00:00:00
welcome everyone to the Wharton serus XM
00:00:02
podcast series on artificial
00:00:04
intelligence sponsored by analytics at
00:00:06
Warton and artificial intelligence at
00:00:08
Wharton I'm Eric bradow professor of
00:00:10
marketing statistics and data science
00:00:12
here at the Wharton School and I'm also
00:00:14
I guess for this role I'm also the vice
00:00:16
of analytics at Wharton and so I'm here
00:00:18
interviewing thought leaders at the
00:00:19
Wharton School on the impact of AI and
00:00:22
business and of course today's episode
00:00:24
which is part of our multi-part series
00:00:26
is no exception today we're going to be
00:00:28
talking about Ai and robotics or if
00:00:30
you'd like Ai and automation we're going
00:00:32
to be I'm going to be talking to my
00:00:33
colleague Lyn woo Lynn is an associate
00:00:35
professor in our operations informations
00:00:37
and decisions Department here at the
00:00:39
Wharton School uh she teaches everyone
00:00:41
undergrads mbas and PhD classes about
00:00:44
the use and impact of emerging
00:00:45
Technologies so Lynn Welcome to our
00:00:47
podcast thank you thank you for having
00:00:49
me H it's great to be here with you um
00:00:53
one of the things I even mentioned to
00:00:54
you briefly off air is that a lot of
00:00:57
people might be listening to this
00:00:58
podcast and serious XM version and
00:01:01
saying wait a second I thought AI was
00:01:04
just just generative AI That's what I'm
00:01:06
hearing about today I'm hearing about
00:01:07
chat GPT and B and open AI what the hell
00:01:11
does AI have to do with automation so I
00:01:13
thought maybe for a few minutes it would
00:01:14
be good if you take our listeners
00:01:16
through a history of artificial
00:01:18
intelligence and what does it mean to
00:01:21
have artificial intelligence in
00:01:22
Automation and then kind of how
00:01:23
generative AI fits in into the broader
00:01:25
class of AI oh thank you that's a great
00:01:28
question so AI as a field has been
00:01:31
existing for decades like from the 1950s
00:01:33
and 60s that's when people started
00:01:35
thinking about coin a term like
00:01:37
artificial intelligence and like you
00:01:39
know you probably if you were around the
00:01:41
70s 80s you probably know about expert
00:01:43
systems which is just bunch of rules if
00:01:46
them classes if you see a and then this
00:01:48
B happens and they think that will Co
00:01:51
cat out lots of you know medical
00:01:52
knowledge or other type of knowledge out
00:01:54
there and that didn't work so well and
00:01:57
then come the neon networks and the
00:01:59
neuro network was happened around the
00:02:01
1780s again and at the time it wasn't uh
00:02:05
performing very well because as you know
00:02:07
neuron Network needs a lot of data and
00:02:09
come the 90s actually we had AI winter
00:02:12
basically near wasn't working a lot of
00:02:14
AI techniques weren't working so they
00:02:17
actually thought AI is not going to
00:02:18
happen for a long time and then we had
00:02:20
the internet explosion we have all the
00:02:22
digital Trac of our you know internet
00:02:25
activities our search activities our
00:02:26
social media our videos our photos that
00:02:30
data explosion ultimately fueled the
00:02:33
current AI Revolution because now the
00:02:35
neuro network has ton of data and then
00:02:39
you can build up very very big neuron
00:02:42
networks which we call Deep learning
00:02:44
okay and deep learning was was you know
00:02:46
was driving the AI Revolution from 2010
00:02:49
to about 2018 and then a special type of
00:02:53
AI techniques called
00:02:55
Transformers which is basically you
00:02:57
think about just make AI neuron networks
00:03:00
run really fast computationally very
00:03:04
efficient and that again saves a ton of
00:03:07
resources and that allows generative AI
00:03:09
so that's like although I'm really
00:03:11
simplify things that that's like just
00:03:13
what it is so if you think of generative
00:03:14
AI is only a tip of the very end and
00:03:17
although it's transformative I I gu I
00:03:20
mean I absolutely agree with you with if
00:03:21
you see chbt and Dolly is amazing but is
00:03:25
really this extension of existing
00:03:28
technology in the Ro in deep learning
00:03:30
yes just in this case predicting it to
00:03:32
language so these are large language
00:03:34
models but there's as you're as we're
00:03:36
going to talk about today they're
00:03:37
applications of AI in many many
00:03:39
