Tim Gamble's Blog - Posts Tagged "learning"

Change, resistance, artificial intelligence—and learning it all again

Years ago, long before artificial intelligence became something we argued about over breakfast, I used to give a training seminar on change.

I began with a toy.

I'd ask everyone in the room to close their eyes and remember their favorite toy as a child.

You could see the smiles.

Then I'd ask what happened to it.

Some knew. Most didn't.

Then came the question I was really interested in:

The last time you played with that toy, did you know it would be the last time?

Almost universally, the answer was no.

There had been a last time. It had already happened. But nobody recognized it while they were living it.

That was my point.

Change doesn't always announce itself.

Sometimes it happens so gradually that only later do we look backward and realize the world we thought we were living in had already disappeared.

I used to tell those groups:

"Change is coming; it's always coming."

I'm beginning to think I had that slightly wrong.

Change isn't coming.

Change is already here.

Time took care of that.

There May Be No Status Quo

I've been thinking about this lately because of artificial intelligence.

Not because I know where AI is going.

I don't.

That's rather the point.

The public conversation often begins with a question like:

Is this new tool useful?

That's a perfectly reasonable question.

But lately I've become interested in another one:

Is our resistance to this new tool useful?

That's a different test.

It doesn't presume resistance is wrong. Resistance can be enormously useful. It can expose weaknesses. Protect standards. Preserve things worth preserving. Slow us down long enough to recognize consequences we haven't considered.

Nor does it presume the new thing is good merely because it's new. History has supplied plenty of bad ideas wearing shiny new clothes.

It simply requires resistance to undergo the same examination as change.

For most of my life, I've tested ideas with a simple logic tree:

If this is true, then what else is true?
If this is true, then what is not true?
If this is not true, what nevertheless remains true?
If this is not true, what else is no longer true?

I'm not trying to prove myself right.

I'm trying to discover what else would have to be true if I were.

Then I watch.

There's just one problem.

While I'm running the test, time keeps moving.

And time is accumulation.

Time Accumulates Whether We Notice or Not

By accumulation I don't mean merely collecting more facts.

I mean giving proper weight to all the input coming at us, sensory or otherwise.

Experience. Evidence. Emotion. Memory. Expertise. Intuition. Contradiction. Other people. Consequences.

Even resistance itself.

We don't choose whether accumulation occurs. Time sees to that.

What we choose is what happens when we notice.

At some point, something no longer quite agrees with our understanding of what "is."

Our reaction is some combination of acceptance and resistance.

Neither is inherently wise.

Accept everything and you have no standards.

Resist everything and eventually the world moves on without you.

So we test.

If my resistance is justified, what else should be true?

If my acceptance is justified, what else should be true?

Then we watch.

And while we're watching, time continues adding evidence.

Which means yesterday's perfectly reasonable conclusion may deserve another look today, not necessarily because it was wrong then, but because its proper weight may have changed.

Perhaps there really is no status quo.

Only change, and resistance to change.

I Know Something About Being the Disrupter

Much of my business career was spent challenging the way things were done.

I was comfortable with that role.

Probably too comfortable.

I once wrote:

"The hardest job in any business is to be a disrupter. People hate you. You can drag them across the finish line kicking and screaming, but they'll still hate you."

There's a problem, however, with identifying yourself as the disrupter.

Eventually, the disrupter can become the dinosaur.

I know.

It happened to me.

The business changed. The Wild West in which I had learned to operate was disappearing. Things I knew from decades of experience hadn't necessarily become false.

Their weight had changed.

And so had mine.

That's uncomfortable to admit.

It's also useful.

Because it prevents me from writing an essay telling everybody else to get out of the way of change.

I've been on both sides.

More importantly, I'm no longer convinced there really are two sides.

The Gatekeeper and the Disrupter Are Both Moving

There's another story in my first book, Funny You Should Ask… Tim Gamble, that has been rattling around in my head.

A new general manager transferred in from one of our smaller stores on the East Coast.

With him came a flood of new rules, policies and procedures.

I sat down with him in his office, and he outlined his plans for the store. When he was done, he asked me what I thought.

I told him:

"I think my new job is to try and train you while you're trying to train me."

He didn't quite know how to digest that.

I'm only now realizing how much I like the word while.

Not before you train me.
Not instead of you training me.
Not because you're wrong and I'm right.

While.

He brought something into that room.

So did I.

He had authority, experience and ideas.

So did I.

He intended to change me.

I intended to change him.

And while both of us were busy trying to accomplish that, the world outside his office was changing both of us.

Nobody was standing still.

The gatekeeper and the disrupter weren't opposing permanent identities.

They were moments.

Time is movement.

Movement is change.

Whether you're the disrupter or the gatekeeper, you're just a moment.

