00:00:00
welcome welcome everyone to the first
00:00:03
episode of the analytics at Wharton and
00:00:05
AI at Wharton podcast series on
00:00:07
artificial intelligence my name is Eric
00:00:10
bradow I'm a professor of marketing and
00:00:12
statistics here at the Wharton School
00:00:13
I'm also Vice dean of analytics and I
00:00:15
will be the host for this multi-part
00:00:17
series on artificial intelligence I can
00:00:20
think of no better way to start that
00:00:22
series with both two of my friends and
00:00:24
two colleagues who actually run our
00:00:26
Center on artificial intelligence the
00:00:28
title of this episode is artificial
00:00:30
intelligence is here as you will hear
00:00:32
we'll do episodes on artificial
00:00:34
intelligence in sports artificial
00:00:35
intelligence in real estate artificial
00:00:37
intelligence in healthcare but I think
00:00:39
it's best to start just with the basics
00:00:42
so I'm very happy to have joined with me
00:00:44
today first my colleague cik hoser our
00:00:47
cardic is the John C Hower Professor um
00:00:50
at the Wharton School he's also as I
00:00:51
mentioned the co-director of our Center
00:00:53
on artificial intelligence at Wharton um
00:00:56
and normally I don't read someone's bio
00:00:59
first of all it's only a few sentences
00:01:00
but I think this actually is important
00:01:01
for our listeners to understand the
00:01:03
breath and also the practicality of
00:01:05
cic's work um his research examins how
00:01:08
AI impacts business and society and
00:01:10
something you'll hear about is that is
00:01:12
what our Center does there's kind of two
00:01:13
prongs second he was a founder of Yodo
00:01:16
where he applied AI to online
00:01:18
advertising and more recently and
00:01:19
currently to jump cut media a company
00:01:22
applying AI to democratize Hollywood and
00:01:25
he also teaches our courses on enabling
00:01:27
Technologies and AI business in society
00:01:29
so cardic welcome thanks for having me
00:01:31
Eric I'm also happy to have my colleague
00:01:33
Stephano poni uh Stephano is the
00:01:36
Sebastian sresky professor of marketing
00:01:38
here at the Wharton School he's also
00:01:40
along with cardic the co-director of our
00:01:42
Center on AI at Wharton and his research
00:01:45
examines how artificial intelligence and
00:01:46
autom automation are changing
00:01:48
consumption and society and similar to
00:01:51
cardic he also teaches our courses on
00:01:52
artificial intelligence brand management
00:01:54
and marketing strategies so Stephano
00:01:56
welcome thank you very much it's great
00:01:58
to be with both of you so maybe CTIC
00:02:00
I'll throw the first question out to you
00:02:02
um while artificial intelligence is now
00:02:05
the big thing that every company is
00:02:07
thinking about what do you see as well
00:02:10
first of all maybe even before what are
00:02:12
challenges facing companies how would
00:02:14
you even Define what artificial
00:02:16
intelligence is because it can mean lots
00:02:18
of things it could mean everything from
00:02:21
taking text and images and stuff like
00:02:24
that and kind of quantifying it or it
00:02:26
could be generative AI which is kind of
00:02:28
a same side of the coin but a different
00:02:30
part how do you even view what does it
00:02:32
mean to say artificial intelligence yeah
00:02:35
artificial intelligence is a field of
00:02:37
computer science which is focused on
00:02:39
getting computers to do the kinds of
00:02:41
things that human that traditionally
00:02:43
requires human intelligence and so what
00:02:46
that is is the moving Target so when uh
00:02:50
computers couldn't play uh you know say
00:02:53
a very simple game like um well chess is
00:02:55
not simple but uh you know maybe even
00:02:58
simpler board games maybe that's the
00:03:00
Target and then when you say uh
00:03:02
computers can play chess and when that's
00:03:04
easy for computers we no longer think of
00:03:06
that as AI but really today when we
00:03:09
think about what is AI it's again
00:03:11
getting computers to do the kinds of
00:03:13
things that require human intelligence
00:03:14
like understand language like navigate
00:03:17
the physical world uh like being able to
00:03:20
learn uh from experiences from data so
00:03:23
