August 11, 2025 • 15 mins
Is Artificial Intelligence a threat to your career, or is it the most powerful tool you'll ever have? In this episode, we cut through the hype and get to the facts by breaking down a groundbreaking new study from Microsoft Research that analysed over 200,000 real-world AI interactions.
Join us as we translate the academic findings into a practical guide for today's professional. You will learn:
- The real story behind AI's role in the workplace—is it a replacement or a collaborator?
- Which specific jobs and tasks are being impacted the most right now.
- Real-world examples of how AI is being used to draft project plans, summarise meetings, and accelerate analysis.
- The "Mindset, Skillset, and Toolset" framework: a clear, actionable plan to not just survive, but thrive in the age of AI.
If you want to move past the headlines and build a future-proof career, this episode is your essential starting point.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
SPEAKER_00 (00:00):
Is artificial intelligence coming for your
job?
It's one of the most significantand persistent questions in the
professional world right now.
It's a topic of discussion inevery industry, from finance to
(00:23):
marketing and on the mind ofevery forward-thinking
professional.
We've been inundated withpredictions and headlines, some
optimistic, some alarming.
But what's actually happeningtoday?
Not in a lab or in a think tank,but in the real world of work.
(00:51):
Welcome to the show.
Today we're moving past thespeculation and looking at the
data.
A groundbreaking new study fromMicrosoft Research titled
Working with AI gives us one ofthe clearest pictures yet of how
this technology is activelychanging our jobs.
And as someone who uses andtrains people on Microsoft
(01:15):
Copilot every single day, thefindings didn't just resonate
with me.
They validated what I've beenexperiencing firsthand.
So stick around.
Over the next 15 minutes, We'regoing to break down what AI is
really doing at work, who isbeing most affected, and most
importantly, how you can build astrategy to thrive in this new
(01:38):
era of collaborativeintelligence.
Now let's get into the details.
The first thing to understandabout this study is its
foundation in reality.
This isn't based on forecasts orwhat-if scenarios.
The researchers analysed over200,000 real conversations
between users and a generativeAI system.
(02:02):
This is workplace anthropologyfor the digital age, showing us
what people are actually usingthis technology for day in and
day out.
The most significant takeaway isthis.
The study found that the jobsmost impacted by AI are what
they call knowledge work.
If your job revolves aroundinformation, ideas and
(02:25):
communication, if you are paidto analyze, strategize and
create, then you're at theforefront of this
transformation.
This is where my own experiencelines up perfectly.
My work is all about takinglarge amounts of information,
making sense of it, and thencommunicating that analysis in
(02:47):
reports and project plans.
In this context, Copilot hasbecome an indispensable research
partner.
The study identified threeprimary categories of AI use,
which I see as the three pillarsof modern knowledge work.
The first being gathering andprocessing information.
(03:10):
Secondly, writing and editingcontent.
And thirdly, communicating withothers.
So let me provide a concreteexample.
Say I was brought into a projectthat had been in progress for
three months.
The traditional way to get up tospeed would be me being involved
(03:33):
in spending hours, maybe even afull day, digging through long
email chains, searching fordocuments in different folders
and systems and trying to piecetogether the key decisions and
action items.
It's inefficient and is prone toerror.
The new way, I asked Copilot tosummarize the entire email
(03:55):
thread and list the keydecisions and outstanding action
items.
In less than a minute, I had aclean bulleted list.
This is a perfect example ofgathering and processing
information.
It's not just a time saver, it'sa powerful accelerator for
productivity and clarity.
(04:17):
It allowed me to enter my firstproject meeting fully informed
and ready to contribute ratherthan a week behind.
This leads to one of the mostfascinating parts of the study.
And this is where the newparadigm of work becomes clear.
(04:40):
The researchers make adistinction between the user
goal and the AI action.
The user goal is what you aretrying to accomplish.
The AI action is the specifictask the AI is performing to
help you.
And here's the critical insight.
(05:02):
In 40% of the conversations theyanalyzed, the user's goal and
the AI's action were completelydifferent.
This isn't simple automationwhere a machine just does your
task for you.
This is collaboration.
It's a dynamic partnershipbetween the human professional
(05:22):
and the AI tool.
The human holds the strategicintent while the AI executes a
component task.
Let me give you another realworld example.
A significant part of my roleinvolves setting up new
SharePoint sites where acritical step is creating a
(05:44):
robust metadata structure.
which are the tags and labelsthat make content findable and
governable.
My user goal is to build ametadata schema.
Before AI, this was a manual,time-consuming process of
reviewing hundreds of documentsto identify common themes.
(06:06):
Now, my process is different.
I can provide Copilot with a setof sample documents, and my
prompt is...
assess these documents andpropose a metadata schema based
on their content.
Copilot's AI action is toanalyze, identify patterns, and
(06:27):
provide a structuredrecommendation.
