AI Raises the Bar: Why the Insight Industry Must Lead on Better Thinking

We shall not cease from exploration
And the end of all our exploring
Will be to arrive where we started
And know the place for the first time.”
— T. S. Eliot

The debate about AI has moved on.

We no longer need to ask whether AI will change the world of knowledge work. It already has. The more important question now is how we design human–AI collaboration so that human judgement is amplified, not diluted.

We are beginning to recognise a crucial truth: as AI gets better, human capability becomes more valuable, not less. The central challenge for organisations is not simply how to use AI more efficiently. It is how to help people work with AI in ways that improve their judgement, sharpen their creativity and strengthen their capacity to take informed, intelligent action.

This is good news for insight professionals.

Long before AI arrived, the best insight professionals were already practising many of the cognitive skills we now need more urgently than ever. They knew how to question, challenge, probe, imagine, connect, doubt, refine and take intellectual responsibility.

AI has not made these capabilities obsolete. It has made them more visible, more valuable and more necessary.

So we found ourselves asking: before AI, what distinguished the outstanding insight professional from the merely competent one?

What made some people stand head and shoulders above others in their ability to think critically and creatively?

What separated those with vitality, curiosity and intellectual energy from those who simply accepted what was in front of them and went with the flow?

The argument of this article is simple. If insight professionals can dial up the best of what they have always been capable of, they are well placed to lead the new agenda around the quality of human thinking in the AI era.

So let’s look at what the insight industry can deliver when it brings its A-game to the party.

Metacognition: thinking about thinking — and enjoying it

The best insight professionals possess a high degree of metacognition, even if they would not necessarily use that word.

They monitor the quality of their own thinking.

They ask themselves:

Do I really understand this?

Have I dug deeply enough?

Have I examined the underlying assumptions?

Have I teased out the hidden constraints?

Am I solving the real problem — or merely the obvious one?

Have I done enough thinking to earn this conclusion?

This is the difference between deep critical thinking and default thinking.

Default thinking accepts the easy frame, the obvious answer, the familiar pattern. Deep thinking asks whether the frame itself may be wrong.

That is why the famous Einstein principle — often paraphrased as spending most of the available time defining the problem before attempting the solution — remains so powerful. Whether or not the wording is exact, the underlying point is profound: the quality of the solution depends on the quality of the question.

It is also why Germaine Greer’s reported listing of “thinking” as a hobby has always been such a wonderful provocation. For some people, thinking is not merely a professional activity. It is a form of engagement with the world.

Steve Jobs captured the discipline beautifully when he observed that simple thinking is hard work, but worth the effort because, once you get there, it can move mountains.

Outstanding insight professionals enjoy this effort. They enjoy the search for cleaner thinking. They enjoy the mental discipline of stripping away clutter until the real issue comes into view.

That is precisely what we need now when working with AI: not less thinking, but better thinking about our thinking.

This matters because metacognition goes to the heart of one of the defining challenges of AI collaboration: understanding the difference between intelligent cognitive offloading and cognitive surrender.

Intelligent cognitive offloading means using AI to take on work it can do powerfully — scanning, structuring, summarising, generating, comparing and reframing — so that human creativity, judgement and strategic thinking can flourish where they matter most.

Cognitive surrender is different. That is the slippery slope where humans become intellectually lazy, default to AI outputs and stop applying the checks, refinements, doubts, instincts and judgement that turn a plausible answer into a genuinely valuable one.

Forensic intellectual energy

Digging deeper into metacognition takes us to the quality we might call forensic intellectual energy.

Years ago, we spoke to a top barrister and asked what had helped him reach the top of his profession. His answer was memorable:

“I have this forensic energy. When I hear the opposing barrister say something flawed, weak or illogical, I can’t wait to jump out of my seat, dismantle the argument, demonstrate the true logic and facts in play, and win the judge and jury over to my point of view.”

That is not just legal skill. It is intellectual aliveness.

It is the refusal to let sloppy thinking pass unchallenged. It is the instinct to interrogate an argument, test its foundations, expose its weaknesses and replace it with something stronger.

In the AI era, this matters enormously.

AI can produce answers that sound fluent, plausible and confident. But the outstanding human collaborator does not simply admire the fluency. They ask:

Is this actually true?

Is the logic sound?

What has been missed?

Where is the weak link?

What assumption is hiding beneath the surface?

That forensic energy is now one of the key safeguards against cognitive surrender.

Integrity of thought

This takes us to integrity of thought.

Stephen King, reflecting on creative writing, talks about a mistake that is not simply factual — not the equivalent of putting a service station on a motorway where no service station exists.

