August 5, 2026 • 29 mins
I'm Sahar Andrade, MB.BCh, a physician turned Neuroleadership Coach and a Forbes Coaches Council member.
AI adoption fails in companies with exhausted leaders because adoption is not a technology decision.
It is a decision-capacity decision. Deloitte's 2026 Global Human Capital Trends found that 60 percent of executives use AI in decision-making and only 5 percent say they manage it well.
That gap is where the failure lives. Under sustained load, the exact thinking AI adoption requires, holding ambiguity and judging unfamiliar output, becomes the most expensive thinking a leadership team can produce. Here is the mechanism, and the three moves that change the sequence.
In this episode: why the standard governance reading of the adoption numbers is incomplete, what sustained load does to the exact thinking an AI decision requires, the five patterns that show up in stalled rollouts, and three moves that change the sequence. Steady before strategy, as an operating requirement rather than a personal virtue.
Sources cited: Deloitte 2026 Global Human Capital Trends (more than 9,000 leaders, 89 countries). EY US Technology Pulse Poll, February 2026 (500 US technology-industry business leaders).
Sahar Andrade, MB.BCh, is a Neuroleadership Coach, Forbes Coaches Council member, and host of AI Café Conversations podcast (top 2% globally in search visibility). Teaching executives how the nervous system shapes leadership under pressure.
The next Pressure Room for executives is Thursday, August 27, 2026, 11am Pacific. Free, 45 minutes. Register at
https://www.saharandrade.com/pressure-room
#AIdecisionfatigue #executivedecisionmaking #neuroscienceofdecisions #leadershipunderpressure #cognitiveoverloadatwork #AIcognitiveload #executiveburnoutneuroscience #AIExhaustion #NeuroscienceLeadership #ExecutiveBurnout #RegulatedLeadership #AILeadership #NeuroscienceOfLeadership #LeadershipBurnout #AIForExecutives #AIAdoption #LeadershipDevelopment #FutureOfWork #AICafeConversations #neuroleadership #ExecutiveLeadership #AITransformation #AIStrategy #ExecutiveCoaching #HumanCenteredAI #LeadershipPodcast #NotechRequired #neuroleadership #humancenteredleadership #HRLeadership #regulatetolead #RegulationFirstLeadership #NervousSystemLeadership #NeuroleadershipCA #NervousSystemAtWork #CoRegulationAtWork #ExecutiveNervousSystem #AILeadershipNeuroscience #NeuroscienceExecutive #ExecutiveCoaching #BurnoutRecovery #WorkplaceWellbeing #AILeadership #LeadershipTraining #NeuroscienceOfpressure #HighPerformerBurnout #LeadershipTraining #PeopleFirst #ChangeManagement
1. Why does AI adoption fail in companies with exhausted leaders?
2. What does sustained pressure do to executive decision-making?
3. Is AI adoption failure a technology problem or a leadership problem?
4. Why do 60 percent of executives use AI but only 5 percent manage it well?
5. What should leaders do before rolling out AI across a team?
6. How do exhausted leaders make AI decisions differently?
---
AI Cafe Conversations: Neuroscience-based AI leadership for executives. Hosted by Sahar (The AI Whisperer) | New episodes Wed & Fri
🔗 Connect: https://www.linkedin.com/in/saharandradespeaker/
📧 Work with me: sahar@saharconsulting.com
🌐 Website: https://www.saharconsulting.com/
📧 Instagram: https://www.instagram.com/saharthereinventcoach
Listen
Watch
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
SPEAKER_00 (00:13):
Picture an AI rollout that goes exactly to
plan.
Eighteen months, read budget.
The tools work, the traininghappens, the dashboard go live.
And a year later, almost nothinghas changed about how decisions
actually get made.
(00:35):
Everyone blames the technology.
In my opinion, they are lookingat the wrong thing.
Welcome to the AI CafeConversation Podcast.
I'm Sahar Andradi, and this iswhere the neuroscience of
pressure meets the future ofleadership with an AI twist.
(00:56):
Before we start, the same line Isay every week.
Everything I share comes from myexperience and my opinion as a
coach.
It's not a diagnosis and it'snot a fact about you.
You decide what fits.
Today's question is one I hearin some form almost every week,
(01:17):
in different words fromdifferent rooms.
Why does AI adoption fail incompanies with exhausted
leaders?
Not companies with badtechnology, not companies with
thin budgets, not companies thatwere slow to start.
(01:38):
Companies with good tools, realmoney, smart people, a serious
plan, and a leadership team thathas been running at full output
for three years straight.
I'm going to make a case today.
For the first few minutes, itwill sound like a soft argument.
Stay with me.
(01:59):
The mechanism underneath it'snot soft at all.
It's one of the most establishedfindings in the study of
pressure and human thinking.
And almost nobody puts it on theAI strategy slide.
And I want to say why thismatters to me specifically.
I spent years in medicine beforeI did this work.
(02:21):
What I carried out of thattraining was not a set of
diagnoses, it was a habit.
