October 7, 2025 • 27 mins
AI was supposed to make us more productive, but something's gone terribly wrong. While AI usage at work has doubled since 2023, a stunning 95% of organizations see zero measurable return on their investment. The culprit? "Workslop" - polished-looking AI-generated content that lacks real substance and creates more work for everyone downstream. Workers are spending nearly two hours dealing with each workslop incident they receive, costing companies millions in lost productivity and something even more valuable: team trust.
In this episode, we break down the workslop phenomenon and why it's particularly dangerous for startups. We explore how AI has turbocharged an old problem (bad PowerPoints and rambling emails have always existed), examine real examples of workslop in the wild, and discuss why recipients view workslop senders as less creative, capable, and trustworthy. More importantly, we give entrepreneurs practical strategies to prevent their teams from falling into the workslop trap - because the difference between AI that augments your team and AI that destroys it comes down to how you implement it.
Whether you're considering AI tools for your startup or already knee-deep in implementation, this episode will help you avoid the $9 million mistake that's tearing teams apart. We'll show you how to spot workslop, build team commitments around quality over speed, and create a culture where AI enhances human judgment rather than replacing it with hollow busywork.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
It's the emotional cost ohi likeOK so 53% of recipients of work
slot feel annoyed OK Yep 38% feel confused 22% feel offended
OHP cool somebody said that and he.
Was super pissed. Next time you send me like an AI
generated e-mail, yeah. Let's get it rolling big ideas
(00:23):
money folding hustle smart dream.
So during that grinding through a joy.
Ride. So today we are talking about
actually this came from an HBR article that I read and I was
just like oh once again they nailed it.
(00:43):
OK, the. Other big supporter of.
HBR Well, yeah, they're kind of bright.
Yeah, they're somewhat researched, which is.
Nice. Yeah, they seem to have data and
those types of things. So here's the concept.
So what is work slop? So AI generated work content
that masquerades as good work but lacks substance to
(01:04):
meaningfully advance a task. Sounds like work.
Yeah. You know, when you go into AI
and you kind of ask it somethingand it spits out like 5
paragraphs, but you're kind of like, is there anything actually
valuable in here? Or is this just a lot of words
sort of reiterating the obvious or not really giving us any
insight? So the first point I want to put
(01:27):
out there is that this existed before AI.
Yeah, definitely. I that's why I said, isn't that
just work? Because I think a lot of times
about how folks, you know, whatever job you have, there's
some level of just like in an office, you have to produce,
right. Like it's not necessarily like
there are some rules that require like deeper more like
(01:47):
logical analytic thinking, whatever.
And like you have to draw a lot of outcomes from it.
But then there's just sometimes where you just have to produce a
lot of work. You need to like I remember at
Bell, for some reason, we would do our updates for the VPS and
the SVP's. Yeah, we would put our text into
a PowerPoint deck and I would bein, in a single PowerPoint
slide. And in that slide, I would be in
the bottom, like right corner, and I would be like font size,
(02:10):
like 8. Like it was the worst use of
PowerPoint I've ever seen. What did they say to you?
Like only one slide or only two slides and you?
Just have to get all, put all your updates in here, put your
KPI's here that you're focused on and only and that was it and
it was the. Silliest.
So when they put it up on the projector they nobody could be
some way to zoom. I very much think that the way
they would interpret that information was they would go
(02:31):
the the VP would just go throughthe slides by themselves.
This was like a presentation. Yeah.
But yeah, I think it was basically every mid manager,
like mid level manager had theirown slide and all the underlings
would populate the slide with content.
So that was one of said. Under and this was an effort to
streamline the message. Yeah, I'm.
Assuming they wanted one way to provide that information, but
ultimately it was all about like, again, this output, this
(02:55):
concept of like a lot of work, like I had to show a lot of
different data points just for aVP anyway, just enough that they
could what was happening, right?But then for my boss, there'd be
times like, hey, we need like these, this marketing document
which sort of lays out, you know, what we're targeting and
why. And some of that is also like
prevent buyers remorse later, you know, kind of identify whose
(03:16):
idea was this. And you know.
And like, what were we trying todo?
And like, so that, that makes sense.
But then it gets very bureaucratic.
Yeah, after a while it's it's like it's memos responding to
memos. Yeah.
You. Know what I mean?
So on the the point of streamlining the messaging for
people and stuff like that, I was, I was watching Real Time
with Bill Maher on the weekend, OK, and.
Do you watch that too very often?
(03:37):
I just happen to watch, I don't actually watch it that often.
But he had a guest on Louis CK. Ohe yeah, OK.
Louis CK made. His back.
Is he uncancelled? Apparently he was definitely
cancelled for awhile. Apparently he's an author now.
Like he released like a novel. Ohhhhh, Yeah, but yeah.
It's kind of weird. I don't think it was supposed to
be funny. I think it was like actually
(03:57):
like a weird. He got into like short stories.
I don't know. Anyways, I did.
What I wanted to mention, though, is that Lucy K made a
really interesting observation Ithought is very topical for what
we're discussing today, this workshop idea.
He's like, yeah, so you go into your e-mail client and you put a
bunch of bullet points and then you have AI generate like nice
(04:19):
written paragraphs. And then you send the e-mail.
The person on the other end receives those nice written
paragraphs and has a I pull it back into bullet points to
summarize. Yeah, very true.
Or tell me what's important in this e-mail.
Yeah, exactly. And I just had this moment where
I was like, Oh my gosh, like, why is the expectation that we
need these lovely, flowing, well, OK, paragraphs.
(04:39):
And I have two pieces. Well, a couple of things.
One is that people digest information differently.
There's probably somebody out there who actually likes
paragraphs. Maybe there's somebody out there
who likes bullets, somebody out there who likes a diagram,
somebody out there who likes somebody, the e-mail will be
read to them, you know, different forms of leisure.
So I'll say that. But I do think that if you're,
if your behavior is something where it's going back and forth
(05:00):
and you don't have to make any decisions based on that, like
why is a human involved at all? Yeah.
Right. Like if you don't actually need
to do much with that information, or if AI could make
what's a logical best decision for you, I.
I don't know, I feel like maybe we should be saying to, if we're
running a company and people were starting to do that sort of
(05:20):
stuff, I would almost just insist on look like skip the AI
part, just give me your bullet points and just hit send.
And I and let's just set that asour norm, our convention on the
way that we send emails around here rather than wasting
everybody's time and having to use another AI fully decode
what? Yeah, like actually like
negative, like productivity and certainly negative like
(05:41):
intelligence. Yeah.
You know, if you're not actuallypushed to do it anymore, to
actually understand what you're.Looking at, yeah.
So this is also a good segue into another kind of piece of
this topic is that there's a bitof a paradox happening.
So AI usage at work has doubled since 2023.