The deeper mistake is when, on rereading a draft, the writer knows that something is intellectually lazy.

A character says something they would never really say.

A plot event happens only because it is convenient.

A piece of dialogue exists merely to move the story along.

Deep down, the writer knows they have cheated. They have allowed something false into the work because it made life easier.

That is a failure of craft.

The same applies to consulting, strategy, analysis and communication.

The best insight professionals have an internal alarm bell that sounds when the thinking is too convenient. They know when an argument has been forced, when a conclusion has been smuggled in, when a recommendation rests on an assumption no one has properly tested.

They do not accept the easy workaround if it violates the integrity of the work.

This is exactly the kind of discipline we need when working with AI. AI can help us move faster. But speed without integrity simply accelerates weak thinking.

The willingness to agonise through the process

Another trait of outstanding insight professionals is their willingness — even their strange enjoyment — to engage in intellectual struggle.

They do not treat a difficult problem like a hot potato to be passed quickly to someone else. They stay with it. They wrestle with it. They go through second, third and fourth drafts. They wake up still thinking about it.

We once employed someone who eventually left the world of knowledge work and marketing because, as they put it, they wanted a job that did not involve so much agonised thinking.

They did not like having to wrestle with a problem overnight. They did not like the repeated process of drafting, rethinking, reframing and refining.

So we asked what they planned to do instead.

Rather surprisingly, they said they were going to train as an airline pilot.

At first, this was slightly disturbing news.

But their point was understandable. They wanted a role where the thinking was more bounded. Learn to fly the plane. Land it safely. Switch off. Sleep at night. Turn up the next day and fly back.

They did not want to redesign the aircraft, rethink the airport, or reimagine the future of aviation.

In creative knowledge work, by contrast, the “boys in the basement” — the subconscious mind — keep working overnight. The first draft is never the final answer. The problem keeps evolving. The thinking keeps deepening.

Excellent insight professionals accept this. Often, they thrive on it.

And in the AI era, this willingness to iterate becomes even more important. AI gives us first drafts at speed. But the human value lies in knowing that the first draft is only the beginning.

The ability to smell the rat — and take responsibility for sorting it out

Another defining trait is the ability to sense when something does not feel right.

The best insight professionals can “smell the rat”.

They notice the statistic that looks too interesting.

They spot the claim that feels too neat.

They hear the story that sounds persuasive but hollow.

They sense when the data is pointing one way but the human reality is pointing another.

In market research, we used to refer to Twyman’s Law, named after the respected researcher Tony Twyman:

“If a statistic looks interesting or unusual, it is probably wrong.”

The principle is not that surprising data should be dismissed. The principle is that interesting data deserves investigation.

The outstanding insight professional does not simply accept things at face value. They take responsibility for checking what does not feel right.

They know that data can be dumb.

They know that beliefs can be blind.

They know that elegant analysis can still be wrong.

In today’s AI-powered world, this instinct is invaluable.

AI may give us an answer that is beautifully structured, impressively worded and apparently authoritative. But the human collaborator still needs to ask:

What feels off here?

What needs checking?

What is being assumed?

What would I need to verify before acting on this?

That instinctive rat-smelling capability may become one of the most important human skills of all.

Treating contradiction and complexity as old friends

There is something special about insight professionals who understand the difference between unnecessary confusion and genuine complexity.

Unnecessary confusion should be stripped out. It clouds the issue, weakens the recommendation and creates noise.

But genuine complexity must be respected. It cannot simply be airbrushed away because it makes the answer less tidy.

The best insight professionals know this distinction.

They can simplify without becoming simplistic. They understand the wisdom often associated with Einstein: make things as simple as possible, but no simpler.

This matters greatly in the AI era.

AI can be very good at producing neat, shiny, convincing answers. But the world is not always neat. Organisations are not neat. Human beings are not neat.

Real life contains ambiguity, contradiction, emotion, compassion, humour, quirkiness, politics, vanity, stupidity and hope.

The best insight professionals do not pretend these things are irrelevant. They build them into the thinking.

Alongside this comes another vital quality: epistemic humility.

Even after arguing strongly for a point of view, the outstanding insight professional retains respect for other perspectives. They know there are limits to their own knowledge, judgement and interpretation.

That humility is not weakness. It is intellectual maturity.

And when working with AI, it becomes essential.

And the good news for insight professionals is that the list of outstanding skills we have on offer putting us at an advantage in the AI era doesn’t end here.

So the second part of the article we look at further critical skills and explain how as an industry we can forge these to our advantage to be at the frontier of the improved human thinking now urgently needed in the AI era.