When something keeps failing inthe same shape across very
different environments, stopblaming the environment.
Go look at the mechanism thatall of them share.
Right now, AI adoption isfailing in the same shape across
(02:45):
very different companies.
So let's go look at themechanism they share.
Let's talk about the crisis.
Let me start with where weactually are, because the
numbers are stranger than theheadlines suggest.
Deloitte's 2026 Global HumanCapital Trends Report surveyed
(03:10):
more than 9,000 business andhuman resource leaders across 89
countries.
60% of executives said they arealready using AI in decision
making.
Only 5% said they manage itwell.
Sit with those two numbers sideby side, 60 and 5.
(03:35):
That is not an adoption problem.
Adoption already happened.
That is a 55-point gap betweenusing a thing and being able to
hold it.
Then look at what Ernest andYoung published in early 2026.
The US technology passportconducted in February surveyed
(03:59):
500 business leaders workinginside the technology industry.
These are the people closest tothe tools, the ones with the
most fluency, the most budget,and the fewest excuses.
78% said AI adoption isoutpacing the organization's
(04:20):
ability to effectively managethe business.
More than half, 52% saiddepartment level AI work is
running with no formal approvalor oversight at all.
I want to be precise about thatsecond study because precision
is the whole job.
(04:41):
That is technology sectorleaders in the United States.
It's not every industry in everycountry, and I'm not going to
stretch it into something it didnot measure.
But if the people who build thetools are telling you the pace
has outrun the ability tomanage, that is a signal worth
(05:04):
taking seriously.
Now, there is a standard readingof these numbers, and you have
probably heard it in your ownbuilding.
It's a governance reading.
The story goes like this: companies moved too fast, they (05:16):
undefined
skipped the policy work, theyneed better frameworks, tighter
controls, a committee, acharter, a risk register, a
center of excellence.
That story is not wrong.
(05:36):
It's incomplete.
Because governance is not adocument.
Governance is a series ofdecisions made by specific human
beings on ordinary weekdaysunder time pressure about things
they have never seen before.
(05:57):
A framework does not evaluate amodel output.
A person does.
A charter does not decidewhether to override a
recommendation that contradicts20 years of experience.
A person does.
In about 90 seconds in a meetingthat is already running late.
(06:19):
And that's where I want to slowthe whole conversation down.
Ask a different question.
What does AI adoption actuallyask of a leader?
Not the version on thetransformation roadmap, the
Tuesday afternoon version.
It asks them to hold twocontradictory things at the same
(06:41):
time.
This tool might be enormouslyuseful.
This tool might be quietly andconfidently wrong.
It asks them to change a planthey announced three months ago
in front of the same people theyannounced it to.
(07:02):
It asked them to say the fourhardest words in executive
language.
I do not know.
It asked them to notice when amachine's confidence is doing
the persuading rather than themachine's evidence.
It asks them to stay inambiguity for two or three
(07:24):
quarters without collapsing itearly just to feel finished.
Every one of those is expensivethinking, not expensive in
dollars, expensive in what itcosts a human being to do.
Compare that to what aleadership team was asked to do
in normal technology cycle 10years ago.
(07:48):
Pick a vendor, set a timeline,track the rollout.
Those are hard jobs.
They are also familiar jobs withknown shifts and known failure
modes.
This is a different order ofask.
It's not harder in the way abigger project is harder, it's
(08:10):
harder in a different currency.
The old ask spend time andpolitical capital.
The one spends attention,tolerance for not knowing, and
the willingness to be publiclywrong in front of people who
report to you.
Most executive teams havesystems for managing the first
(08:33):
currency.
Almost none have systems formanaging the second.
And here is the part that almostnever makes the slide.
The capacities it requires areprecisely the capacity that gets
quite first when a leadershipteam has been carrying sustained
(08:54):
load.
Let me be careful with the wordhere, because words matter.
And I'm careful about this one.
When I say exhausted, I'm notnaming a condition in any
person.
I'm describing a load state thatleaders describe to be
constantly in their ownlanguage, running on fluids or
(09:17):
running on empty.
A full calendar with no thinkingtime anywhere in it.
Making calls at 7 at night thatshould have been made at 9 in
the morning.
That is strain.
It is a description of load, nota verdict about a person.
And in my experience, straindoes something very specific to
(09:42):
exactly the kind of thinkingthat AI adoption demands.
So what's the cost of that?
It is what actually happensunder that kind of load.
Researchers who study stress andprefrontal function have
described the same pattern formore than 20 years.
(10:04):
The prefrontal cortex is theregion most associated with the
slow, flexible, effortful sideof thinking, holding several
possibilities open at once,weighing an unfamiliar input
against experience, stopping afast answer before it leaves
(10:24):
your mouth, changing directionswhen new information shows up
and makes the old directionwrong.
Under a short burst of pressure,that systems performs
beautifully.
It is built for the spread.
Pressure sharpens it.
(10:45):
But under pressure that issustained, held across weeks and
months with no real recoveryinside it, the picture changes.