all of that really is included in in AI
00:03:26
do you put any separation between what I
00:03:29
call I'll call it tra maybe I'm not even
00:03:31
using the right words traditional AI
00:03:33
which again back in my old days we've
00:03:36
had our AI around like how do you take
00:03:37
an image and turn it into something how
00:03:39
do we take video how do we take text
00:03:42
that's one form of AI versus what what's
00:03:44
got everybody excited today which is
00:03:45
chat GPT which is a form of large
00:03:47
language model do you put any
00:03:49
differentiation there or that's just a
00:03:52
way for us to understand like one is
00:03:54
kind of like creation of data and the
00:03:56
other one's kind of like using it in an
00:03:58
application of forecasting language yeah
00:04:00
I I feel there is some distinction But
00:04:02
ultimately they're closely related
00:04:04
because what we think of as the more
00:04:06
traditional AI or predictive AI it's all
00:04:10
about taking data and understanding the
00:04:12
landscape of the data and to be able to
00:04:13
say in this region of the data let's say
00:04:15
you're predicting whether an um you know
00:04:18
an image is about uh Bob or is it about
00:04:22
uh Lisa and so you kind of say you know
00:04:24
in the image space this region if the
00:04:27
shape of the colors are like this the
00:04:30
shape of the eyes are like this then
00:04:31
it's Bob in that area it's Lisa and so
00:04:33
on so it's mostly understanding the
00:04:35
space of data and being able to say with
00:04:38
emails is it fraudulent or not and
00:04:40
saying which portion of the space does
00:04:42
it have one value versus the other now
00:04:44
once you start getting really good at
00:04:46
predicting that then you can start to
00:04:48
use those predictions to create and
00:04:50
that's where it's the next step where it
00:04:52
becomes generative AI where now you're
00:04:55
predicting what's the next word you may
00:04:57
as well use it to start generating text
00:04:59
and start generating sentences essays
00:05:01
and and novels and so on so Stephano let
00:05:04
me ask you a question so if one went to
00:05:05
your website on the Wharton website and
00:05:08
while by the way just for our listeners
00:05:09
um Stephano has a a lot of deep training
00:05:12
in statistics but most people would say
00:05:14
you're not a computer scientist you're
00:05:16
not a mathematician um what the hell do
00:05:19
you have to do with artificial
00:05:20
intelligence like what role does
00:05:22
Consumer psychology play in artificial
00:05:24
intelligence today like isn't it just
00:05:26
for us math
00:05:28
types
00:05:30
so if you talk to companies and you ask
00:05:32
them why did your analytics program fail
00:05:36
you almost never hear the answer because
00:05:39
the model didn't work because the
00:05:41
techniques didn't deliver it's never
00:05:43
about the technical stuff it's always
00:05:45
about people it's about lack of vision
00:05:47
it's about a lack of alignment between
00:05:49
decision makers and analyst it's about a
00:05:51
lack of clarity about why we do
00:05:53
analytics so I think that a Behavioral
00:05:55
Science perspective on analytics can
00:05:58
bring a lot of benef benefit to try to
00:06:00
understand how do we connect decisions
00:06:02
in companies to the data that we have
00:06:05
and so that takes both the technical
00:06:07
skills and the human insights the
00:06:10
psychology insights and so I think
00:06:12
bringing those together I find that has
00:06:15
a a lot of value and a lot of uh you
00:06:17
know potential insights that can a lot
00:06:19
of low hanging fruits in fact in
00:06:20
companies I think as a follow-up
00:06:22
question um you know we all read these
00:06:25
articles that say you know 70% of the
00:06:28
jobs are going to go away and you know
00:06:30
robots or automation or AI is going to
00:06:32
put me out of business should should
00:06:34
employees be happy with what's going on
00:06:37
in AI or the answer is it depends who
00:06:39
you are and what you're doing what are
00:06:40
your thoughts and then cardock I'd love
00:06:42
to get your thoughts on that including
00:06:43
the work you're doing at Jump cut
00:06:45
because we all know one of the biggest
00:06:46