different areas and also I think it's
00:03:41
also valuable for our listeners to know
00:03:43
you know a lot of us well a lot of us
00:03:45
trained back in the old days as
00:03:47
statisticians but even all of us are
00:03:49
familiar with regression which is
00:03:50
obviously a form of linear prediction
00:03:53
models and neural Nets obviously and
00:03:55
deep learning just allow for much more
00:03:58
complicated interactions between
00:04:00
variables and that's what I think I
00:04:01
maybe you could just comment briefly on
00:04:03
this that's why these Simple Rules tend
00:04:05
not to work well because the way real
00:04:08
life Works isn't just simple if thens
00:04:11
it's not like a linear regression more
00:04:13
is necessarily Better or Worse there are
00:04:15
these complicated relationships that's
00:04:18
exactly right so often we cannot
00:04:20
describe a relationship using a simple
00:04:22
representation or a linear model what
00:04:25
neur net word does is it you can have
00:04:28
link very variety of different
00:04:30
functional form in a way that we cannot
00:04:32
describe any at all but machine knows
00:04:36
how to transform input output through
00:04:38
many different layers of Transformations
00:04:41
well let's actually talk about the main
00:04:42
topic and first of all thanks Lynn for
00:04:44
bringing that up because again a big
00:04:46
part of this series A lot of people are
00:04:48
talking about generative Ai and you've
00:04:50
just pointed out it is just a special
00:04:52
case and a specific application area so
00:04:54
let's talk about this a matter of fact
00:04:56
one of the things I present to my MBA
00:04:57
students which is interesting that it's
00:04:59
something that you wrote down in advance
00:05:00
as a question is that previous Studies
00:05:03
have indicated that between 40 to 70% of
00:05:05
jobs could be automated so let's start
00:05:08
with like which jobs do you think could
00:05:10
be automated and two what would prevent
00:05:14
that from happening or there is kind of
00:05:15
no way back at this point we're talking
00:05:18
about generative AI specifically like
00:05:21
the TP like this could be or it could be
00:05:23
for example now Automation and and
00:05:27
artificial intelligence is going to
00:05:28
allow Rob OTS to do lots of physical
00:05:31
tasks that we couldn't do before so it
00:05:33
could be Vision based types of jobs and
00:05:35
opportunities it could be I don't know
00:05:38
robots doing surgery it could be robots
00:05:40
replacing people in Plants it could be
00:05:42
in lots of different ways okay great so
00:05:44
let me be more concrete I talk about
00:05:47
robotics okay so how how physical robots
00:05:51
can change labor composition right
00:05:54
so contrary to the popular notion that
00:05:57
40 60% of employees are actually getting
00:05:59
laid off when they have no jobs so in
00:06:01
our longitudal research over about 20
00:06:03
years using Canadian data where we
00:06:05
capture every single robots being being
00:06:08
used in the in the firm in establishment
00:06:10
and look at the you know what kind of
00:06:13
employees are working with the robots
00:06:15
who are there and who are not there
00:06:17
anymore and the the revenues their firm
00:06:20
practices in terms of HR Management in
00:06:23
terms of uh how uh people are being
00:06:26
rewarded so we found it uh interestingly
00:06:29
that when firms adopt robots they
00:06:32
actually hired more
00:06:34
people they did not lay off more people
00:06:37
very interesting why so why is that it
00:06:39
actually turns out um it's a robot
00:06:42
adopters who are becoming much more
00:06:45
productive and much more efficient so
00:06:48
they grow their pie bigger and hire more
00:06:51
people so it turns out it's not a robot
00:06:55
displacing people directly is a robot
00:06:58
nonadopters are no longer competitive
00:07:01
and they are laying off people okay so
00:07:03
this is actually as a statistician this
00:07:05
is an interesting what I'll call self-
00:07:07
selection problem so the fact that
00:07:09
someone has adopted robotics is
00:07:12
indicative of the likely growth pattern
00:07:15
of the firms and faster growing firms
00:07:18
tend to hire more people is that the is
00:07:22
is from a selection bias story is that
00:07:25
the selection argument that's being made
00:07:27
uh that's a very good uh point so we we
00:07:30
actually looked at the type of people
00:07:31
that are being laid off that was going
00:07:33
to be my next question right so I think