The research describes a shiftin weight.
Flexible, effortful thinkingbecomes more costly to produce.
Fast, familiar, habitualresponding becomes cheaper and
(11:07):
more available.
The brain does not break, iteconomizes.
It stops paying full price fordeliberation and starts reaching
for whatever worked last time.
I want to be clear here.
This is not a design flaw.
It's not a story about anyonebeing less capable.
(11:30):
In a genuine emergency, a fast,familiar answer is exactly what
keeps people alive.
The system is doing what it'sevolved to do.
It's answering the question itthinks it's being asked, which
is how we get through this week.
(11:50):
But now, put the same brain inan AI adoption meeting.
There is no familiar answeravailable.
That is the entire point of thetechnology.
Nothing about it resembles whatworked last time because nothing
like this existed last time.
(12:12):
So the pattern I see in myexperience is not leaders
refusing AI.
Almost nobody refuses itanymore.
That fight ended.
The pattern is leaders reachingfor the closest familiar shape
and calling it a decision.
Watch how that looks inside areal organization.
(12:33):
Five patterns.
Buying is a familiar motion witha known process, a known
approval path, and a knownfeeling of completion.
Redesigning how work actuallyflows through company has none
(12:56):
of these things.
So the purchase happens and theredesign gets scheduled for a
quarter that never arrives.
The loudest vendor wins, not thebest fit for the problem.
The one who removed the mostambiguity from the room in 45
minutes.
(13:16):
Under load, certainty is theproduct being bought, and the
software is the deliverymechanism.
They pilot everything and scalenothing.
A pilot is a decision thatpostpones a decision.
In isolation, that is genuineprudence.
(13:38):
Twelve pilots later, across fourdepartments with no scaling path
attached to any of them, it'ssomething else.
It's a leadership team buyingtime, it does not actually have.
They push the judgment down.
The word used in the room isempowerment.
(14:00):
What actually moved down was thediscomfort.
And in the vacuum, people startusing the tools quietly on their
(14:22):
own.
That is what the earnest young52% is describing.
Department level AI running withno approval or oversight.
That is not rebellion.
And I would not treat it as adiscipline issue.
That is a workforce movingfaster than the decision layer
(14:43):
above is able to move.
Now stack those five patternsacross four orders and count
what they actually cost.
You get spend without change.
An organization that has boughtAI and has not become an AI
organization in any way thatshows up in a number anyone
(15:06):
would defend to avoid.
You get the 5% from the Deloittedata, and now you know why it's
5 and not 50%.
You get a winding gap betweenwhat the workforce is already
doing and what the leadershiplayer has approved, which is the
gap every real risk lifts.
(15:28):
Not because people are careless,but because oversight requires a
decision, and decisions requirecapacity.
And you get something muchharder to put on a dashboard.
You get a senior team thatquietly stops believing its own
transformation school.
That belief is not a soft asset.
(15:51):
It's the thing that makeseveryone three levels down pick
up the extra work.
When it goes, nothing announcesit.
The initiative just gets slower,the updates get vaguer, and
nobody can say exactly when itstarted.
(16:14):
I want to name the misreadinghere because this is where the
conversation usually goessideways.
This is not a capabilityproblem, it's a load problem.
And in my experience, the teamsthat end up in this exact
position are usually the oneswho absorb the most for the
longest time with the leastnoise about it.
(16:36):
They carry the reorganization,the hiring freeze, the customer
escalations, and the last threetechnology cycles.
The load is evidence that theywere willing to hold, it is also
the reason the next thing doesnot attend.
(17:01):
So what actually moves this?
I want to say the uncomfortablefirst thing.
First, the sequence mostcompanies are running is
backwards.
They are adding decision toolsto a decision layer that is
already at capacity.
(17:22):
And they are surprised when notools do not stick.
Steady before strategy.
Steady before strategy.
That is the order.
And I mean it operationally, notas appealing.
Three moves, none of them aresoft.
All three belong on the rolloutplan, not on the wellness
(17:44):
calendar.
Move one.
Cut decision volume before youadd decision tools.
Before a single new system goesin, take the standing decision
load of the senior team andshrink it on purpose.
With recurring decisions can bedelegated with a written
threshold instead of a standingmeeting.
(18:07):
Which approvals exist onlybecause of a problem that was
solved in 2023 and neverremoved?
Which meetings are decisiontheater where the decision was
already made and the room existsto distribute the risk.
Most senior teams can find realcapacity here in one afternoon
(18:28):
of honest auditing.
Not because they are wasteful,but because decision load
accumulates the way thataccumulates.
Quietly and in a direction thatonly goes one way unless someone
deliberately reverses it.
One caution on these movesthough.
(18:49):
Do not run the audit as asurvey.
Ask a senior team what they candrop and what they will hand you
the three things they alreadydisliked.
Run it against the calendarinstead.
Take four weeks of the actualschedule, mark every block where
a decision was expected of thesenior team and count.
(19:15):
The number is usually higherthan anyone predicts.
And the surprise is this it'sthe useful part.