Totally. Makes sense, yeah.
OK. But 95% of organizations see no
measurable ROI on their AI investment.
(06:04):
None. 95% don't see any measurable ROI.
OK. So there I heard, I heard this
morning we're shooting this on on Monday, I heard Scott
Galloway's office hours most recent one and they were talking
about what's the situation whereAI is going to take your job.
(06:25):
OK. And it's kind of similar to what
I just said. It's a lot of this like there
doesn't need to be it's like repetitive.
There doesn't need to be any situation where there's like a
human like making a decision. Sure.
A lot of this going back and forth and I think when we talk
about AI generally speaking as like a force multiplier, which
is what I think a lot of the even the the AI Bros sort of in
(06:45):
Silicon Valley talk about as yeah, I think a lot of the time
that doesn't apply if the value of what you're producing is very
limited, you know what I mean? So like if you're, so if you're
going to see gains as an organization, I think
fundamentally it's a lot of these like low level jobs.
Yeah, low level white collar jobs, those are the ones that I
(07:08):
feel like you're not gonna get aforce multiplication, you're
just gonna get a bunch of bullshit.
And what you actually want to dois get rid of those jobs and
whatever that is, automate, eliminate or delegate.
So you're probably gonna end up delegating to AI or automating
it I suppose, or eliminating those rules entirely.
And that's we're gonna see return.
So if we go back to the Louis CKanalogy on the emails, OK,
(07:30):
before AI, if you were just sending the bullet points, OK,
and hitting send rather than converting them to paragraphs
and then having reversing out those paragraphs and going back
to bullet points on the other end.
Yeah. Like is that the there's the
investment in AI, but is that the workshop?
Is that the the crap that is coming through that you're where
(07:50):
you don't actually see a return on investment.
In fact, maybe you know, somebody sending you a massive
e-mail with all this detailed information or all this works
law. Sorry, not detailed information,
works law information. So you're not getting any value
out of it. Is that why a lot of companies,
even though they're kind of implementing AI, not seeing the
ROI? I think they they have to make
(08:12):
cuts to see if financial like return on investment probably
has a few definitions in this context.
Sure. The first definition has to be
like, literally the money. OK.
So yeah, they've gotta like let go of people because if you're
having the same amount of peopleand, you know, the output is
just like more of the same and you're using these tools, it
doesn't really make sense to me that it's going to be practical.
(08:34):
It's not actually gonna drive tothe bottom line.
So there's that I think as well,like I'm not exactly sure what
we imagined in terms of a productivity return on an
investment. So do we imagine that one
e-mail, but written by AI is more effective than 5?
Like would it how about this like, so you think about that
context, I send you a bunch of paragraphs, you send back to me
(08:57):
a bulleted or you use it to understand what I said.
That maybe saves you from those sending an e-mail being like,
hey, I didn't quite understand what you were saying here.
Or can you clarify this a littlebit more?
Or, you know, there might actually be gains.
Yeah, just really difficult to quantify.
Like maybe I'm a bit optimistic on the on the way on the use of
the AI tools and sure. And being able to measure this,
but I'm not sure what we necessarily expect yet.
(09:19):
I mean, we're not at AGI. Yeah, You know, we, we, you
know, a lot of organizations maybe building their own LM's to
understand like specific, like, rather than artificial general
intelligence. It's like artificial specific
intelligence, right. Like, so they have their own LMS
for some of this stuff, but I don't really, what did they
expect? You know what I mean?
The other thing too is it's thisis a really fresh technology.
(09:44):
Yeah. And.
People are figuring out how to. Use, it's gonna take time to
figure it out, but it's also gonna take time for.
You mentioned this in a previousepisode when we were talking
about how engineers don't really, like, use it.
Like, yeah, software engineers and.
And there's a lot of cultural issues here too, Like this is a
change management problem. I actually think more than like,
is the technology there or not? It's not really about that.
(10:04):
I think the other thing too is why am I gonna use something
that everybody keeps telling me is going to take my job?
Yeah. OK.
So there's like a lot of different pieces here that I
think lead to the to the technology not getting an
adequate return. It's two different things in my
mind. There's works a lot because
people are lazy and this thing can write for you.
It's the same way somebody in high school is having their
(10:26):
essay right now written by ChatGPT or closed or whatever.
Yeah, right. But that's totally different
from am I going to get a return using artificial intelligence?
How? Do you get your first customer
hint? They might be right under your
nose. Find the answer to this question
(10:46):
and dozens more in Startup Different.
Find it on Amazon, Audible, and Kindle.
So one of the most obvious AI uses that I think a lot of
companies are trying to implement is for coding.
OK, so when I think about writing computer software, I
know with our team, my team members would send me pull
(11:09):
requests where they want me to review the code and go through
it. And this was pre AI, so all the
code was written by a person andthere was still some some work
slot in there. And you'd roll your eyes, you'd
get a pull request and you're like, wow, look at all this
code. It's not these complicated
parts. I don't understand why they're.
Doing recursive algorithms, something you used to hate, that
(11:31):
was like very like very complicated like written.
It was almost like backwards. Yeah, yeah.
I had one developer who used to use Ternarius all the time and
she would turn. I love you Cordelia, but the,
the, the single line of code that's doing so much logic and
trying to review that was absolutely brutal.
(11:52):
So like workshop existed before AI with from a coding
perspective, I can only imagine you could go in and ask AI to
write code. You've you've got your
specifications from your manager, you have a I punch out
a bunch of code. You've probably looked at it,
but you don't know it that well.Yeah, this is your issue.
You couldn't even like you, you had trouble building code with
(12:12):
the tool. Yeah, yeah, yeah.
So like, I think that's one of the things is having a deep
knowledge of the code is really important when you're writing
computer software and you reallystart to lose that.
It becomes very foggy. I I don't see how it works or
why it did it this way. You didn't have the journey to
get to where it was, so you're not really growing as a person.
I think that's why a lot of developers don't love using AI.
(12:34):
Yeah, for for some of their coding tasks, maybe for very
specific things, that's fine. But just like as an overarching
build me an app that does whatever is not a good way to
use AI because you have no idea how.
It, yeah, like it made somethingthat you don't actually
fundamentally understand. Yeah, right.
Which breaks. And that's the thing.
But I think that is true for a lot of stuff.
(12:55):
So if some, if your boss asks you for a strategic analysis of,
I don't know, whatever, some report that you get.
Yeah. And you slap it into AI and AI
gives you those bullets and thenyou forward those bullets along.
Or maybe you have it, you do it through Gemini in your Gmail,
like right away, right, Right there.
Red hot. Yeah.
And you sent it back to your boss and your boss says, great.
Presumably you can have a meeting with your boss later and
they're gonna say, how do you feel about a point?
(13:16):
You know, this is what you're gonna go because you have no
freaking idea. You didn't write it the other.