It might come back withsuggestions for tags like
project name, status,department, and document type.
I, as the human expert, thentake that recommendation,
validate it, refine it, andimplement it.
I am still directing the workand making the final decisions,
(06:50):
but the AI has performed theheavy lifting of the initial
analysis.
I am the architect, the AI isthe surveyor.
That is the collaborative modelthe study so clearly identifies.
It's a loop.
Human strategy informs AI actionand AI output informs human
(07:12):
refinement.
So now, If knowledge workers arethe most affected, which roles
are least affected?
Unsurprisingly, the study foundthat jobs requiring physical
labour, operating machinery orproviding hands-on services have
the lowest AI applicabilityscores at this time.
(07:36):
Roles like roofers, truckdrivers and nursing assistants
aren't the primary focus of thiswave of generative AI.
But for those of us whose workis centred on information, the
impact is undeniable.
The study lists the jobs withthe highest applicability
scores.
At the very top are interpretersand translators.
(07:59):
Following closely are roles likesales representatives, customer
service agents, writers andvarious administrative clerks.
The common thread among allthese roles is a heavy reliance
on communication and informationmanagement.
I see this in my projectmanagement work every day.
(08:20):
I used to spend a significantamount of time creating the
first draft of a project planfrom a blank page.
Now I can give Copilot a clearprompt, such as, we need to
launch a new employee onboardingportal by quarter four.
(08:40):
The stakeholders are HR, IT andcommunications.
Generate a detailed project planwith key phases, deliverables
and estimated timelines.
It will produce a comprehensivestarting point.
It's never a final product, butit gets me about 70% of the way
(09:02):
there in a matter of seconds.
My professional value is thenapplied to refining that last
30%.
to asking the smart questions,to anticipating the roadblocks
that AI can't see, and toapplying the nuanced context of
our organization and the teamdynamics.
(09:25):
The AI provides the structure, Iprovide the strategic wisdom.
This is a fundamental shift inwhere we create value.
Our expertise becomes less aboutthe creation of the initial
draft and more about thecuration and refinement of the
AI's output.
(09:56):
So now let's take a moment toexplore the deeper implications
here.
The study shows that AI is mostsuccessful when assisting with
writing and research.
but less so with tasks requiringdeep data analysis or visual
design.
This tells us somethingimportant about the current
(10:18):
state of the technology.
It's a language and informationprocessor at its core.
This means we need to becomeexperts at knowing when to use
AI and for what purpose.
Trying to use it for the wrongtask is like using a hammer to
turn a screw.
(10:39):
It's inefficient and you won'tget the right results.
The skill is in the discernment.
So what does this mean for youand your career?
How do you move from being apassive observer to an active
participant in this shift?
I believe it comes down to athree-part strategy.
(11:02):
Mindset, skill set and tool set.
So let's take a look at this.
First, mindset.
It's time to stop seeing AI as athreat and start seeing it as a
powerful tool for augmentation.
Think of it as a new juniormember of your team, one that is
(11:22):
incredibly fast, has access tovast information, and can
execute repetitive tasksflawlessly.
Your role is to be the manager,the director, the strategist,
This mindset shift from fear tocuriosity is the essential first
step.
(11:43):
Second is skill set.
The skills that are becomingmost valuable are not the ones
that can be automated.
They are the uniquely humanskills that AI complements.
These include critical thinking,the ability to evaluate the AI's
output, spot biases and andidentify errors.
(12:08):
Strategic questioning.
Knowing how to ask the rightquestions, which is what we now
call prompt engineering, to getthe most valuable output from
the AI.
Creative synthesis.
Taking the AI-generated contentand combining it with your own
ideas and insights to createsomething new and valuable.
(12:32):
Emotional intelligence.
The ability to communicate,collaborate and lead teams.
Skills that remain firmly in thehuman domain.
Those are the skill sets thatare becoming the most valuable.
And thirdly, tool set.
This means getting hands-on.
(12:53):
You cannot learn this by readingarticles alone.
You have to actively use thetools.
Start small.
Ask it to rephrase an email orsuggest a few headlines for a
presentation.
Then move to more complex tasks.
Use it to create the first draftof that report you need to
(13:14):
write.
Overcoming the blank page isoften the hardest part.
Ask it to summarize longmeetings and extract the key
action items.
This improves accountability andfollow through.
Also have it analyze data setsand identify patterns or
anomalies you might have missed.
(13:43):
The study from MicrosoftResearch confirms what many of
us are now seeing on the ground.
AI is not leading to a worldwithout jobs for knowledge
workers.
It's leading to a world where wecan offload the tedious and
repetitive parts of our work tofocus on what we as humans do
best.
(14:04):
Creativity, strategic thinking,complex problem solving and
building relationships.
The professionals who will excelin the coming years are the ones
who master this new way ofworking.