People defend a calendar muchless than they defend an
opinion.
This is not a wellnessintervention, it's a capacity
engineering.
(19:35):
You are clearing room for theexpensive thinking the rollout
is about to demand.
Before it demands it.
And this is the first place AIgenuinely earns its seat at this
table.
Not as the things that makes thedecision, but as the thing that
clears the ground before it.
(19:56):
Summarize the reading nobody hadtime to do.
Take the preparation load offthe table so the judgment load
(20:17):
can have the room in place.
That's a real and specific useof the technology, and it's
available in almost everycompany right now, no matter
what you use.
Use chat, plot, copilot, Gemini,perplexy, whatever you use.
It is almost never the use thatgets funded first because it
(20:43):
does not photograph well in aboard tank.
Move number two.
Name one human owner forjudgment, not just develop,
deployment.
One for judgment, notdeployment.
Almost every rollout has anowner for delivery.
Timeline, budget, training,license, utilization, adoption
(21:05):
metrics.
That role is well understood andusually well standard.
Almost none have a named ownerfor the harder question.
Was this output any good?
How would we know?
Those are two different jobs andthey need two different people.
Delivery is a project with anend date.
(21:27):
Judgment is a standingresponsibility with no end date,
but it needs a name attached toit.
Review rhythm, a small amount ofprotected time, and the explicit
authority to say no out loudwithout it being a career event.
(21:48):
The Droid gap is exactly thisgap.
60% are using it.
Managing it.
Managing it is a job.
In most companies, nobody hasbeen given that job, and
everyone assumes someone elsehas it.
(22:11):
Move number three.
Put recovery on the rolloutcalendar as a line item, not as
a benefit, not as a wellnessweek in October, as a scheduling
constraint on the transformationplan itself, written by the same
people who wrote the Meisters.
(22:32):
If the plan asks a leadershipteam of 18 months of continuous
ambiguity with no protectedrecovery inside it, the plan has
a mechanical problem, not amorale problem.
You would not run criticalinfrastructure at 100%
utilization for 18 months andthen call the failure a
(22:55):
surprise.
You would call it a capacityplanning error, and you would
fix it in the planning.
That layer has a utilizationnumber two.
(23:16):
Almost nobody measures it, andalmost everybody spends it.
Now, what actually changes whena leader team is grounded going
into this?
They can hold an unfamiliaroutput long enough to actually
evaluate it, instead of reachingfor the closest familiar shape
(23:40):
and moving on.
They can say, I don't know, yetin a room full of people, which
is the single most usefulsentence in an AI transition
because it keeps the questionopen long enough for a real
answer to arrive.
They can revise in publicwithout it costing them their
(24:03):
standing, because they are notdefending a position they took
while depleted.
They can tell the differencebetween a tool that is generally
useful and a tool that simplyremoves their discomfort for 45
minutes.
In my experience, that singledistinction is worth more than
(24:26):
any vendor evaluation metrics Ihave seen.
And they can let theorganization move fast in the
places where speed is cheap andslow down in the narrow places
where a wrong call is expensive.
That's what a good judgmentunder uncertainty actually looks
like.
It's not caution, it's selectivecaution aimed correctly.
(24:51):
This is the territory my brainframework works, and it's why I
teach steadiness as an operatingrequirement rather than a
personal value.
Now, I can hear some of you veryclearly right now say this is an
engineering and governanceproblem, not a feelings problem.
We need better controls, cleaneraccountability, and a real risk
(25:15):
framework, we do not need calmerexecutives.
And here is my answer, and it'snot an assurance.
You're right that you needcontrols.
My argument is that you will notget them.
Controls are not a document youinstall.
Controls are a series of hardjudgment calls made by entire
(25:37):
person or leader or people aboutunfamiliar systems on ordinary
Tuesdays.
The charter does not make thecall.
A person does.
And the research on sustainedpressure and prefrontal function
says that the exact capacity acontrol decision requires,
holding ambiguity, weighing anunfamiliar input, stopping a
(26:03):
fast answer before it leavesyour mouth is the capacity that
becomes most expensive toproduce under long load.
So the governance work and thecapacity work are not two
projects competing for the samebudget and the same attention.
One is the precondition for theother.
(26:25):
Build the framework, absolutely.
Then ask the second questionnobody asks in that meeting,
which is who specifically willbe executing this?
On what week, on top of whatelse?
You can write the best AIgovernance framework in your
industry.
If the people who have toexecute it are running at full
(26:49):
output with no room, thatframework will be a beautiful
document describing somethingnobody is actually doing.
And it will pass every auditbecause audits read documents.
That is not a feeling argument,that is a mechanism argument.
(27:10):
And in my opinion, it's thereason the 60 and the five sit
so far apart.
In study after study, year afteryear, across industries that
have almost nothing else incommon.
So here is where I would leavethis.
(27:31):
If an AI rollout has falled, thehonest first question is not
which tool, it's not whichvendor, and it's not whether the
wait for the next model.
It is this.