Piece is that your boss is gonnabe like, wow, look at all this.
I have to like decipher all thisand see if it's actually good.
And this is, there's kind of this resentful, you know,
situation. That create it create 2 two
thoughts 11 is that I remember with the RFPs yeah I couldn't
(13:40):
really get anyone else to do them right so request for
proposals, big government issueddocuments or public organization
documents you have to respond tothere's usually the average RFP
takes 32 hours to respond to OK it's a beast and I couldn't have
people do it This is well beforeAI because they would copy and
paste old answers but didn't really understand what they were
doing they just trying to go. Off keyword Keyword questions.
(14:01):
Yeah, and I would get super frustrated and it would actually
be negative utility. I'd have to go back and do it.
That's the one thing I want to say there.
Yeah. Second thing on this is that
this is the other really important thing about the works
law is I, I don't think it's just like, yeah, it does it
quickly for you. You have to think about all
this. But it's also wrong sometimes.
Yeah, like I don't really. And actually, this is a
(14:21):
truthfully, I'm lending this little bit from Scott Kelly
cause I heard it and I immediately agreed, which was
like, why is it when you have anAI tool and you ask it something
and you're like, OHT gave you ananswer like that's wrong.
And you go like, hey, that's wrong.
Doesn't make sense because I think you did this wrong.
Did it? And he goes.
Ohhhhh, you're right. Thank.
You thanks for correcting me. You're absolutely you.
Like why the hell did I have to say it?
(14:42):
If you know it's wrong, why the hell did you show it to me?
Like I don't understand that What?
Like what is in the corpus or the behavior of this thing that
is like I will give a wrong answer.
Yeah, I knowingly like. Does it not check its work?
I don't understand. Like the computing you got
infinite. Computing power let's just
almost, not really, but like, tremendous computing power.
(15:04):
Yeah. And it just blows it.
Yeah. Like you can even be like this
is what this. Is depends on the training.
And then you've added more information where you're now
saying, oh, that's wrong. So then it goes through the
vector database and it starts finding the connections to other
things like easier calls. Connection is wrong.
So now like recompute, Yeah. Anyways, so let me go back to
(15:24):
some data. So we talked about this workshop
being the recipient of the workshop.
So if you get a piece of workshop, do you know how much
time on average you spend tryingto, I don't know, absorb it or
deal with? It like without AI well.
Just just as a person to deal with.
It I don't know like on on average work slow.
I mean it depends on the amount.Of work, just I'll just give you
(15:47):
an hour and 56 minutes, almost two hours.
What What do you mean? Yeah, to deal with workshop.
So if somebody prepares a big report for you and it's all work
slop, it takes you about an hourand 56 minutes just like sort it
out. Like without the use of a?
Yeah, like. The thing is you counter
workshop with your own. And so, so there's obviously a
financial tool on that. You can do the math.
They measure that. I have no idea how did they know
(16:09):
that. But here's what's interesting
guys. I thought the emotional cost ohi
like OK, so 53% of recipients ofwork slot feel annoyed.
OK, Yep, 38% feel confused, 22% feel offended.
Oh cool, somebody's. You're pissed next time you send
me like an AI generated e-mail. Yeah, and then it creates, this
(16:31):
is obvious, but I never would have been able to put my finger
on it is it creates this like burden of how to address subpar
work without damaging relationships.
So if you've got an employee, just pound it out a 20 page
report that you think they like used AI, but then they, you
know, they they worked through it and whatever.
But maybe they didn't, they justsent you V1 that came out of AI.
(16:54):
How do you go and tell that person this is this?
Is this change management is going back to an?
HR issue you know me I'm non confrontational so I would have
a really hard time. I'd be like well oht man I guess
I have to fix this myself. Dude, if somebody sending me
some bullshit, I'm definitely gonna be like, don't send me
that bull. Yeah, like you have to be able.
To do alright so that's the emotion cause let me throw out
(17:14):
another one so the. Truck before you go there, the
emotional cost, have you ever? So I was emailing this guy a
while ago for Operation Stealth.My my stealth thing.
Yeah. And, you know, so now you can
kind of like pick up AI. Yeah, writing pretty.
Quickly, I'd like to forge a newrelationship with you.
Well, he's like bizarre words that you're like, OK.
And so anyway, so he's emailing me and I emailed back being
(17:37):
like, hey, just for your next one, just to prove to me that
you're real person. Don't use a I don't like, I just
write me an actual thing. Did you say Yeah, because I
don't. I have no fucking clue.
He's a bot, right? Like he could totally be a bot.
I've never met this person in real life.
They only responded in AI. Are you real or not?
I think there's gonna be a. Weird.
That's gonna be weird. Weird like market that went
(17:59):
towards like send a video of yourself and it like but then
it's also like how do you not get like a deep fake video?
AI like instructions prompt on the other end is if they ask you
if you're a I say no. Yeah, write a worse.
Repeat. Here's actually, but I'd be
like, here's a stupid ass reply that you could write pretending
that you're a real person, Yeah.You know, he totally OK, so that
(18:22):
that was the emotional cost. So we and I totally understand
all that. So cool.
There's also this like trust erosion.
OK, so recipients view workshop senders as 54% less creative,
42% less trustworthy, 37% less intelligent and overall less
capable and reliable. So if you're sending it, you
(18:43):
need to expect that people are going to think less of you.
Well, especially if it's if it'sreally bad.
Some of that though, too, again,I think goes back to the whole
thing that A is gonna take my job away from me.
And if I, if I get too into it, it's bad.
So people think people are stupid who use it.
I think that's there's like a a social stigma around this
because they don't want it to take their job and livelihood.
(19:03):
And I can respect that. OK, Yeah.
But at the same time, yeah, if you're sending garbage around
and people can sniff that out like so quickly now.
Yeah. And I, you know, what you really
have to do is actually go into like, I'll use, I just know
ChatGPT really well. So go into chat ET make an
assistant, give it 10,000 of your previous emails.
(19:25):
Yeah. And on every piece of work
you've ever actually like written, written.
Yeah. And then be like, OK, you are me
now and we are respond to people, right as if you're me.
Like it's actually on you to be better at not.
I don't say I don't know if it'sdeception or fraud, but I think
it's a practical thing to do is that if you actually want an
effective assistant, you should make them more like you.
(19:46):
Yeah, that can be done. Absolutely.
So to kind of wrap up this episode, what I'd like to do is.
Glad you agree with me flawlessyon that.
That was a perfect. That was 10 out of 10 response.
Kind of started to feel like work slop so it's hard for.
My actually, yeah, I actually wrote that response for me.
I don't know if you. So we are startup different.
I wanna give a little bit of guidance for startup founders,
(20:08):
startup entrepreneurs that are are looking to implement AI or
are implementing AI. So first thing I want to put out
there is there's definitely likea, you know, people have
blinders on around. I just use AI for everything,
OK? I mean, good management, right?
