’re moving faster… possibly in the wrong direction… with amplified confidence.
That’s not intelligence.
That’s automation of assumption.

What we should be advocating for is something more demanding.

AI that pushes back.
AI that forces clearer problem definition.
AI that doesn’t say “here’s what to do,” but instead asks,
“Are you sure this is the real problem?”
“What would have to be true for this to work?”
“What are you assuming away?”

That kind of AI doesn’t feel efficient… it feels slow… frustrating… occasionally irritating.

Which is exactly why it works.
Which is exactly what I want… so that’s how I use it.
Because judgment isn’t about speed or task completion.
It’s about discernment.
It’s about knowing when more data clarifies… and when it distorts.
When optimization sharpens… and when it dulls difference into sameness.
When an answer creates momentum… and when it quietly locks you into the wrong path.

If AI does your thinking for you, it weakens leadership.

If AI helps you think better, it strengthens it.
Which brings us to the next issue… many organizations don’t actually want better thinking.
They want faster justification and task completion.
They want certainty they can point to.
A recommendation they can defend.
A system they can defer to… or blame… if something doesn’t work.

The best leadership has never been about having the best answer.

It’s been about taking responsibility for the choice… even when the data can’t save you.
Said another way, how a company uses AI reveals whether leaders know where they’re going… or are simply moving faster.
So maybe the real strategic question in this moment isn’t how quickly AI can help us decide.
Maybe it’s whether we strengthen judgment instead of replacing it.
Because speed is abundant now… and abundance makes it boring.
Answers are cheap.
Thinking… real thinking… is still scarce.

And that’s exactly why this conversation with Roger matters… enjoy the conversation…

Before you go… ThinkersOne is a new way for organizations to buy bite-sized and personalized thought leadership video content (live and recorded) from the best Thinkers in the world. If you’re looking to add excitement and big smarts to your meetings, corporate events, company off-sites, “lunch & learns” and beyond, check it out.