What is the actual decisioncapacity of the people we are
asking to hold this?
And what have we done to protectit?
(27:54):
That question is uncomfortablein a way a vendor comparison
never is.
A vendor comparison is about themarket.
This one is about the rule.
And it's also the one thatchanges the number.
And if the answer is that nobodyhas ever measured it, that's not
a failure.
(28:14):
That is just the first thing onthe list.
And it's a shorter list thanmost transformation plans I have
seen.
I run a live session every monthcalled the Pressure Room for
Executives.
It's free.
It's 45 minutes, and it's wherewe work on exactly this out
loud, together, with one modeland one protocol, you can use
(28:38):
the same afternoon.
The next one is Thursday, August27th at 11 a.m.
Pacific.
Register at saharandy.com slashpressure-room.
I will have the link in thedescription.
That is the only link you needfor this episode, and it's in
the show notes.
Thank you for spending this timewith me.
(29:00):
Leadership does not fail.
Nervous systems too.
Steady before strategy.
As I always say before I leave,show me some love.
Like, subscribe, comment, rateour podcast.
Thank you for your support formaking us one of the top 2%
(29:23):
global podcasts that exist.
I really appreciate you.
Till we meet again on the Fridayshort podcast, the Forbes like
article.
Peace out.
This is Sahar Andradi.
Picture an AI rollout that goes exactly to
plan.
Eighteen months, read budget.
The tools work, the traininghappens, the dashboard go live.
And a year later, almost nothinghas changed about how decisions
actually get made.
(00:35):
Everyone blames the technology.
In my opinion, they are lookingat the wrong thing.
Welcome to the AI CafeConversation Podcast.
I'm Sahar Andradi, and this iswhere the neuroscience of
pressure meets the future ofleadership with an AI twist.
(00:56):
Before we start, the same line Isay every week.
Everything I share comes from myexperience and my opinion as a
coach.
It's not a diagnosis and it'snot a fact about you.
You decide what fits.
Today's question is one I hearin some form almost every week,
(01:17):
in different words fromdifferent rooms.
Why does AI adoption fail incompanies with exhausted
leaders?
Not companies with badtechnology, not companies with
thin budgets, not companies thatwere slow to start.
(01:38):
Companies with good tools, realmoney, smart people, a serious
plan, and a leadership team thathas been running at full output
for three years straight.
I'm going to make a case today.
For the first few minutes, itwill sound like a soft argument.
Stay with me.
(01:59):
The mechanism underneath it'snot soft at all.
It's one of the most establishedfindings in the study of
pressure and human thinking.
And almost nobody puts it on theAI strategy slide.
And I want to say why thismatters to me specifically.
I spent years in medicine beforeI did this work.
(02:21):
What I carried out of thattraining was not a set of
diagnoses, it was a habit.
When something keeps failing inthe same shape across very
different environments, stopblaming the environment.
Go look at the mechanism thatall of them share.
Right now, AI adoption isfailing in the same shape across
(02:45):
very different companies.
So let's go look at themechanism they share.
Let's talk about the crisis.
Let me start with where weactually are, because the
numbers are stranger than theheadlines suggest.
Deloitte's 2026 Global HumanCapital Trends Report surveyed
(03:10):
more than 9,000 business andhuman resource leaders across 89
countries.
60% of executives said they arealready using AI in decision
making.
Only 5% said they manage itwell.
Sit with those two numbers sideby side, 60 and 5.
(03:35):
That is not an adoption problem.
Adoption already happened.
That is a 55-point gap betweenusing a thing and being able to
hold it.
Then look at what Ernest andYoung published in early 2026.
The US technology passportconducted in February surveyed
(03:59):
500 business leaders workinginside the technology industry.
These are the people closest tothe tools, the ones with the
most fluency, the most budget,and the fewest excuses.
78% said AI adoption isoutpacing the organization's
(04:20):
ability to effectively managethe business.
More than half, 52% saiddepartment level AI work is
running with no formal approvalor oversight at all.
I want to be precise about thatsecond study because precision
is the whole job.
(04:41):
That is technology sectorleaders in the United States.
It's not every industry in everycountry, and I'm not going to
stretch it into something it didnot measure.
But if the people who build thetools are telling you the pace
has outrun the ability tomanage, that is a signal worth
(05:04):
taking seriously.
Now, there is a standard readingof these numbers, and you have
probably heard it in your ownbuilding.
It's a governance reading.
The story goes like this: companies moved too fast, they (05:16):
undefined
skipped the policy work, theyneed better frameworks, tighter
controls, a committee, acharter, a risk register, a
center of excellence.
That story is not wrong.
(05:36):
It's incomplete.
Because governance is not adocument.
Governance is a series ofdecisions made by specific human
beings on ordinary weekdaysunder time pressure about things
they have never seen before.
(05:57):
A framework does not evaluate amodel output.
A person does.
A charter does not decidewhether to override a
recommendation that contradicts20 years of experience.
A person does.
In about 90 seconds in a meetingthat is already running late.