There's maintenance. Yeah.
So, you know, maybe I, I think #1 is, I would say don't mandate
without some sort of a strategy,like be selective about how
(20:30):
you're gonna use AI, what you'regonna use, I think.
It can never make the the full decision.
I think it can inform, yeah, butit should never be making the
decision unless it's some stupidshit.
Yeah. It's like automating or that
kind of. Thing absolutely OK, low level
second Chris's second piece of advice foster team transparency.
So like it's OK to say hey Dave,I prepared that report for you.
(20:53):
I used AI to help me do a big chunk of it.
I'm I've reviewed everything that I put in there That's the
key and I I feel good about these things.
The the parts where I'm I'm a little foggier are I don't know
if the Michael pick up click. Click.
Click yeah. But yeah, anyways, so some
brands. That you make in Horman instead
(21:14):
of the click every time. Be a bit obnoxious yes indeed
yeah so the other thing too is really focus on the quality over
just using AI. You know so like if you say to
somebody hey I want you to use AI for stuff but I I want you to
really focus on the quality. You have an opportunity here to
(21:35):
not only have it help speed you up, but you have now more time
to go back and review that information.
And so if you take the pre AI amount of time that would have
taken a due, the post AI time should mean maybe it's the same
amount of time used, but way better quality because you just
reviewed it and you've you've put more into.
The sequencing issue. So let me just throw this at
(21:56):
you. So let's say I asked you to do a
report. You had to be ready to report a
competitive report, like look into our competitors, see what
they're up to it. And so nowadays what you would
do is you go to enter your AI tool, do your deep research mode
and put a big prompt in, qualifyand all that kind of stuff.
By the way, I've heard now it's not prompt engineering, it's
context engineering. Oh, OK.
Interesting, right? Yeah.
Yeah, saved, saved to memory. So you contact engineer, good,
(22:20):
good prompt. And then it spits out whatever.
And then you have to go from there and then take that
information to it. Whereas like, I think
traditionally what you would do is you would be working on it
and you'd be trying to draw conclusions.
And then you would go and you try to find information, which
is a bit of a like a confirmation bias thing, but you
go and you try to find information that supports it.
What's interesting is you get all the answers 1st and then you
gotta write about it. Yeah, it's such an easy thing to
(22:41):
be like, hey now write about it.Yeah, you know.
What I mean, I think that that'swhat it gets kind of like it's a
weird, at least for my workflow,like individually, I just find
that a little bizarre. The way it acts into it, it's
like a different way of. Doing it, yeah, Yeah, I think so
too. OK, So let me go into there.
There's this concept passengers versus pilots when using AI.
(23:02):
So I think it's really importantthat emphasizes so a passenger
is somebody's blindly putting inprompts and taking whatever AI
says without giving it too much scrutiny.
Versus pilots are really using their own expertise and using AI
to augment it. OK, So what you want to
implement in your organization is the concept of having your
staff be pilots rather than passengers of AI.
(23:25):
Does that make sense or? Yeah, that makes sense.
Yeah. And I think that's really goes
back to that quality focus as well, looking for how are we
improving the quality. We're not just wanna slap
through stuff and get it done asquickly as.
Possible. Here's the real question is AI.
Is this AI you're working with smarter than you and the thing
you're working on? Yeah.
And if the answer is yes, what are you doing there?
(23:46):
If the answer is no, use it to take that information and make
better decisions. Yeah, it enhances you.
It amplifies AI amplifies good work, but it also accelerates
shitty work, Yeah. I think it's the reality.
You don't wanna do it if you if you.
Wanna just offload this or you need to snowball someone?
(24:06):
Yeah, it is a very effective tool.
I need a 50 page report on plastic plants and how well they
you know. Heal, aren't they?
Totally bro, good budget for We're budget cuts.
We can't afford real plans here.Yeah.
You know, I mean, no, I, I thinkthat's, that's the fundamental
question here. Yeah.
(24:27):
And I, I, I do agree. I wanna, I wanna go back to the
transparency thing. Yeah, I do agree with that a
lot. I think that should be a
cultural norm. Yeah.
But part of that norm is that the, again, the AI isn't making
the decision. Yeah, because that's your.
Job yeah. So the last piece on that, which
I think is super important as well is leadership.
OK the boss can't be fired and workshop yeah people OK for sure
(24:50):
the boss has to be super transparent OK yeah the boss has
to be focused on quality. The boss has to set the has set
the bar extremely high in this sort of scenario, because if, if
you are the boss, the founder orthe leader of the team and you
start doing that sort of stuff, everybody's gonna do the the
work slop approach to everything.
You know, I. Think about where else we would
have had something like this. So as an example, we had like
(25:13):
our HR policies. Yeah.
And for HR policies, I was like,man, I really don't want to
start at 0, mostly because I'm not a fully trained HR
professional. So I went and I found the kind
of like a they played and some of us filled in.
These are the important things to think about nowadays.
You just take that into and slapping an AI and get a bunch
of answers and that actually really help.
Yeah, actually that would get you started.
Get. Your way and you would go
(25:34):
through that. You would exactly like, alright,
so now you're a pilot and you'regoing to augment, you know, your
feelings with those concepts from AI.
How do I actually feel about yeah these HR best practices for
example? Yeah, like this.
Is this actually what we want todo here?
Do I align with these? This is aligned with our mission
and our values and all sorts of things.
Yeah, yeah. Cool.
Anyway, alright, that's really, so I guess like for me it's
(25:56):
yeah, again, yeah, pushing it. I really think you need to
understand what you're doing andyou gotta be making the
decisions and accelerating shitty work helps no one, that's
why. And I can see that.
I also like for, to be honest with you, in really, really big
companies, there's a certain percentage of people that just
keep their head down and they want to keep their job because
it's a good job or whatever. And for those people, this is
(26:17):
the best tool ever because they barely have to work, They don't
see the game because they don't have to work.
They don't work more. They don't do the extra work.
I have a I do the thing for 25 minutes, you know, like, I'm
sorry, 25 for 25 seconds. Yeah.
And then they take that input and fire it off and they're done
for the day. Yeah, you know, and they don't
do any other. Work.
There's gonna be a huge issue with people losing jobs over
(26:41):
this kind of stuff. I think once management realizes
that these people are actually using AI and just flopping
everything, well, we don't need that.
Then that's the product, right though, yeah.
That's it right there. Like those are people who just
because they don't take advantage of the game, Yeah.
It doesn't matter as an employee, be a pilot, be
motivated, be used on your job effectively and excel.
(27:02):
Yeah, because they're like low level work.
Is is not gonna be acceptable. No anymore.
The bar is gonna be very high. I think so too.