00:07:35
I can talk about statistics the methods
00:07:37
we did but I think it's really probably
00:07:38
easier intuitive to understand but let
00:07:40
me just understand if I hire a robot if
00:07:43
I hire if I build a robot or utilize a
00:07:45
robot to do a specific type of task are
00:07:48
you suggesting that the firm hires more
00:07:50
people to do that type of task or the
00:07:53
firm hires more people in general to do
00:07:55
other tasks now this task is done more
00:07:57
efficiently this firm can produce more
00:07:59
units and now I need more people in
00:08:01
other jobs it's ladder your it's really
00:08:03
ladder the ladder okay but no one is
00:08:05
doing every everything like robots are
00:08:08
not doing everything a human can do yet
00:08:11
so you are robots are replacing part of
00:08:14
a Tas a human worker used to do but
00:08:16
increase the demand of the other task
00:08:18
that other that person used to do so if
00:08:22
you are a manager and you mostly looking
00:08:24
at quality assurance issues if that
00:08:27
robot's taking care of that you can do
00:08:29
more leadership you can do more people
00:08:31
management like you do other work that
00:08:33
robots you know can do yet I see now
00:08:37
your paper finds that the case for
00:08:40
employment this this this to me I I love
00:08:43
it when people find things that are
00:08:44
counterintuitive your paper finds is
00:08:46
what I've written down here that the
00:08:48
case for employment is reversed for
00:08:50
managers that is more robots can equal
00:08:53
fewer bosses so tell tell us about that
00:08:56
cuz that must have been one of the to me
00:08:58
intriguing finding I I think most people
00:08:59
would say no it's the I'll call it lower
00:09:02
level laborer that's going to lose their
00:09:04
job but managers someone has to make
00:09:06
decisions you're fine yeah that's also a
00:09:09
very intriguing finding we we thought we
00:09:11
got a WR at first so we really double
00:09:12
checked triple by dble checked and then
00:09:15
we find out why um it's are two effects
00:09:17
one direct effect right I mentioned the
00:09:20
monitoring technology can't figure out
00:09:23
what your employees are doing right you
00:09:24
think about a laptop well you can't even
00:09:27
figure out the every key employee can
00:09:29
type if you if you want to right and uh
00:09:33
uh you know in a warehouse situation
00:09:34
you're clocked in each box coming in
00:09:37
they know exactly what you what you've
00:09:38
done so you don't need a foreman or a
00:09:41
you know person who will see you're
00:09:42
actually doing the work because there's
00:09:44
all automatically captured and then your
00:09:46
performance reports automatically
00:09:47
generated so that part of the manager TI
00:09:51
is being directly substituted so you see
00:09:54
the very beginning within zero one year
00:09:56
of adoption you see a slight dip in man
00:09:59
managerial work but then after 2 three
00:10:02
year you see a really big downturn a
00:10:05
very sharp downturn so why is that right
00:10:07
that's not no longer a direct
00:10:08
substitution story anymore okay what
00:10:10
happened is that the composition of
00:10:13
Labor the employees at the firm changed
00:10:16
right so if you think about what robots
00:10:18
can do it turns out robots are really
00:10:20
good at doing the middle skilled work
00:10:23
part so they're they're not very good at
00:10:25
you know doing thought leadership you
00:10:27
know thinking about what to do
00:10:29
that's what humans are doing we need to
00:10:31
figure out our goals we need to you know
00:10:32
figure out what to do and execute it and
00:10:34
write plans are execute it and then but
00:10:37
robots are still not very good at some
00:10:39
of our like what our hands can do like
00:10:42
top tail SS are like you know like the
00:10:43
residual task somebody still need to you
00:10:47
know pick up a ball and put in a box
00:10:49
like you know in various shapes and form
00:10:51
our hands can handle infinite shapes
00:10:54
where a robot you got to train a lot of
00:10:56
time just hand handle one type of shape
00:10:57
so that versatility
00:10:59
although people are still are are
00:11:00
working on making a lot of Anis on this
00:11:02
area but the versatility of the robots
00:11:04
is not there yet you can train a very
00:11:06
narrow task but you cannot generalize
00:11:10
very quickly to lots of tasks and do you
00:11:12
think the you know if you've like the
00:11:15