(06:19):
And that's where I want to slowthe whole conversation down.
Ask a different question.
What does AI adoption actuallyask of a leader?
Not the version on thetransformation roadmap, the
Tuesday afternoon version.
It asks them to hold twocontradictory things at the same
(06:41):
time.
This tool might be enormouslyuseful.
This tool might be quietly andconfidently wrong.
It asks them to change a planthey announced three months ago
in front of the same people theyannounced it to.
(07:02):
It asked them to say the fourhardest words in executive
language.
I do not know.
It asked them to notice when amachine's confidence is doing
the persuading rather than themachine's evidence.
It asks them to stay inambiguity for two or three
(07:24):
quarters without collapsing itearly just to feel finished.
Every one of those is expensivethinking, not expensive in
dollars, expensive in what itcosts a human being to do.
Compare that to what aleadership team was asked to do
in normal technology cycle 10years ago.
(07:48):
Pick a vendor, set a timeline,track the rollout.
Those are hard jobs.
They are also familiar jobs withknown shifts and known failure
modes.
This is a different order ofask.
It's not harder in the way abigger project is harder, it's
(08:10):
harder in a different currency.
The old ask spend time andpolitical capital.
The one spends attention,tolerance for not knowing, and
the willingness to be publiclywrong in front of people who
report to you.
Most executive teams havesystems for managing the first
(08:33):
currency.
Almost none have systems formanaging the second.
And here is the part that almostnever makes the slide.
The capacities it requires areprecisely the capacity that gets
quite first when a leadershipteam has been carrying sustained
(08:54):
load.
Let me be careful with the wordhere, because words matter.
And I'm careful about this one.
When I say exhausted, I'm notnaming a condition in any
person.
I'm describing a load state thatleaders describe to be
constantly in their ownlanguage, running on fluids or
(09:17):
running on empty.
A full calendar with no thinkingtime anywhere in it.
Making calls at 7 at night thatshould have been made at 9 in
the morning.
That is strain.
It is a description of load, nota verdict about a person.
And in my experience, straindoes something very specific to
(09:42):
exactly the kind of thinkingthat AI adoption demands.
So what's the cost of that?
It is what actually happensunder that kind of load.
Researchers who study stress andprefrontal function have
described the same pattern formore than 20 years.
(10:04):
The prefrontal cortex is theregion most associated with the
slow, flexible, effortful sideof thinking, holding several
possibilities open at once,weighing an unfamiliar input
against experience, stopping afast answer before it leaves
(10:24):
your mouth, changing directionswhen new information shows up
and makes the old directionwrong.
Under a short burst of pressure,that systems performs
beautifully.
It is built for the spread.
Pressure sharpens it.
(10:45):
But under pressure that issustained, held across weeks and
months with no real recoveryinside it, the picture changes.
The research describes a shiftin weight.
Flexible, effortful thinkingbecomes more costly to produce.
Fast, familiar, habitualresponding becomes cheaper and
(11:07):
more available.
The brain does not break, iteconomizes.
It stops paying full price fordeliberation and starts reaching
for whatever worked last time.
I want to be clear here.
This is not a design flaw.
It's not a story about anyonebeing less capable.
(11:30):
In a genuine emergency, a fast,familiar answer is exactly what
keeps people alive.
The system is doing what it'sevolved to do.
It's answering the question itthinks it's being asked, which
is how we get through this week.
(11:50):
But now, put the same brain inan AI adoption meeting.
There is no familiar answeravailable.
That is the entire point of thetechnology.
Nothing about it resembles whatworked last time because nothing
like this existed last time.
(12:12):
So the pattern I see in myexperience is not leaders
refusing AI.
Almost nobody refuses itanymore.
That fight ended.
The pattern is leaders reachingfor the closest familiar shape
and calling it a decision.
Watch how that looks inside areal organization.
(12:33):
Five patterns.
Buying is a familiar motion witha known process, a known
approval path, and a knownfeeling of completion.
Redesigning how work actuallyflows through company has none
(12:56):
of these things.
So the purchase happens and theredesign gets scheduled for a
quarter that never arrives.
The loudest vendor wins, not thebest fit for the problem.
The one who removed the mostambiguity from the room in 45
minutes.
(13:16):
Under load, certainty is theproduct being bought, and the
software is the deliverymechanism.
They pilot everything and scalenothing.
A pilot is a decision thatpostpones a decision.
In isolation, that is genuineprudence.
(13:38):
Twelve pilots later, across fourdepartments with no scaling path
attached to any of them, it'ssomething else.
It's a leadership team buyingtime, it does not actually have.
They push the judgment down.
The word used in the room isempowerment.
(14:00):
What actually moved down was thediscomfort.
And in the vacuum, people startusing the tools quietly on their
(14:22):
own.
That is what the earnest young52% is describing.
Department level AI running withno approval or oversight.
That is not rebellion.
And I would not treat it as adiscipline issue.
That is a workforce movingfaster than the decision layer
(14:43):
above is able to move.