Cool. Cool.
Yeah, we solved it. You.
Did it no more. Work Polish.
Also, one important thing I wantto just throw out there.
Go Blue Jays. Let's go.
Let's take out those Yankees. Alright, See you later, folks.
Hey. Let's get it rolling.
(27:25):
Big ideas, Money, hustle, Smart dream.
So why turn that grinding through a joy ride?
It's the emotional cost ohi likeOK so 53% of recipients of work
slot feel annoyed OK Yep 38% feel confused 22% feel offended
OHP cool somebody said that and he.
Was super pissed. Next time you send me like an AI
generated e-mail, yeah. Let's get it rolling big ideas
(00:23):
money folding hustle smart dream.
So during that grinding through a joy.
Ride. So today we are talking about
actually this came from an HBR article that I read and I was
just like oh once again they nailed it.
(00:43):
OK, the. Other big supporter of.
HBR Well, yeah, they're kind of bright.
Yeah, they're somewhat researched, which is.
Nice. Yeah, they seem to have data and
those types of things. So here's the concept.
So what is work slop? So AI generated work content
that masquerades as good work but lacks substance to
(01:04):
meaningfully advance a task. Sounds like work.
Yeah. You know, when you go into AI
and you kind of ask it somethingand it spits out like 5
paragraphs, but you're kind of like, is there anything actually
valuable in here? Or is this just a lot of words
sort of reiterating the obvious or not really giving us any
insight? So the first point I want to put
(01:27):
out there is that this existed before AI.
Yeah, definitely. I that's why I said, isn't that
just work? Because I think a lot of times
about how folks, you know, whatever job you have, there's
some level of just like in an office, you have to produce,
right. Like it's not necessarily like
there are some rules that require like deeper more like
(01:47):
logical analytic thinking, whatever.
And like you have to draw a lot of outcomes from it.
But then there's just sometimes where you just have to produce a
lot of work. You need to like I remember at
Bell, for some reason, we would do our updates for the VPS and
the SVP's. Yeah, we would put our text into
a PowerPoint deck and I would bein, in a single PowerPoint
slide. And in that slide, I would be in
the bottom, like right corner, and I would be like font size,
(02:10):
like 8. Like it was the worst use of
PowerPoint I've ever seen. What did they say to you?
Like only one slide or only two slides and you?
Just have to get all, put all your updates in here, put your
KPI's here that you're focused on and only and that was it and
it was the. Silliest.
So when they put it up on the projector they nobody could be
some way to zoom. I very much think that the way
they would interpret that information was they would go
(02:31):
the the VP would just go throughthe slides by themselves.
This was like a presentation. Yeah.
But yeah, I think it was basically every mid manager,
like mid level manager had theirown slide and all the underlings
would populate the slide with content.
So that was one of said. Under and this was an effort to
streamline the message. Yeah, I'm.
Assuming they wanted one way to provide that information, but
ultimately it was all about like, again, this output, this
(02:55):
concept of like a lot of work, like I had to show a lot of
different data points just for aVP anyway, just enough that they
could what was happening, right?But then for my boss, there'd be
times like, hey, we need like these, this marketing document
which sort of lays out, you know, what we're targeting and
why. And some of that is also like
prevent buyers remorse later, you know, kind of identify whose
(03:16):
idea was this. And you know.
And like, what were we trying todo?
And like, so that, that makes sense.
But then it gets very bureaucratic.
Yeah, after a while it's it's like it's memos responding to
memos. Yeah.
You. Know what I mean?
So on the the point of streamlining the messaging for
people and stuff like that, I was, I was watching Real Time
with Bill Maher on the weekend, OK, and.
Do you watch that too very often?
(03:37):
I just happen to watch, I don't actually watch it that often.
But he had a guest on Louis CK. Ohe yeah, OK.
Louis CK made. His back.
Is he uncancelled? Apparently he was definitely
cancelled for awhile. Apparently he's an author now.
Like he released like a novel. Ohhhhh, Yeah, but yeah.
It's kind of weird. I don't think it was supposed to
be funny. I think it was like actually
(03:57):
like a weird. He got into like short stories.
I don't know. Anyways, I did.
What I wanted to mention, though, is that Lucy K made a
really interesting observation Ithought is very topical for what
we're discussing today, this workshop idea.
He's like, yeah, so you go into your e-mail client and you put a
bunch of bullet points and then you have AI generate like nice
(04:19):
written paragraphs. And then you send the e-mail.
The person on the other end receives those nice written
paragraphs and has a I pull it back into bullet points to
summarize. Yeah, very true.
Or tell me what's important in this e-mail.
Yeah, exactly. And I just had this moment where
I was like, Oh my gosh, like, why is the expectation that we
need these lovely, flowing, well, OK, paragraphs.
(04:39):
And I have two pieces. Well, a couple of things.
One is that people digest information differently.
There's probably somebody out there who actually likes
paragraphs. Maybe there's somebody out there
who likes bullets, somebody out there who likes a diagram,
somebody out there who likes somebody, the e-mail will be
read to them, you know, different forms of leisure.
So I'll say that. But I do think that if you're,
if your behavior is something where it's going back and forth
(05:00):
and you don't have to make any decisions based on that, like
why is a human involved at all? Yeah.
Right. Like if you don't actually need
to do much with that information, or if AI could make
what's a logical best decision for you, I.
I don't know, I feel like maybe we should be saying to, if we're
running a company and people were starting to do that sort of
(05:20):
stuff, I would almost just insist on look like skip the AI
part, just give me your bullet points and just hit send.
And I and let's just set that asour norm, our convention on the
way that we send emails around here rather than wasting
everybody's time and having to use another AI fully decode
what? Yeah, like actually like
negative, like productivity and certainly negative like
(05:41):
intelligence. Yeah.
You know, if you're not actuallypushed to do it anymore, to
actually understand what you're.Looking at, yeah.
So this is also a good segue into another kind of piece of
this topic is that there's a bitof a paradox happening.
So AI usage at work has doubled since 2023.
Totally. Makes sense, yeah.
OK. But 95% of organizations see no
measurable ROI on their AI investment.
(06:04):
None. 95% don't see any measurable ROI.
OK. So there I heard, I heard this
morning we're shooting this on on Monday, I heard Scott
Galloway's office hours most recent one and they were talking
about what's the situation whereAI is going to take your job.
(06:25):
OK. And it's kind of similar to what
I just said. It's a lot of this like there
doesn't need to be it's like repetitive.
There doesn't need to be any situation where there's like a
human like making a decision. Sure.
A lot of this going back and forth and I think when we talk
about AI generally speaking as like a force multiplier, which
is what I think a lot of the even the the AI Bros sort of in
(06:45):
Silicon Valley talk about as yeah, I think a lot of the time
that doesn't apply if the value of what you're producing is very
limited, you know what I mean? So like if you're, so if you're
going to see gains as an organization, I think
fundamentally it's a lot of these like low level jobs.