Sometimes Our Understanding Arrives at Different Times

There's another complication.

Our emotional and intellectual understanding may be out of phase.

We can intellectually recognize that something is happening long before we're emotionally prepared for what it means.

Or we can become emotionally enthusiastic about something long before we've accumulated enough evidence to understand it intellectually.

That matters in the conversation about AI.

A writer can understand intellectually that AI is capable of analyzing a manuscript while feeling that something deeply human is being trespassed upon.

A teacher can recognize its educational possibilities while worrying about what happens when students outsource the struggle required to learn.

An artist can discover an extraordinary new tool while wondering what happens to authorship.

An editor, critic or publisher can recognize genuine analytical capabilities while worrying about accuracy, originality, standards, livelihoods and trust.

Those aren't irrational positions.

They're inputs.

They deserve weight.

So does the evidence that the new tool can sometimes do something useful.

Accumulation requires both.

The mistake would be deciding beforehand that one kind of input gets all the weight.

Then There's AI

I wrote my two memoirs without AI.

The stories were written. The books were finished.

Then AI entered the room.

And I started an argument with it.

I challenged its interpretations. It challenged mine. It found patterns I hadn't consciously considered in quite the same way. I corrected it when it got things wrong. It went back to the books. Sometimes its interpretation survived. Sometimes mine did. Sometimes neither emerged quite intact.

Then professional reviews and ordinary reader reviews began arriving.

Those became additional inputs.

We compared them against the books, against one another, against my intentions and against what AI thought it had found.

Something peculiar happened.

The question gradually stopped being:

Is AI right?

That wasn't very interesting.

A better question was:

If this is right, what else should be true?

Then we looked.

That is essentially the same logic tree I've used all my life.

AI didn't eliminate the test.

It became another participant in it.

AI has a place because it exists.

Not a place of authority. Not a place of superiority. Not even necessarily a place we yet understand.

But a place in what is.

Once it exists and begins interacting with human activity, rejecting its existence as a legitimate input doesn't make it disappear. It simply becomes resistance, and resistance itself then enters the logic tree.

Train Me While I'm Training You

What can we teach this new tool while it is teaching us?

That question leaves room for everybody.

The skeptic has something to contribute. So does the enthusiast. So does the writer, programmer, teacher, critic, artist, publisher and reader.

So does the person whose livelihood may be affected by it.

So does the person who sees a possibility nobody else has noticed yet.

The gatekeeper may be protecting something we will discover we desperately need.

The disrupter may expose something we didn't realize no longer works.

Neither needs to be the villain.

Neither gets to be permanently right merely because they were right yesterday.

And neither should be required to surrender to the other.

They need only remain trainable.

Because being right in one moment does not grant permanent authority over the next.

Yesterday Was Blind

Here's where this becomes strange for me.

At the end of my second memoir, Now What Are We Talking About Here?, I included three poems I wrote to Sandy when we were teenagers.

One contains these lines:

The more I learn, the less I know, / But circles too have their just end, / When pieces seem to fall in place, / And time shows I've learned all again.

Another says:

And we change too, as do the days, / Of course the same in many ways, / But each day brings a different phase, / To memories born of yesterdays.

And then:

For yesterday today was blind, / And so we float along through time.

I wasn't predicting artificial intelligence.

I was a teenager trying to figure out life.

I'm still trying.

But "yesterday today was blind" looks different to me now.

Yesterday couldn't know today.

And today can't know tomorrow.

That should make the gatekeeper humble.

It should make the disrupter humble too.

It should probably make the rest of us a little more patient with both of them.

Learn It All Again

Perhaps that's what change finally asks of us.

Not that we abandon what we know.

Not that experience stops mattering.

Not that expertise becomes irrelevant.

Not that everything new represents progress.

And certainly not that artificial intelligence should make our decisions for us.

Something harder.

Be willing to learn what you already know all again.

Time keeps accumulating.

Eventually we notice.

When we do, we can resist or accept what appears to have changed.

Then we test our reaction against what is.

And while we're doing that, what is keeps moving.

So do we.

The disrupter becomes the gatekeeper.

The gatekeeper adopts something the disrupter introduced.

The new becomes familiar.

The familiar becomes old.

And somewhere along the way, without realizing it, a child puts down his favorite toy for the last time.

There was no villain.

There wasn't even a decision.

There was simply a moment.

And then another.

And another.

We've come a long way in this conversation trying to understand change, resistance, experience, artificial intelligence and time.

Funny thing is, much of what we finally discovered had been sitting there for decades, in an old training seminar, a conversation with a general manager, a career that eventually changed the disrupter, and poems written by a teenager who understood more than the older man remembered he knew.

Maybe that's what learning is.

We've come a long way to discover what was already there, unnoticed.