explosion of Transformer models so the
00:11:18
ability to handle richer and bigger and
00:11:21
larger data sets in real time is even
00:11:24
going to threaten that that in some
00:11:26
sense robots are going to be better at
00:11:29
very let's call them fine motor tasks
00:11:31
and complex tasks and it's just a matter
00:11:34
of time even before let's call it the pi
00:11:36
expands to even more fine motor types of
00:11:38
jobs people are definitely working on
00:11:40
that but I do want to caution you that
00:11:43
for robotics area it's the data
00:11:46
generation it just not there as a AI
00:11:49
because AI you can simulate lots of
00:11:51
different kind of datas right like you
00:11:52
were pictures your videos automatically
00:11:54
there on the internet you can just grab
00:11:56
it right but if you want to train robots
00:11:59
but you got to train them in realistic
00:12:02
settings when humans are involved so why
00:12:04
can't you have let's as a form of by the
00:12:06
way the term you might call it a form of
00:12:08
supervised learning why can't you have a
00:12:10
bunch of humans doing a task I actually
00:12:13
have video cameras with extraordinary
00:12:15
maybe the person has sensors all over
00:12:17
her his their body and now all of a
00:12:19
sudden I've got this extraordinarily
00:12:21
Rich data set of movements and actions
00:12:23
and stuff and matter fact I'm going to
00:12:24
use AI to train the model to start using
00:12:28
robotics to replace those tasks is that
00:12:31
because companies are not doing these
00:12:33
types of I'll call it measurement
00:12:34
experiments in large scale or why can't
00:12:36
that be done I think you can do that but
00:12:39
again it's only if you're a very narrow
00:12:41
set of tasks unless that task has
00:12:44
extremely high value to the firm that
00:12:46
makes sense right and also most of the
00:12:49
robots you can't just plug in a robot in
00:12:51
a firm and then expect it to to be like
00:12:53
autonomously working on its own so
00:12:55
there's always a human monitoring or you
00:12:58
know working with robots and many lot
00:13:00
the cobots like a lot of the new
00:13:01
development and using uh you know more
00:13:03
advanced technique are cobots and these
00:13:05
cobots is usually a human machine
00:13:08
interaction in those scenarios right you
00:13:11
actually have to generate human robot
00:13:13
interactive data and that's something
00:13:15
you cannot just simulate you can't you
00:13:17
actually have to capture right you you
00:13:19
can't you can put a sensor all over me
00:13:20
or robots but you got to catch me every
00:13:23
everywhere and that's just hard to
00:13:25
capture right like you know you can't
00:13:26
just put in a you know car in a road
00:13:29
right I capture all this stuff but when
00:13:30
the hum involved is expensive right
00:13:32
capturing that data is really expensive
00:13:34
so in terms that data like I would say
00:13:36
robotics at least 10 years behind in AI
00:13:39
in generating the techniques that can do
00:13:41
what gener AI can do I see we're here on
00:13:43
the Wharton serus XM podcast series on
00:13:45
artificial intelligence sponsored by
00:13:47
analytics at Wharton and AI at Wharton
00:13:49
I'm here talking to associate professor
00:13:51
of operations information decisions ly
00:13:53
woo uh we're talking about uh Ai and
00:13:56
Robotics so let me ask you um how do
00:14:00
firms how should firms think about their
00:14:02
employees now or how should employees
00:14:05
think about the skill sets they need to
00:14:07
develop to kind of continue to thrive in
00:14:10
a world where I think we agree it's
00:14:12
unlikely robots are going to be doing
00:14:14
less overtime probably more overtime
00:14:17
what Can employees do and how should
00:14:19
firms think about managing them I think
00:14:22
one of the biggest thing human tend to
00:14:24
underestimate is that you we have a lot
00:14:27
of tacit knowledge about how work is
00:14:29
being done right so robots can only
00:14:32
observe you know a small part a lot of
00:14:35
our knowledge not codified for machine
00:14:38
to consume so the more knowledge you
00:14:40
have that's not codified the better off
00:14:43
you are because there's no way robots
00:14:44
can use that knowledge to replace you
00:14:47
but then again that's very abstract
00:14:48
right but if you've been if you have
00:14:49
been in a firm in the industry for a
00:14:51
long time you have a deep expertise
00:14:54
knowledge by the industry or by that
00:14:55