Yeah, low level white collar jobs, those are the ones that I
(07:08):
feel like you're not gonna get aforce multiplication, you're
just gonna get a bunch of bullshit.
And what you actually want to dois get rid of those jobs and
whatever that is, automate, eliminate or delegate.
So you're probably gonna end up delegating to AI or automating
it I suppose, or eliminating those rules entirely.
And that's we're gonna see return.
So if we go back to the Louis CKanalogy on the emails, OK,
(07:30):
before AI, if you were just sending the bullet points, OK,
and hitting send rather than converting them to paragraphs
and then having reversing out those paragraphs and going back
to bullet points on the other end.
Yeah. Like is that the there's the
investment in AI, but is that the workshop?
Is that the the crap that is coming through that you're where
(07:50):
you don't actually see a return on investment.
In fact, maybe you know, somebody sending you a massive
e-mail with all this detailed information or all this works
law. Sorry, not detailed information,
works law information. So you're not getting any value
out of it. Is that why a lot of companies,
even though they're kind of implementing AI, not seeing the
ROI? I think they they have to make
(08:12):
cuts to see if financial like return on investment probably
has a few definitions in this context.
Sure. The first definition has to be
like, literally the money. OK.
So yeah, they've gotta like let go of people because if you're
having the same amount of peopleand, you know, the output is
just like more of the same and you're using these tools, it
doesn't really make sense to me that it's going to be practical.
(08:34):
It's not actually gonna drive tothe bottom line.
So there's that I think as well,like I'm not exactly sure what
we imagined in terms of a productivity return on an
investment. So do we imagine that one
e-mail, but written by AI is more effective than 5?
Like would it how about this like, so you think about that
context, I send you a bunch of paragraphs, you send back to me
(08:57):
a bulleted or you use it to understand what I said.
That maybe saves you from those sending an e-mail being like,
hey, I didn't quite understand what you were saying here.
Or can you clarify this a littlebit more?
Or, you know, there might actually be gains.
Yeah, just really difficult to quantify.
Like maybe I'm a bit optimistic on the on the way on the use of
the AI tools and sure. And being able to measure this,
but I'm not sure what we necessarily expect yet.
(09:19):
I mean, we're not at AGI. Yeah, You know, we, we, you
know, a lot of organizations maybe building their own LM's to
understand like specific, like, rather than artificial general
intelligence. It's like artificial specific
intelligence, right. Like, so they have their own LMS
for some of this stuff, but I don't really, what did they
expect? You know what I mean?
The other thing too is it's thisis a really fresh technology.
(09:44):
Yeah. And.
People are figuring out how to. Use, it's gonna take time to
figure it out, but it's also gonna take time for.
You mentioned this in a previousepisode when we were talking
about how engineers don't really, like, use it.
Like, yeah, software engineers and.
And there's a lot of cultural issues here too, Like this is a
change management problem. I actually think more than like,
is the technology there or not? It's not really about that.
(10:04):
I think the other thing too is why am I gonna use something
that everybody keeps telling me is going to take my job?
Yeah. OK.
So there's like a lot of different pieces here that I
think lead to the to the technology not getting an
adequate return. It's two different things in my
mind. There's works a lot because
people are lazy and this thing can write for you.
It's the same way somebody in high school is having their
(10:26):
essay right now written by ChatGPT or closed or whatever.
Yeah, right. But that's totally different
from am I going to get a return using artificial intelligence?
How? Do you get your first customer
hint? They might be right under your
nose. Find the answer to this question
(10:46):
and dozens more in Startup Different.
Find it on Amazon, Audible, and Kindle.
So one of the most obvious AI uses that I think a lot of
companies are trying to implement is for coding.
OK, so when I think about writing computer software, I
know with our team, my team members would send me pull
(11:09):
requests where they want me to review the code and go through
it. And this was pre AI, so all the
code was written by a person andthere was still some some work
slot in there. And you'd roll your eyes, you'd
get a pull request and you're like, wow, look at all this
code. It's not these complicated
parts. I don't understand why they're.
Doing recursive algorithms, something you used to hate, that
(11:31):
was like very like very complicated like written.
It was almost like backwards. Yeah, yeah.
I had one developer who used to use Ternarius all the time and
she would turn. I love you Cordelia, but the,
the, the single line of code that's doing so much logic and
trying to review that was absolutely brutal.
(11:52):
So like workshop existed before AI with from a coding
perspective, I can only imagine you could go in and ask AI to
write code. You've you've got your
specifications from your manager, you have a I punch out
a bunch of code. You've probably looked at it,
but you don't know it that well.Yeah, this is your issue.
You couldn't even like you, you had trouble building code with
(12:12):
the tool. Yeah, yeah, yeah.
So like, I think that's one of the things is having a deep
knowledge of the code is really important when you're writing
computer software and you reallystart to lose that.
It becomes very foggy. I I don't see how it works or
why it did it this way. You didn't have the journey to
get to where it was, so you're not really growing as a person.
I think that's why a lot of developers don't love using AI.
(12:34):
Yeah, for for some of their coding tasks, maybe for very
specific things, that's fine. But just like as an overarching
build me an app that does whatever is not a good way to
use AI because you have no idea how.
It, yeah, like it made somethingthat you don't actually
fundamentally understand. Yeah, right.
Which breaks. And that's the thing.
But I think that is true for a lot of stuff.
(12:55):
So if some, if your boss asks you for a strategic analysis of,
I don't know, whatever, some report that you get.
Yeah. And you slap it into AI and AI
gives you those bullets and thenyou forward those bullets along.
Or maybe you have it, you do it through Gemini in your Gmail,
like right away, right, Right there.
Red hot. Yeah.
And you sent it back to your boss and your boss says, great.
Presumably you can have a meeting with your boss later and
they're gonna say, how do you feel about a point?
(13:16):
You know, this is what you're gonna go because you have no
freaking idea. You didn't write it the other.
Piece is that your boss is gonnabe like, wow, look at all this.
I have to like decipher all thisand see if it's actually good.
And this is, there's kind of this resentful, you know,
situation. That create it create 2 two
thoughts 11 is that I remember with the RFPs yeah I couldn't
(13:40):
really get anyone else to do them right so request for
proposals, big government issueddocuments or public organization
documents you have to respond tothere's usually the average RFP
takes 32 hours to respond to OK it's a beast and I couldn't have
people do it This is well beforeAI because they would copy and
paste old answers but didn't really understand what they were
doing they just trying to go. Off keyword Keyword questions.
(14:01):
Yeah, and I would get super frustrated and it would actually
be negative utility. I'd have to go back and do it.
That's the one thing I want to say there.