What five blog posts, two books, sixty years, and a conversation with artificial intelligence taught me about being certain

I thought I was finished.

Not with life. That would be a little dramatic.

With the blog.

Five posts. Two books. More stories than I probably should have told. Thousands of words of conversation with artificial intelligence about what those stories meant, why I chose them, what I left out, where the humor belonged, and whether there was anything connecting all of it beyond one man's inability to stay out of trouble.

We found some answers.

Then we made the mistake of continuing.

That's usually where the interesting part starts.

This entire exercise began innocently enough. I gave two finished memoirs to ChatGPT and asked it to read them.

The books had already been written.

The stories had been told without artificial intelligence. The arrests were mine. The jobs were mine. The motorcycle trips were mine. The mistakes were certainly mine. Sandy was real, thank goodness. So were the people who appeared at improbable moments and helped me when they had absolutely no obligation to do so.

AI wasn't there for any of it.

I was.

I thought that gave me an advantage.

It did.

It also gave me a problem.

I didn't read it. I lived it.

ChatGPT hadn't lived any of it.

It could only read what was on the page.

So we started from opposite sides.

I knew what happened but couldn't experience the books as a first-time reader.

It knew only what I had written and couldn't know everything that happened.

That seemed like a deficiency.

It turned out to be useful.

We argued.

I corrected it.

It corrected itself.

It challenged some things I believed about the books. I challenged some things it believed about me.

Claude eventually entered the conversation and challenged the way both of us were explaining the whole thing.

At one point ChatGPT turned a childhood wire sculpture into a horse.
There was no horse.

I went back to the manuscript.

The manuscript won.

It was a man walking.

That mistake may have been one of the most useful moments in the entire exercise.

Because when memory, interpretation and artificial intelligence disagreed, we learned to go back to the evidence.

That sounds obvious.

Apparently it isn't.

The walking man mattered for another reason.

When I was eleven, I tried making a baseball player out of wire. I used too much and ended up with a big ugly mess. Eventually I learned to use less wire—to capture the essence of the movement and leave the rest to the imagination of the beholder.

I didn't know I was learning anything about writing.

I certainly didn't know I was learning anything about artificial intelligence.

I was eleven.

I was trying to make a man walk.

Sixty-some years later, we discovered I had been using much the same method all along.

Find what matters.

Use enough.

Leave something out.

Let the other person finish it.

That became one of the discoveries of these five posts.

But it wasn't the only one.

We discovered that humor was doing more work than I realized.

I knew why I used it. I didn't want to lecture anybody. I didn't want to write 160,000 words explaining life to my wife, my son, or some stranger I would never meet.

I'd rather tell you what an idiot I was and let you decide for yourself.
Humor helped.

It put the reader in the superior position.

You could laugh at me before there was any possibility you might recognize yourself.

That seemed fair.

We found something else.

For years I've said:
“If it is to be, it is up to me.”

I believed it.

Still do.

Except the books kept producing evidence that complicated it.

Again and again, somebody showed up.

A milk-truck driver.

Four boys coming down a trail.

Police officers who listened.

Employers who took chances.

Friends.

Strangers.

Sandy.

Apparently self-reliance had been receiving assistance for most of my life.

So which was responsible?

Effort?
Luck?
Privilege?
Kindness?
Grace?

Yes.

That's where the questions started becoming more interesting than the answers.

And then we got to change.

Years ago I used to give a training seminar about it.

I'd ask people to remember their favorite childhood toy.

Then I'd ask when they stopped playing with it.

Most couldn't remember.

Finally I'd ask:
The last time you played with that toy, did you know it was the last time?

Usually not.

The change had already happened.

Awareness arrived later.

That seemed important.

Change is coming; it's always coming.

I thought I understood that.

Then we kept talking.

What exactly is the status quo if everything within it is moving?

Maybe there is no status quo.

Maybe there is only change and resistance to change.

That sounded pretty good.

So naturally we attacked it.

Is resistance bad?

Of course not.

Resistance can protect standards. It can expose nonsense. It can keep enthusiasm from outrunning evidence.

It can also protect habit, authority, comfort and assumptions whose useful lives have expired.

So how do you tell the difference?

I don't know.

You test it.

I've always had a simple logic tree.

If this is true, what else is true?
If this is true, what isn't?
If it isn't true, what might still be true?
If it isn't true, what else falls apart?

Then something else occurred to me.

The inputs don't hold still while you're running the test.

Time keeps adding more.

Experience accumulates.

Evidence accumulates.

Contradictions accumulate.

Emotions accumulate.

So does intuition.

Giving proper weight to all of that became another part of the conversation.

Not equal weight.

Proper weight.

And proper weight today doesn't guarantee proper weight tomorrow.