firm is always going to be good because
00:14:58
that know knowledge cannot be easily
00:14:59
transferable outside has anyone I don't
00:15:01
know I'm sure someone has studied this
00:15:03
are employees strategic like for example
00:15:07
maybe I you know it's the classic I
00:15:09
teach my student but I don't teach them
00:15:11
everything I know I teach my student you
00:15:14
know strategically enough but if I give
00:15:17
away everything then in theory maybe I
00:15:19
could be replaced by a robot is there
00:15:21
any studies that have been done on the
00:15:22
Strategic nature of employees I am sure
00:15:26
it has been done I think more will be
00:15:28
will have to be done because as you know
00:15:30
lot of PE lot of firm these days are
00:15:32
using human the TR robots I am sure
00:15:35
those kind of strategic Behavior will
00:15:36
become more and more uh prevalent there
00:15:38
are some literature in in the back in
00:15:40
the 6070s but I think that needs to be
00:15:41
updated I see so do firms have a choice
00:15:46
like is there any choice firms really
00:15:48
have or in some sense like you know I do
00:15:51
a lot of work in sports do I really need
00:15:54
some former 30y old sorry 50 60y old
00:15:57
baseball player throwing batting
00:16:00
practice why don't I just get a robot to
00:16:02
do that you know I study things in
00:16:05
high-tech do I really need some sort of
00:16:07
H human supervision anymore where I can
00:16:09
just have a robot do it are there any
00:16:12
specific like how should confirms really
00:16:15
compete in scale nowadays without
00:16:18
adopting Robotics are there's any or
00:16:20
another way to frame it is is there any
00:16:21
industry you think that is essentially
00:16:24
immune to it based on my study in
00:16:26
Canadian data I think he's not you
00:16:29
because adopters are really just killing
00:16:30
the competition and unless nobody in
00:16:32
your industry collect to decide not to
00:16:34
adopt robot I think you're then but you
00:16:37
you worry about new entrance coming in
00:16:39
with a robot so I think you have to be
00:16:41
forward looking at least and think about
00:16:43
how robots is going to be affect your
00:16:45
firm and productivity but the most
00:16:47
important thing is not like oh my God my
00:16:50
competitor do robots I need to do
00:16:51
immediately and I I I would caution that
00:16:54
because the value of adopting robots is
00:16:58
is really to help you understand your
00:17:00
existing firm processes and how to help
00:17:02
you understand which part can be
00:17:04
automated and which part can be
00:17:07
strengthened to complement that
00:17:09
automation but in fact I would say nine
00:17:11
out of 10 90% of value come from
00:17:14
studying that process understanding
00:17:17
where it can be automated where it
00:17:19
cannot be automated and how to
00:17:20
strengthen that relationship between
00:17:22
human a machine that collaboration so
00:17:24
this is not a a do robot buy some
00:17:26
expensive hardware and get some
00:17:29
consultant and you're done right this is
00:17:30
required people in your firm been doing
00:17:33
that process for a long time and and
00:17:35
from them they can tell oh you know this
00:17:37
part of process makes sense actually
00:17:39
yeah sure the robots can do it but I
00:17:40
don't think it's a good idea because
00:17:42
this is a very risky operation if
00:17:44
something happen to the you know it's
00:17:46
like a something happen to the person
00:17:48
it's very dangerous or high like Risk
00:17:50
situation it should not be should not be
00:17:52
automated even if you it can be done so
00:17:54
it's really understanding that process
00:17:56
deciding human decision year understand
00:17:58
understanding where it can be automated
00:18:00
where it shouldn't even if it can that
00:18:02
is where the value come from so let me
00:18:05
ask a a question about it's really a
00:18:06
two-part question but it's really the
00:18:08
same question which is so how big an
00:18:11
effect are we talking about here so a
00:18:13
firm hires a robot and then some measure
00:18:16
and this is the second part but it's
00:18:17
really the same question like first of
00:18:19
all what are the in our language as
00:18:21
academics what are the dependent
00:18:23
variables that people worry about so
00:18:25
let's let's imagine you have a model
00:18:26
that says if I adopt robots to a certain
00:18:29
degree to do a certain task my let's say
00:18:32
I care about sales or market share or
00:18:35