Yeah. Second thing on this is that
this is the other really important thing about the works
law is I, I don't think it's just like, yeah, it does it
quickly for you. You have to think about all
this. But it's also wrong sometimes.
Yeah, like I don't really. And actually, this is a
(14:21):
truthfully, I'm lending this little bit from Scott Kelly
cause I heard it and I immediately agreed, which was
like, why is it when you have anAI tool and you ask it something
and you're like, OHT gave you ananswer like that's wrong.
And you go like, hey, that's wrong.
Doesn't make sense because I think you did this wrong.
Did it? And he goes.
Ohhhhh, you're right. Thank.
You thanks for correcting me. You're absolutely you.
Like why the hell did I have to say it?
(14:42):
If you know it's wrong, why the hell did you show it to me?
Like I don't understand that What?
Like what is in the corpus or the behavior of this thing that
is like I will give a wrong answer.
Yeah, I knowingly like. Does it not check its work?
I don't understand. Like the computing you got
infinite. Computing power let's just
almost, not really, but like, tremendous computing power.
(15:04):
Yeah. And it just blows it.
Yeah. Like you can even be like this
is what this. Is depends on the training.
And then you've added more information where you're now
saying, oh, that's wrong. So then it goes through the
vector database and it starts finding the connections to other
things like easier calls. Connection is wrong.
So now like recompute, Yeah. Anyways, so let me go back to
(15:24):
some data. So we talked about this workshop
being the recipient of the workshop.
So if you get a piece of workshop, do you know how much
time on average you spend tryingto, I don't know, absorb it or
deal with? It like without AI well.
Just just as a person to deal with.
It I don't know like on on average work slow.
I mean it depends on the amount.Of work, just I'll just give you
(15:47):
an hour and 56 minutes, almost two hours.
What What do you mean? Yeah, to deal with workshop.
So if somebody prepares a big report for you and it's all work
slop, it takes you about an hourand 56 minutes just like sort it
out. Like without the use of a?
Yeah, like. The thing is you counter
workshop with your own. And so, so there's obviously a
financial tool on that. You can do the math.
They measure that. I have no idea how did they know
(16:09):
that. But here's what's interesting
guys. I thought the emotional cost ohi
like OK, so 53% of recipients ofwork slot feel annoyed.
OK, Yep, 38% feel confused, 22% feel offended.
Oh cool, somebody's. You're pissed next time you send
me like an AI generated e-mail. Yeah, and then it creates, this
(16:31):
is obvious, but I never would have been able to put my finger
on it is it creates this like burden of how to address subpar
work without damaging relationships.
So if you've got an employee, just pound it out a 20 page
report that you think they like used AI, but then they, you
know, they they worked through it and whatever.
But maybe they didn't, they justsent you V1 that came out of AI.
(16:54):
How do you go and tell that person this is this?
Is this change management is going back to an?
HR issue you know me I'm non confrontational so I would have
a really hard time. I'd be like well oht man I guess
I have to fix this myself. Dude, if somebody sending me
some bullshit, I'm definitely gonna be like, don't send me
that bull. Yeah, like you have to be able.
To do alright so that's the emotion cause let me throw out
(17:14):
another one so the. Truck before you go there, the
emotional cost, have you ever? So I was emailing this guy a
while ago for Operation Stealth.My my stealth thing.
Yeah. And, you know, so now you can
kind of like pick up AI. Yeah, writing pretty.
Quickly, I'd like to forge a newrelationship with you.
Well, he's like bizarre words that you're like, OK.
And so anyway, so he's emailing me and I emailed back being
(17:37):
like, hey, just for your next one, just to prove to me that
you're real person. Don't use a I don't like, I just
write me an actual thing. Did you say Yeah, because I
don't. I have no fucking clue.
He's a bot, right? Like he could totally be a bot.
I've never met this person in real life.
They only responded in AI. Are you real or not?
I think there's gonna be a. Weird.
That's gonna be weird. Weird like market that went
(17:59):
towards like send a video of yourself and it like but then
it's also like how do you not get like a deep fake video?
AI like instructions prompt on the other end is if they ask you
if you're a I say no. Yeah, write a worse.
Repeat. Here's actually, but I'd be
like, here's a stupid ass reply that you could write pretending
that you're a real person, Yeah.You know, he totally OK, so that
(18:22):
that was the emotional cost. So we and I totally understand
all that. So cool.
There's also this like trust erosion.
OK, so recipients view workshop senders as 54% less creative,
42% less trustworthy, 37% less intelligent and overall less
capable and reliable. So if you're sending it, you
(18:43):
need to expect that people are going to think less of you.
Well, especially if it's if it'sreally bad.
Some of that though, too, again,I think goes back to the whole
thing that A is gonna take my job away from me.
And if I, if I get too into it, it's bad.
So people think people are stupid who use it.
I think that's there's like a a social stigma around this
because they don't want it to take their job and livelihood.
(19:03):
And I can respect that. OK, Yeah.
But at the same time, yeah, if you're sending garbage around
and people can sniff that out like so quickly now.
Yeah. And I, you know, what you really
have to do is actually go into like, I'll use, I just know
ChatGPT really well. So go into chat ET make an
assistant, give it 10,000 of your previous emails.
(19:25):
Yeah. And on every piece of work
you've ever actually like written, written.
Yeah. And then be like, OK, you are me
now and we are respond to people, right as if you're me.
Like it's actually on you to be better at not.
I don't say I don't know if it'sdeception or fraud, but I think
it's a practical thing to do is that if you actually want an
effective assistant, you should make them more like you.
(19:46):
Yeah, that can be done. Absolutely.
So to kind of wrap up this episode, what I'd like to do is.
Glad you agree with me flawlessyon that.
That was a perfect. That was 10 out of 10 response.
Kind of started to feel like work slop so it's hard for.
My actually, yeah, I actually wrote that response for me.
I don't know if you. So we are startup different.
I wanna give a little bit of guidance for startup founders,
(20:08):
startup entrepreneurs that are are looking to implement AI or
are implementing AI. So first thing I want to put out
there is there's definitely likea, you know, people have
blinders on around. I just use AI for everything,
OK? I mean, good management, right?
There's maintenance. Yeah.
So, you know, maybe I, I think #1 is, I would say don't mandate
without some sort of a strategy,like be selective about how
(20:30):
you're gonna use AI, what you'regonna use, I think.
It can never make the the full decision.
I think it can inform, yeah, butit should never be making the
decision unless it's some stupidshit.
Yeah. It's like automating or that
kind of. Thing absolutely OK, low level
second Chris's second piece of advice foster team transparency.
So like it's OK to say hey Dave,I prepared that report for you.
(20:53):
I used AI to help me do a big chunk of it.
I'm I've reviewed everything that I put in there That's the
key and I I feel good about these things.