Now stack those five patternsacross four orders and count
what they actually cost.
You get spend without change.
An organization that has boughtAI and has not become an AI
organization in any way thatshows up in a number anyone
(15:06):
would defend to avoid.
You get the 5% from the Deloittedata, and now you know why it's
5 and not 50%.
You get a winding gap betweenwhat the workforce is already
doing and what the leadershiplayer has approved, which is the
gap every real risk lifts.
(15:28):
Not because people are careless,but because oversight requires a
decision, and decisions requirecapacity.
And you get something muchharder to put on a dashboard.
You get a senior team thatquietly stops believing its own
transformation school.
That belief is not a soft asset.
(15:51):
It's the thing that makeseveryone three levels down pick
up the extra work.
When it goes, nothing announcesit.
The initiative just gets slower,the updates get vaguer, and
nobody can say exactly when itstarted.
(16:14):
I want to name the misreadinghere because this is where the
conversation usually goessideways.
This is not a capabilityproblem, it's a load problem.
And in my experience, the teamsthat end up in this exact
position are usually the oneswho absorb the most for the
longest time with the leastnoise about it.
(16:36):
They carry the reorganization,the hiring freeze, the customer
escalations, and the last threetechnology cycles.
The load is evidence that theywere willing to hold, it is also
the reason the next thing doesnot attend.
(17:01):
So what actually moves this?
I want to say the uncomfortablefirst thing.
First, the sequence mostcompanies are running is
backwards.
They are adding decision toolsto a decision layer that is
already at capacity.
(17:22):
And they are surprised when notools do not stick.
Steady before strategy.
Steady before strategy.
That is the order.
And I mean it operationally, notas appealing.
Three moves, none of them aresoft.
All three belong on the rolloutplan, not on the wellness
(17:44):
calendar.
Move one.
Cut decision volume before youadd decision tools.
Before a single new system goesin, take the standing decision
load of the senior team andshrink it on purpose.
With recurring decisions can bedelegated with a written
threshold instead of a standingmeeting.
(18:07):
Which approvals exist onlybecause of a problem that was
solved in 2023 and neverremoved?
Which meetings are decisiontheater where the decision was
already made and the room existsto distribute the risk.
Most senior teams can find realcapacity here in one afternoon
(18:28):
of honest auditing.
Not because they are wasteful,but because decision load
accumulates the way thataccumulates.
Quietly and in a direction thatonly goes one way unless someone
deliberately reverses it.
One caution on these movesthough.
(18:49):
Do not run the audit as asurvey.
Ask a senior team what they candrop and what they will hand you
the three things they alreadydisliked.
Run it against the calendarinstead.
Take four weeks of the actualschedule, mark every block where
a decision was expected of thesenior team and count.
(19:15):
The number is usually higherthan anyone predicts.
And the surprise is this it'sthe useful part.
People defend a calendar muchless than they defend an
opinion.
This is not a wellnessintervention, it's a capacity
engineering.
(19:35):
You are clearing room for theexpensive thinking the rollout
is about to demand.
Before it demands it.
And this is the first place AIgenuinely earns its seat at this
table.
Not as the things that makes thedecision, but as the thing that
clears the ground before it.
(19:56):
Summarize the reading nobody hadtime to do.
Take the preparation load offthe table so the judgment load
(20:17):
can have the room in place.
That's a real and specific useof the technology, and it's
available in almost everycompany right now, no matter
what you use.
Use chat, plot, copilot, Gemini,perplexy, whatever you use.
It is almost never the use thatgets funded first because it
(20:43):
does not photograph well in aboard tank.
Move number two.
Name one human owner forjudgment, not just develop,
deployment.
One for judgment, notdeployment.
Almost every rollout has anowner for delivery.
Timeline, budget, training,license, utilization, adoption
(21:05):
metrics.
That role is well understood andusually well standard.
Almost none have a named ownerfor the harder question.
Was this output any good?
How would we know?
Those are two different jobs andthey need two different people.
Delivery is a project with anend date.
(21:27):
Judgment is a standingresponsibility with no end date,
but it needs a name attached toit.
Review rhythm, a small amount ofprotected time, and the explicit
authority to say no out loudwithout it being a career event.
(21:48):
The Droid gap is exactly thisgap.
60% are using it.
Managing it.
Managing it is a job.
In most companies, nobody hasbeen given that job, and
everyone assumes someone elsehas it.
(22:11):
Move number three.
Put recovery on the rolloutcalendar as a line item, not as
a benefit, not as a wellnessweek in October, as a scheduling
constraint on the transformationplan itself, written by the same
people who wrote the Meisters.
(22:32):
If the plan asks a leadershipteam of 18 months of continuous
ambiguity with no protectedrecovery inside it, the plan has
a mechanical problem, not amorale problem.
You would not run criticalinfrastructure at 100%
utilization for 18 months andthen call the failure a
(22:55):
surprise.
You would call it a capacityplanning error, and you would
fix it in the planning.
That layer has a utilizationnumber two.