They will be the ones who learnto collaborate effectively with
their AI partners.
You've already taken animportant step by seeking to
(14:26):
understand this topic moredeeply.
The next step Thank you fortuning in.
Join us next time as we continueto explore the intersection of
(14:48):
technology and the future ofwork.
UNKNOWN (14:53):
Music
Is artificial intelligence coming for your
job?
It's one of the most significantand persistent questions in the
professional world right now.
It's a topic of discussion inevery industry, from finance to
(00:23):
marketing and on the mind ofevery forward-thinking
professional.
We've been inundated withpredictions and headlines, some
optimistic, some alarming.
But what's actually happeningtoday?
Not in a lab or in a think tank,but in the real world of work.
(00:51):
Welcome to the show.
Today we're moving past thespeculation and looking at the
data.
A groundbreaking new study fromMicrosoft Research titled
Working with AI gives us one ofthe clearest pictures yet of how
this technology is activelychanging our jobs.
And as someone who uses andtrains people on Microsoft
(01:15):
Copilot every single day, thefindings didn't just resonate
with me.
They validated what I've beenexperiencing firsthand.
So stick around.
Over the next 15 minutes, We'regoing to break down what AI is
really doing at work, who isbeing most affected, and most
importantly, how you can build astrategy to thrive in this new
(01:38):
era of collaborativeintelligence.
Now let's get into the details.
The first thing to understandabout this study is its
foundation in reality.
This isn't based on forecasts orwhat-if scenarios.
The researchers analysed over200,000 real conversations
between users and a generativeAI system.
(02:02):
This is workplace anthropologyfor the digital age, showing us
what people are actually usingthis technology for day in and
day out.
The most significant takeaway isthis.
The study found that the jobsmost impacted by AI are what
they call knowledge work.
If your job revolves aroundinformation, ideas and
(02:25):
communication, if you are paidto analyze, strategize and
create, then you're at theforefront of this
transformation.
This is where my own experiencelines up perfectly.
My work is all about takinglarge amounts of information,
making sense of it, and thencommunicating that analysis in
(02:47):
reports and project plans.
In this context, Copilot hasbecome an indispensable research
partner.
The study identified threeprimary categories of AI use,
which I see as the three pillarsof modern knowledge work.
The first being gathering andprocessing information.
(03:10):
Secondly, writing and editingcontent.
And thirdly, communicating withothers.
So let me provide a concreteexample.
Say I was brought into a projectthat had been in progress for
three months.
The traditional way to get up tospeed would be me being involved
(03:33):
in spending hours, maybe even afull day, digging through long
email chains, searching fordocuments in different folders
and systems and trying to piecetogether the key decisions and
action items.
It's inefficient and is prone toerror.
The new way, I asked Copilot tosummarize the entire email
(03:55):
thread and list the keydecisions and outstanding action
items.
In less than a minute, I had aclean bulleted list.
This is a perfect example ofgathering and processing
information.
It's not just a time saver, it'sa powerful accelerator for
productivity and clarity.
(04:17):
It allowed me to enter my firstproject meeting fully informed
and ready to contribute ratherthan a week behind.
This leads to one of the mostfascinating parts of the study.
And this is where the newparadigm of work becomes clear.
(04:40):
The researchers make adistinction between the user
goal and the AI action.
The user goal is what you aretrying to accomplish.
The AI action is the specifictask the AI is performing to
help you.
And here's the critical insight.
(05:02):
In 40% of the conversations theyanalyzed, the user's goal and
the AI's action were completelydifferent.
This isn't simple automationwhere a machine just does your
task for you.
This is collaboration.
It's a dynamic partnershipbetween the human professional
(05:22):
and the AI tool.
The human holds the strategicintent while the AI executes a
component task.
Let me give you another realworld example.
A significant part of my roleinvolves setting up new
SharePoint sites where acritical step is creating a
(05:44):
robust metadata structure.
which are the tags and labelsthat make content findable and
governable.
My user goal is to build ametadata schema.
Before AI, this was a manual,time-consuming process of
reviewing hundreds of documentsto identify common themes.
(06:06):
Now, my process is different.
I can provide Copilot with a setof sample documents, and my
prompt is...
assess these documents andpropose a metadata schema based
on their content.
Copilot's AI action is toanalyze, identify patterns, and
(06:27):
provide a structuredrecommendation.
It might come back withsuggestions for tags like
project name, status,department, and document type.
I, as the human expert, thentake that recommendation,
validate it, refine it, andimplement it.
I am still directing the workand making the final decisions,
(06:50):
but the AI has performed theheavy lifting of the initial
analysis.
I am the architect, the AI isthe surveyor.
That is the collaborative modelthe study so clearly identifies.
It's a loop.
Human strategy informs AI actionand AI output informs human
(07:12):
refinement.