The the parts where I'm I'm a little foggier are I don't know
if the Michael pick up click. Click.
Click yeah. But yeah, anyways, so some
brands. That you make in Horman instead
(21:14):
of the click every time. Be a bit obnoxious yes indeed
yeah so the other thing too is really focus on the quality over
just using AI. You know so like if you say to
somebody hey I want you to use AI for stuff but I I want you to
really focus on the quality. You have an opportunity here to
(21:35):
not only have it help speed you up, but you have now more time
to go back and review that information.
And so if you take the pre AI amount of time that would have
taken a due, the post AI time should mean maybe it's the same
amount of time used, but way better quality because you just
reviewed it and you've you've put more into.
The sequencing issue. So let me just throw this at
(21:56):
you. So let's say I asked you to do a
report. You had to be ready to report a
competitive report, like look into our competitors, see what
they're up to it. And so nowadays what you would
do is you go to enter your AI tool, do your deep research mode
and put a big prompt in, qualifyand all that kind of stuff.
By the way, I've heard now it's not prompt engineering, it's
context engineering. Oh, OK.
Interesting, right? Yeah.
Yeah, saved, saved to memory. So you contact engineer, good,
(22:20):
good prompt. And then it spits out whatever.
And then you have to go from there and then take that
information to it. Whereas like, I think
traditionally what you would do is you would be working on it
and you'd be trying to draw conclusions.
And then you would go and you try to find information, which
is a bit of a like a confirmation bias thing, but you
go and you try to find information that supports it.
What's interesting is you get all the answers 1st and then you
gotta write about it. Yeah, it's such an easy thing to
(22:41):
be like, hey now write about it.Yeah, you know.
What I mean, I think that that'swhat it gets kind of like it's a
weird, at least for my workflow,like individually, I just find
that a little bizarre. The way it acts into it, it's
like a different way of. Doing it, yeah, Yeah, I think so
too. OK, So let me go into there.
There's this concept passengers versus pilots when using AI.
(23:02):
So I think it's really importantthat emphasizes so a passenger
is somebody's blindly putting inprompts and taking whatever AI
says without giving it too much scrutiny.
Versus pilots are really using their own expertise and using AI
to augment it. OK, So what you want to
implement in your organization is the concept of having your
staff be pilots rather than passengers of AI.
(23:25):
Does that make sense or? Yeah, that makes sense.
Yeah. And I think that's really goes
back to that quality focus as well, looking for how are we
improving the quality. We're not just wanna slap
through stuff and get it done asquickly as.
Possible. Here's the real question is AI.
Is this AI you're working with smarter than you and the thing
you're working on? Yeah.
And if the answer is yes, what are you doing there?
(23:46):
If the answer is no, use it to take that information and make
better decisions. Yeah, it enhances you.
It amplifies AI amplifies good work, but it also accelerates
shitty work, Yeah. I think it's the reality.
You don't wanna do it if you if you.
Wanna just offload this or you need to snowball someone?
(24:06):
Yeah, it is a very effective tool.
I need a 50 page report on plastic plants and how well they
you know. Heal, aren't they?
Totally bro, good budget for We're budget cuts.
We can't afford real plans here.Yeah.
You know, I mean, no, I, I thinkthat's, that's the fundamental
question here. Yeah.
(24:27):
And I, I, I do agree. I wanna, I wanna go back to the
transparency thing. Yeah, I do agree with that a
lot. I think that should be a
cultural norm. Yeah.
But part of that norm is that the, again, the AI isn't making
the decision. Yeah, because that's your.
Job yeah. So the last piece on that, which
I think is super important as well is leadership.
OK the boss can't be fired and workshop yeah people OK for sure
(24:50):
the boss has to be super transparent OK yeah the boss has
to be focused on quality. The boss has to set the has set
the bar extremely high in this sort of scenario, because if, if
you are the boss, the founder orthe leader of the team and you
start doing that sort of stuff, everybody's gonna do the the
work slop approach to everything.
You know, I. Think about where else we would
have had something like this. So as an example, we had like
(25:13):
our HR policies. Yeah.
And for HR policies, I was like,man, I really don't want to
start at 0, mostly because I'm not a fully trained HR
professional. So I went and I found the kind
of like a they played and some of us filled in.
These are the important things to think about nowadays.
You just take that into and slapping an AI and get a bunch
of answers and that actually really help.
Yeah, actually that would get you started.
Get. Your way and you would go
(25:34):
through that. You would exactly like, alright,
so now you're a pilot and you'regoing to augment, you know, your
feelings with those concepts from AI.
How do I actually feel about yeah these HR best practices for
example? Yeah, like this.
Is this actually what we want todo here?
Do I align with these? This is aligned with our mission
and our values and all sorts of things.
Yeah, yeah. Cool.
Anyway, alright, that's really, so I guess like for me it's
(25:56):
yeah, again, yeah, pushing it. I really think you need to
understand what you're doing andyou gotta be making the
decisions and accelerating shitty work helps no one, that's
why. And I can see that.
I also like for, to be honest with you, in really, really big
companies, there's a certain percentage of people that just
keep their head down and they want to keep their job because
it's a good job or whatever. And for those people, this is
(26:17):
the best tool ever because they barely have to work, They don't
see the game because they don't have to work.
They don't work more. They don't do the extra work.
I have a I do the thing for 25 minutes, you know, like, I'm
sorry, 25 for 25 seconds. Yeah.
And then they take that input and fire it off and they're done
for the day. Yeah, you know, and they don't
do any other. Work.
There's gonna be a huge issue with people losing jobs over
(26:41):
this kind of stuff. I think once management realizes
that these people are actually using AI and just flopping
everything, well, we don't need that.
Then that's the product, right though, yeah.
That's it right there. Like those are people who just
because they don't take advantage of the game, Yeah.
It doesn't matter as an employee, be a pilot, be
motivated, be used on your job effectively and excel.
(27:02):
Yeah, because they're like low level work.
Is is not gonna be acceptable. No anymore.
The bar is gonna be very high. I think so too.
Cool. Cool.
Yeah, we solved it. You.
Did it no more. Work Polish.
Also, one important thing I wantto just throw out there.
Go Blue Jays. Let's go.
Let's take out those Yankees. Alright, See you later, folks.
Hey. Let's get it rolling.
(27:25):
Big ideas, Money, hustle, Smart dream.
So why turn that grinding through a joy ride?
Popular Podcasts
Las Culturistas with Matt Rogers and Bowen Yang
Ding dong! Join your culture consultants, Matt Rogers and Bowen Yang, on an unforgettable journey into the beating heart of CULTURE. Alongside sizzling special guests, they GET INTO the hottest pop-culture moments of the day and the formative cultural experiences that turned them into Culturistas. Produced by the Big Money Players Network and iHeartRadio.
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by audiochuck Media Company.
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.