There's a bad question people keep asking about AI.

Is it going to decide for us?

I understand why the question comes up.

But in my daily use, that isn't where the interesting part sits.

I don't ask AI to decide for me.

I ask it to help my thinking mature faster.

That isn't the same thing.

Judgement rarely arrives in one go

In intellectual work, we sometimes picture a decision as a clean moment.

You look at the elements.

You make the call.

You move on.

In practice, a lot of decisions mature through small successive corrections.

An idea looks good in the morning. Then an objection appears. An example is missing. A group of users was never taken into account. A source contradicts the original hunch. A sentence sounds right, but it hides a weakness. An assumption seemed solid, and then you work out that it rests mostly on an impression.

Judgement isn't only about choosing.

It's about recognising what matters.

And that often takes time.

You reread. You walk. You talk it over. You sleep on it. You come back to the text two days later. A detail resurfaces. A question comes up. Something you hadn't seen becomes obvious.

That delay is part of the work.

But it also costs something.

What AI speeds up

AI doesn't remove the time an idea takes to mature.

It speeds up some of the steps.

It can surface very quickly what you would have seen later anyway:

• a blind spot;
• a contradiction;
• a shaky assumption;
• a fuzzy definition;
• a missing example;
• an obvious objection;
• a missing source;
• a consequence nobody dealt with;
• a mix-up between two neighbouring subjects.

There's nothing magic about it.

It isn't because AI "knows better".

It's because it can quickly test an idea against several frames: the goal of the work, the sources available, the decisions already made, the rules of method, the texts already written, the questions you tend to forget.

It gives you a first resistance.

And that resistance is valuable.

An objection is not a decision

One distinction still has to stay very clear.

When AI flags a gap, it isn't making the call.

It's offering something to look at.

Judging it is then down to the human.

A blind spot can be real.

It can also be beside the point.

A contradiction may need resolving.

It may also be worth keeping, because it says something true about the problem.

An objection can matter.

It can also come from reasoning that is too general, too cautious, too by-the-book.

The danger would be to take every piece of feedback from the AI as a correction to apply.

That would mean swapping your judgement for compliance.

That isn't what I'm after.

I want AI to widen the surface an idea gets tested against.

Not to choose for me what survives.

Using it well: challenge, don't conclude

In my work, the best questions to put to an AI aren't always requests for output.

They're often requests for pushback.

For instance:

• what's missing?
• what moves too fast?
• which distinction isn't clear?
• what objection could a reader raise?
• which assumption is still implicit?
• which passage repeats an idea already covered elsewhere?
• what isn't proven enough?
• which part of the reasoning rests mostly on my own hunch?

These questions change the relationship.

The AI is no longer there only to write.

It's there to put the thinking to work.

It pushes. It tests. It objects. It asks you to be more precise. It flags that the idea may well work, but that it isn't stable enough yet.

That isn't pleasant in the comfortable sense.

But it's useful.

Why context changes the quality of the feedback

An AI in a chat window can already produce a list of blind spots.

Sometimes that's useful.

But it often stays generic.

It can tell you to be clearer about the audience, add an example, define your terms, soften the claim.

Those are good pieces of advice.

But they're advice you could give to almost any text.

With a Context, the feedback becomes more situated.

The AI can compare what you've just written with what already exists in the case. It can see that an idea has already been covered in another section. It can spot that an agreed decision isn't being followed. It can notice that an important source hasn't been used yet. It can flag that a passage contradicts a nuance you kept somewhere else.

The feedback no longer comes only from a general ability to write or analyse.

It comes from the meeting of:

• an intention;
• raw input;
• a memory;
• rules;
• the real state of the work.

That's what makes it more useful.

AI gives you handles

There's one very concrete thing I value in how this works.

AI turns a vague impression into a handle on the work.

I used to feel that a text wasn't right yet without knowing exactly why.

Something was missing.

But what?

An example? A transition? An objection? A definition? A consequence? A sentence that stays too abstract?

AI helps cut that unease into pieces.

It can say: this passage repeats a point you've already made, this section comes too early, this distinction isn't prepared enough, this promise isn't kept, this conclusion opens a subject the text never covered.

After that, I can decide.

I can correct.

I can refuse.

I can keep the tension as it is.

But I'm no longer facing a vague feeling.

I have handles.

Judgement matters more, not less

The more useful AI becomes, the more human judgement matters.

That's counter-intuitive.

You might think AI shrinks the space left for judgement, because it produces, analyses, summarises, rewrites and criticises.

I think the opposite.

It increases the number of proposals, objections, variants and possible paths.

So you have to judge more.

You have to know:

• which objection deserves to be dealt with;
• which suggestion is too generic;
• which simplification betrays the idea;
• which nuance weighs the text down for nothing;
• which contradiction should stay visible;
• which source is enough;
• which risk is acceptable;
• which judgement call really binds the work.

AI speeds up the circulation of possibilities.

But it doesn't automatically hand you the right choice.

Judgement doesn't disappear.

It moves up a level.

What this changes for a product manager

In product management, this difference is essential.

A PM isn't only short of documents.

They're often short of the time to test the signals against each other.

One customer comment looks important. Another contradicts it. A technical constraint changes the scope of the problem. A commercial opportunity pushes one way. A product reality pushes the other. An assumption looks attractive, but the evidence is thin.

The risk isn't only that you don't have enough information.

The risk is that you don't test the information enough before turning it into a decision.

In that setting, AI can help you slow down intellectually while speeding up the work.

The phrasing is odd, but it seems right to me.

It speeds up the gathering, the testing, the rewriting, the hunt for gaps.

And because it does all that fast, it leaves more room for the real slow part: judgement.

Do I believe this?

Does this objection count?

Is this decision robust?

Is this idea worth pursuing?

Are we solving the right problem?

Don't confuse speed with delegation

The risk, obviously, is confusing speed with delegation.

Because AI is fast, you can be tempted to take its answer for a conclusion.

Because it phrases things well, you can believe the idea is clear.

Because it hands you a structured list, you can believe the problem is covered.

But good wording proves nothing.

A list that looks complete is no guarantee that the right subjects are in it.

A well-written objection isn't necessarily relevant.

The human's role is precisely not to be hypnotised by the form.

They have to stay responsible for the meaning, the business context, the level of evidence, the final decision.

AI can speed up how an idea matures.

It can make thinking more inspectable.

It can show you sooner what's missing.

But it doesn't automatically turn an idea into a truth.

A better intellectual solitude

There's one last, more personal side to this.

A large part of intellectual work happens alone.

Even inside a company, even with a team, even with meetings, there's always a moment where you have to think on your own: write a note, prepare a decision, put a hunch into words, work out why something bothers you.

AI changes that solitude.

It doesn't remove it.

It makes it less silent.

You can put down an idea that isn't finished yet and get friction straight away. Not automatic validation. Not a truth. Friction.

Something answers.

Something questions.

Something forces you to be precise.

It's in that friction that the idea matures.

And that, for me, is where AI becomes genuinely interesting.

Not when it replaces judgement.

When it gives judgement something to work on, sooner.