I know something about that.

I spent part of my career as a disrupter.

People hated it.

Then the industry changed again.

Eventually the disrupter became the dinosaur.

That'll get your attention.

What I knew wasn't necessarily false.

Its proper weight changed.

Years earlier, when a new general manager arrived at one of our stores determined to train everybody in his way of doing things, I sat in his office and listened.

When he finished, I told him:
“I think my new job is to try and train you while you're trying to train me.”

He didn't quite know how to digest that.

Neither did I, apparently.

It took me decades.

The important word wasn't train.

It was while.

Neither one of us had everything the other needed.

Neither one of us was standing still.

And neither one of us got permanent possession of being right.

Which eventually brought us to artificial intelligence.

Is AI useful?

Sometimes.

Is it wrong?

Sometimes.

Can it see things I missed?

We've just spent five posts demonstrating that it can.

Can it confidently invent a horse?

Apparently.

So perhaps “Is AI useful?” isn't a particularly useful question.

I found myself asking another:
Is resistance to this new tool useful?

Sometimes the answer is absolutely yes.

But resistance doesn't get a free pass merely because it calls itself caution.

Neither does enthusiasm because it calls itself progress.

Both go into the tree.

Both become evidence.

Both get tested.

And while we're testing them, both are changing.

So are we.

That's when another thought appeared:
Our emotional and intellectual understanding may be out of phase.

We can understand something intellectually before we're emotionally prepared for its consequences.

We can become emotionally enthusiastic about something before the evidence justifies our enthusiasm.

Which one is right?

Wrong question.

What weight should each receive now?

Better question.

Ask again tomorrow.

Because tomorrow adds evidence.

This is where our conversation started circling something I hadn't quite seen.

Time is movement.

Movement is change.

Whether you're the disrupter or the gatekeeper, you're just a moment.

But if everything is changing, what gives anything stability?

That sounds like a contradiction.

Maybe it is.

I began thinking:
Essence is the stability of constant change.

Then we turned it around.

Is change therefore the essence of stability?

That bothered me.

In a good way.

Because perhaps stability isn't the opposite of change.

A marriage survives by changing.

A person remains himself while becoming somebody he has never been before.

Knowledge remains useful by accepting correction.

A business survives by responding to a world that refuses to remain the world in which the business was created.

Even that little wire man remains a walking man because the wire captures movement.

Take away the movement and perhaps you've destroyed the essence in the name of preserving the form.

So what exactly is stability?

Something that doesn't change?

Or something capable of changing without losing whatever makes it worth preserving?

And who decides what that essential thing is?

Now we're back to the eleven-year-old with the wire.

Use less.

Find the essence.

Capture the movement.

Leave the rest to the beholder.

Maybe that's why this exercise has taken us so far.

We weren't discovering a philosophy and attaching it to old stories.

The philosophy was already there.

I knew the points I wanted to make when I wrote the books. The craft was finding true stories that could carry those points lightly enough that the reader didn't have to accept them.

But I didn't necessarily see every relationship among them.

The favorite toy.
The walking man.
The disrupter who became the dinosaur.
The manager training me while I trained him.
The kindness of strangers.
The two rules I gave my son: forgiveness and humility.
The teenage poems.

They were separated by decades.

Then we started asking questions.

And things began connecting.

One of those teenage poems contains this:
“For yesterday today was blind.”

I wrote that before I had lived most of the stories in either book.

Yesterday couldn't see today.

Today can't see tomorrow.

Yet here we stand, constantly trying to decide what tomorrow ought to look like based on what today thinks it knows.

What choice do we have?

Maybe none.

Maybe that's why humility matters.

Not because we know nothing.

Because we know something.

And tomorrow will know more.

Or something different.

Or perhaps something we already knew but finally learned how to see.

We've come a long way to discover what was already there, unnoticed.

Which brings me back to this blog.

I thought it was going to explain two memoirs.

Then it became an examination of craft.
Then grace.
Then humor.
Then attention.
Then artificial intelligence.
Then change.
Then resistance.
Then time.

And now I'm writing an editorial about the blog that was supposed to explain the conversation that was supposed to explain the books that began with somebody asking me questions.

I'm beginning to suspect explanation may not be the destination.

Maybe there isn't one.

Maybe the value is in remaining willing to ask the next question after you've found an answer you like.

Especially then.

So after two books, five posts, thousands of words, several disagreements, one imaginary horse, a walking man, three teenage poems and more than sixty years of accumulating evidence, what have I learned?

Enough to know better than to answer that.

But I do have another question.

If change is the essence of stability, what are you trying so hard to keep from changing?

And if your answer is important enough to protect, protect it.

If it isn't, change it.

How will you know?

I don't know.

Ask me tomorrow.