(23:16):
Almost nobody measures it, andalmost everybody spends it.
Now, what actually changes whena leader team is grounded going
into this?
They can hold an unfamiliaroutput long enough to actually
evaluate it, instead of reachingfor the closest familiar shape
(23:40):
and moving on.
They can say, I don't know, yetin a room full of people, which
is the single most usefulsentence in an AI transition
because it keeps the questionopen long enough for a real
answer to arrive.
They can revise in publicwithout it costing them their
(24:03):
standing, because they are notdefending a position they took
while depleted.
They can tell the differencebetween a tool that is generally
useful and a tool that simplyremoves their discomfort for 45
minutes.
In my experience, that singledistinction is worth more than
(24:26):
any vendor evaluation metrics Ihave seen.
And they can let theorganization move fast in the
places where speed is cheap andslow down in the narrow places
where a wrong call is expensive.
That's what a good judgmentunder uncertainty actually looks
like.
It's not caution, it's selectivecaution aimed correctly.
(24:51):
This is the territory my brainframework works, and it's why I
teach steadiness as an operatingrequirement rather than a
personal value.
Now, I can hear some of you veryclearly right now say this is an
engineering and governanceproblem, not a feelings problem.
We need better controls, cleaneraccountability, and a real risk
(25:15):
framework, we do not need calmerexecutives.
And here is my answer, and it'snot an assurance.
You're right that you needcontrols.
My argument is that you will notget them.
Controls are not a document youinstall.
Controls are a series of hardjudgment calls made by entire
(25:37):
person or leader or people aboutunfamiliar systems on ordinary
Tuesdays.
The charter does not make thecall.
A person does.
And the research on sustainedpressure and prefrontal function
says that the exact capacity acontrol decision requires,
holding ambiguity, weighing anunfamiliar input, stopping a
(26:03):
fast answer before it leavesyour mouth is the capacity that
becomes most expensive toproduce under long load.
So the governance work and thecapacity work are not two
projects competing for the samebudget and the same attention.
One is the precondition for theother.
(26:25):
Build the framework, absolutely.
Then ask the second questionnobody asks in that meeting,
which is who specifically willbe executing this?
On what week, on top of whatelse?
You can write the best AIgovernance framework in your
industry.
If the people who have toexecute it are running at full
(26:49):
output with no room, thatframework will be a beautiful
document describing somethingnobody is actually doing.
And it will pass every auditbecause audits read documents.
That is not a feeling argument,that is a mechanism argument.
(27:10):
And in my opinion, it's thereason the 60 and the five sit
so far apart.
In study after study, year afteryear, across industries that
have almost nothing else incommon.
So here is where I would leavethis.
(27:31):
If an AI rollout has falled, thehonest first question is not
which tool, it's not whichvendor, and it's not whether the
wait for the next model.
It is this.
What is the actual decisioncapacity of the people we are
asking to hold this?
And what have we done to protectit?
(27:54):
That question is uncomfortablein a way a vendor comparison
never is.
A vendor comparison is about themarket.
This one is about the rule.
And it's also the one thatchanges the number.
And if the answer is that nobodyhas ever measured it, that's not
a failure.
(28:14):
That is just the first thing onthe list.
And it's a shorter list thanmost transformation plans I have
seen.
I run a live session every monthcalled the Pressure Room for
Executives.
It's free.
It's 45 minutes, and it's wherewe work on exactly this out
loud, together, with one modeland one protocol, you can use
(28:38):
the same afternoon.
The next one is Thursday, August27th at 11 a.m.
Pacific.
Register at saharandy.com slashpressure-room.
I will have the link in thedescription.
That is the only link you needfor this episode, and it's in
the show notes.
Thank you for spending this timewith me.
(29:00):
Leadership does not fail.
Nervous systems too.
Steady before strategy.
As I always say before I leave,show me some love.
Like, subscribe, comment, rateour podcast.
Thank you for your support formaking us one of the top 2%
(29:23):
global podcasts that exist.
I really appreciate you.
Till we meet again on the Fridayshort podcast, the Forbes like
article.
Peace out.
This is Sahar Andradi.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Betrayal Weekly
Betrayal Weekly is back for a new season. Every Thursday, Betrayal Weekly shares first-hand accounts of broken trust, shocking deceptions, and the trail of destruction they leave behind. Hosted by Andrea Gunning, this weekly ongoing series digs into real-life stories of betrayal and the aftermath. From stories of double lives to dark discoveries, these are cautionary tales and accounts of resilience against all odds. From the producers of the critically acclaimed Betrayal series, Betrayal Weekly drops new episodes every Thursday. If you would like to share your story, you can reach out to the Betrayal Team by emailing them at betrayalpod@gmail.com and follow us on Instagram at @betrayalpod and @glasspodcasts. Please join our Substack for additional exclusive content, curated book recommendations, and community discussions. Sign up FREE by clicking this link Beyond Betrayal Substack. Join our community dedicated to truth, resilience, and healing. Your voice matters! Be a part of our Betrayal journey on Substack.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com