employee retention or employee
00:18:37
satisfaction what depend what outcome
00:18:39
variables do people tend to study when
00:18:42
they think about the adoption of robots
00:18:44
that's number one and second you know
00:18:46
how big an effect size are we talking
00:18:48
about here like if all the ways that a
00:18:50
firm could improve efficiency and
00:18:51
profitability is this in the top five
00:18:53
and how big an effect are we talking
00:18:54
about that's a great question so
00:18:56
obviously as a as you know a Economist
00:18:58
adti we capture any variable we have in
00:19:00
our you know Arsenal right Revenue
00:19:03
employee satisfaction individual
00:19:05
productivity performance right and
00:19:07
that's a great thing about the Canadian
00:19:08
data set is that we actually have a lot
00:19:10
of that data we look at employee
00:19:12
satisfaction there are their incentive
00:19:15
pay system so we were able to capture a
00:19:17
lot of this um so yeah it it is a very
00:19:20
important study in terms of what firm
00:19:22
should like what the effecta size is u i
00:19:25
want what did you find like how big an
00:19:27
effect are we talking about what happens
00:19:29
I'm just saying let's say you know the
00:19:31
way we always tend to think about it is
00:19:32
there's always three populations at
00:19:35
least the way I think about it is what
00:19:37
happens to firm let's even just say
00:19:39
we're thinking about profits or Surplus
00:19:41
what happens to the firm what happens to
00:19:43
customers and what happens to society
00:19:45
yeah let me just go by the revenue like
00:19:47
proct profit first right so if you think
00:19:50
about a profit maximizing firm right you
00:19:54
should you should like so you know I
00:19:56
adop one robot I got $10 and if my robot
00:19:58
is only costing $1 right that means I
00:20:00
should just buy more robot until I got I
00:20:04
got I reached exactly 10 robots and then
00:20:06
I got the same same same bio and some
00:20:08
right I put in $10 robots I get $10 of
00:20:10
output out right right that's should be
00:20:12
the problem math firm right so the fact
00:20:14
what we found is that robot is 10 times
00:20:17
that factor share so that's 10 times was
00:20:21
value so I you put in one you get 10
00:20:23
back that's what we found but there has
00:20:25
to be some sort of diminishing marginal
00:20:27
returns to robots it's like the second
00:20:28
robot's got to be worth maybe a little
00:20:30
bit less of course my cost might go down
00:20:32
as well yeah exactly so there's that but
00:20:34
I what I what I what I want to emphasize
00:20:37
This Not That robots responsible the the
00:20:40
whole 10 here nine of them come from
00:20:44
process Improvement I got it the study
00:20:46
that the stuff you do to make robot work
00:20:49
so the robots one because that's makes
00:20:51
sense you possibly maximize it one but
00:20:52
the nine is what you do with it I see
00:20:55
and what about for employees like um so
00:20:58
you could imagine one of two things one
00:20:59
is the employees that are still
00:21:01
remaining now that the firm's making
00:21:03
more money you could argue that the
00:21:04
employees are in better shape than
00:21:06
before I don't know what do you find in
00:21:08
terms of employees that's great uh what
00:21:10
we found is that is precisely when human
00:21:12
are working with robots like robots make
00:21:15
different kind of Errors than humans
00:21:18
right so ifite you and I work together
00:21:19
we could collude in a sense let's Shir
00:21:21
today right it'll be say our machine
00:21:23
died so let's just have a coffee right
00:21:25
I'm not saying it's going to happen but
00:21:26
we could right and a manager sitting far
00:21:28
away where cannot see you or know a
00:21:32
headquarter couldn't figure out okay
00:21:34
they cannot directly observe observe you
00:21:36
oh okay I guess the machine wasn't
00:21:38
working I I repair it right but when you
00:21:41
are working with a robot you can't do
00:21:43
that anymore all right because robots
00:21:44
are doing these things these kind of
00:21:46
Errors won't is less likely to occur but
00:21:49
because human or robot have different
00:21:51
errors that means when you're working
00:21:54
with a robot I can tell your performance
00:21:56
I can capture your performance more
00:21:58
accurately than you work with the human
00:22:01
so what happened is that firm adopting
00:22:04
robots also changed the performance pay
00:22:06
practices I see so they actually become