So now, If knowledge workers arethe most affected, which roles
are least affected?
Unsurprisingly, the study foundthat jobs requiring physical
labour, operating machinery orproviding hands-on services have
the lowest AI applicabilityscores at this time.
(07:36):
Roles like roofers, truckdrivers and nursing assistants
aren't the primary focus of thiswave of generative AI.
But for those of us whose workis centred on information, the
impact is undeniable.
The study lists the jobs withthe highest applicability
scores.
At the very top are interpretersand translators.
(07:59):
Following closely are roles likesales representatives, customer
service agents, writers andvarious administrative clerks.
The common thread among allthese roles is a heavy reliance
on communication and informationmanagement.
I see this in my projectmanagement work every day.
(08:20):
I used to spend a significantamount of time creating the
first draft of a project planfrom a blank page.
Now I can give Copilot a clearprompt, such as, we need to
launch a new employee onboardingportal by quarter four.
(08:40):
The stakeholders are HR, IT andcommunications.
Generate a detailed project planwith key phases, deliverables
and estimated timelines.
It will produce a comprehensivestarting point.
It's never a final product, butit gets me about 70% of the way
(09:02):
there in a matter of seconds.
My professional value is thenapplied to refining that last
30%.
to asking the smart questions,to anticipating the roadblocks
that AI can't see, and toapplying the nuanced context of
our organization and the teamdynamics.
(09:25):
The AI provides the structure, Iprovide the strategic wisdom.
This is a fundamental shift inwhere we create value.
Our expertise becomes less aboutthe creation of the initial
draft and more about thecuration and refinement of the
AI's output.
(09:56):
So now let's take a moment toexplore the deeper implications
here.
The study shows that AI is mostsuccessful when assisting with
writing and research.
but less so with tasks requiringdeep data analysis or visual
design.
This tells us somethingimportant about the current
(10:18):
state of the technology.
It's a language and informationprocessor at its core.
This means we need to becomeexperts at knowing when to use
AI and for what purpose.
Trying to use it for the wrongtask is like using a hammer to
turn a screw.
(10:39):
It's inefficient and you won'tget the right results.
The skill is in the discernment.
So what does this mean for youand your career?
How do you move from being apassive observer to an active
participant in this shift?
I believe it comes down to athree-part strategy.
(11:02):
Mindset, skill set and tool set.
So let's take a look at this.
First, mindset.
It's time to stop seeing AI as athreat and start seeing it as a
powerful tool for augmentation.
Think of it as a new juniormember of your team, one that is
(11:22):
incredibly fast, has access tovast information, and can
execute repetitive tasksflawlessly.
Your role is to be the manager,the director, the strategist,
This mindset shift from fear tocuriosity is the essential first
step.
(11:43):
Second is skill set.
The skills that are becomingmost valuable are not the ones
that can be automated.
They are the uniquely humanskills that AI complements.
These include critical thinking,the ability to evaluate the AI's
output, spot biases and andidentify errors.
(12:08):
Strategic questioning.
Knowing how to ask the rightquestions, which is what we now
call prompt engineering, to getthe most valuable output from
the AI.
Creative synthesis.
Taking the AI-generated contentand combining it with your own
ideas and insights to createsomething new and valuable.
(12:32):
Emotional intelligence.
The ability to communicate,collaborate and lead teams.
Skills that remain firmly in thehuman domain.
Those are the skill sets thatare becoming the most valuable.
And thirdly, tool set.
This means getting hands-on.
(12:53):
You cannot learn this by readingarticles alone.
You have to actively use thetools.
Start small.
Ask it to rephrase an email orsuggest a few headlines for a
presentation.
Then move to more complex tasks.
Use it to create the first draftof that report you need to
(13:14):
write.
Overcoming the blank page isoften the hardest part.
Ask it to summarize longmeetings and extract the key
action items.
This improves accountability andfollow through.
Also have it analyze data setsand identify patterns or
anomalies you might have missed.
(13:43):
The study from MicrosoftResearch confirms what many of
us are now seeing on the ground.
AI is not leading to a worldwithout jobs for knowledge
workers.
It's leading to a world where wecan offload the tedious and
repetitive parts of our work tofocus on what we as humans do
best.
(14:04):
Creativity, strategic thinking,complex problem solving and
building relationships.
The professionals who will excelin the coming years are the ones
who master this new way ofworking.
They will be the ones who learnto collaborate effectively with
their AI partners.
You've already taken animportant step by seeking to
(14:26):
understand this topic moredeeply.
The next step Thank you fortuning in.
Join us next time as we continueto explore the intersection of
(14:48):
technology and the future ofwork.
UNKNOWN (14:53):
Music
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.
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
The Breakfast Club
The World's Most Dangerous Morning Show, The Breakfast Club, With DJ Envy, Jess Hilarious, And Charlamagne Tha God!