00:22:08
like the they're rewarding the high
00:22:10
performers more so than before like you
00:22:13
are actually good performers is not
00:22:14
because you were lucky or some other
00:22:16
stuff so good performers in some ways
00:22:18
should be happier with robots cuz in
00:22:21
some sense there's less measurement
00:22:23
error in their performance y that's
00:22:25
exactly what we found there's there less
00:22:28
team based uh promotions uh incentive
00:22:30
pay but mostly more individual based
00:22:33
performance pay to what extent is going
00:22:35
to happen in the future I don't know
00:22:36
because in we we see a lot of problems
00:22:38
in the individual performance pay too
00:22:40
but at least right now the the the swing
00:22:42
the pendulum is more to individually
00:22:44
based incentive pay and how about you
00:22:46
know policy makers so how should like
00:22:49
should this be a regulated market or is
00:22:52
this just you know I think most people
00:22:55
on the surface would say this is bad for
00:22:58
society because it's replacing paying
00:23:00
jobs for humans although on the other
00:23:02
hand there could be greater efficiencies
00:23:04
how should policy makers and Society
00:23:07
think about it I think um I think the
00:23:10
genie is out a bottle you can't put it
00:23:12
back anymore you can't just say let's
00:23:14
not use AI because the US doesn't use AI
00:23:16
the other country would and so as we we
00:23:19
have to be competitive the national
00:23:22
front as well in terms of Regulation I
00:23:24
think is really important we we actually
00:23:26
study the phenomena to Greater detail we
00:23:28
now we see gener AI people getting
00:23:31
worried but we really have to see
00:23:34
evidence does it actually make consumer
00:23:37
worse off right does it actually hurt
00:23:39
people because in my because you know
00:23:41
remember in that my robot study is
00:23:43
premis by saying robots will kill 40 to
00:23:47
50% of employees employment but we found
00:23:49
opposite we actually found employment to
00:23:51
have gone up quite a bit so we really
00:23:54
have to have large imperical evidence
00:23:56
first before we make any significant
00:23:58
policy changes because we don't want to
00:24:01
you
00:24:02
know we don't want to throttle all this
00:24:05
process we want to AI to
00:24:07
grow before we kill it completely just
00:24:10
without having evidence that it's going
00:24:11
to actually have significant I'm not
00:24:14
saying it will have it won't have harm
00:24:16
it it will but we need to watch and
00:24:18
monitor closely before we make any
00:24:19
decisions so maybe in the last minute or
00:24:22
so that we have could you tell me if we
00:24:24
were sitting here 10 years from now and
00:24:26
maybe we will be what so as we look back
00:24:29
on those 10 years what are we saying
00:24:31
this has been the big advance in Ai and
00:24:33
Robotics oh I think generative AI would
00:24:35
is for sure the most prominent evance in
00:24:40
NLP it pretty much wiped out all the
00:24:43
existing natural language processing
00:24:45
techniques out there this is de facto
00:24:47
replaced it so I think uh this entire
00:24:50
field needs to be revolutionalized uh so
00:24:53
in terms of in terms of I think the next
00:24:56
10 years we're going to see may not be
00:24:58
necessarily a technology advancement in
00:25:00
that area per se is really about
00:25:03
application how is that P marketing how
00:25:05
does it apply to schools and education
00:25:07
how does it apply to you know writing
00:25:10
how does it apply to reporting like as
00:25:12
Industries are starting to find novel
00:25:15
use as technology then that's where
00:25:17
we're going to see how amazing that just
00:25:19
like think of internet internet was
00:25:20
invented 1970s but we didn't really see
00:25:23
made how the potential of it 192000 with
00:25:26
right with with the internet with
00:25:29
e-commerce so I think uh generi may be
00:25:32
faster but it will still be a decade or
00:25:34
two until we see the real impact well
00:25:37
I'd like to thank my colleague uh
00:25:39
associate professor of operations
00:25:40
information decisions Lin moo for
00:25:42
joining me today on the Wharton sirusxm
00:25:44
podcast series on artificial
00:25:46
intelligence today we've been talking
00:25:47
about Ai and automation uh we have
00:25:50
plenty of episodes to go but Lynn thank
00:25:51
you for joining me today thank you for
00:25:53
having me this is so much
00:25:56
fun