issues in the current writer strike was
00:06:49
actually what's going to happen with
00:06:50
artificial intelligence so I'd love to
00:06:52
hear your thoughts from this psychology
00:06:53
or the employee motivation perspective
00:06:55
and then what are you seeing actually
00:06:56
out in the real
00:06:58
world the academic answer to any
00:07:01
question will be it depends it depends
00:07:02
but uh in my research what I've been
00:07:04
looking at is the extent to which people
00:07:06
perceive automation as a threat and uh
00:07:09
what we find is that often times when
00:07:12
tasks that are being automated by AI for
00:07:15
example um are tasks that have some kind
00:07:18
of meaning to the person that they are
00:07:20
Central to the way that they see
00:07:21
themselves for example in their
00:07:23
professional identity they can create a
00:07:25
lot of threat so you have psychological
00:07:26
threats and then you have the objective
00:07:28
threats of maybe jobs on the line and
00:07:30
maybe you'll feel happy about knowing
00:07:32
that I try out the professor job on some
00:07:34
of this scoring algorithms and we are
00:07:37
fairly safe for now at least well and so
00:07:40
cardic let me ask you and let me just
00:07:41
preface this with saying um You probably
00:07:44
don't even know about this 15 years ago
00:07:46
I wrote a paper with a former colleague
00:07:48
and a doctoral student about how to use
00:07:51
it was I didn't call it AI back then but
00:07:53
how to basically in large scale compute
00:07:56
features of advertisements and optim L
00:07:59
design advertisements based on a massive
00:08:02
number of features and I remember the
00:08:04
reaction I first thought I was going to
00:08:06
get rich I went to every big media
00:08:08
agency and said you know you can fire
00:08:11
all your creative people I know how to
00:08:13
create these ads using mathematics and I
00:08:15
was looked at like I had four heads so
00:08:17
can you bring us up to the year 2023 can
00:08:20
you tell us what you're doing at Jump
00:08:21
cut and kind of what role AI machine
00:08:24
learning plays in your company and just
00:08:26
what you see going on in the creative
00:08:28
world yeah yeah yeah and I'll I'll
00:08:30
connect that to also what you and
00:08:31
Stefano just brought up about Ai and
00:08:34
jobs and and exposure to Ai and so on um
00:08:39
I just came from a real estate
00:08:41
conference and the panel before I spoke
00:08:43
was talking about hey this artificial
00:08:46
intelligence it's not really
00:08:47
intelligence it just uh replicates
00:08:49
whatever in some data we this true human
00:08:51
intelligence is creative problem solving
00:08:53
and so on and I was sharing over there
00:08:56
that there are multiple studies now that
00:08:58
talk about what can AI do and cannot do
00:09:00
for example my colleague Daniel Rock has
00:09:02
a study where he shows that just llms
00:09:06
meaning large language models like Chad
00:09:08
GPT and before the advances of the last
00:09:11
6 months this is as of early 2023 they
00:09:14
found that 50% of jobs have at least 10%
00:09:17
of their tasks exposed to llms 20% of
00:09:19
jobs have more than 50% of their tasks
00:09:22
exposed to llm and that's not all of AI
00:09:24
That's Just large language models and
00:09:26
that's also 10 months ago and people all
00:09:29
so underestimate the nature of
00:09:30
exponential
00:09:32
change exponential and I've been working
00:09:34
with gpd2 gpt3 you know the earlier
00:09:37
models of this and I can say every year
00:09:39
the change is order of magnitude and so
00:09:43
you know it's coming and it's going to
00:09:44
affect all kinds of jobs um now as of
00:09:47
today I can say that multiple research
00:09:50
studies I don't mean two three four but
00:09:52
several dozen research studies that have
00:09:54
looked at AI use in multiple settings
00:09:57
including creative settings like writing
00:09:59
poems or problem solving and so on find
00:10:02
that AI today already can match humans
00:10:06
but human plus AI today beats both human
00:10:10
alone and AI alone so for me the big
00:10:13
opportunity with AI is we are going to
00:10:16
see productivity boost like we've never
00:10:18
seen before in the history of humanity
00:10:22
and that kind of productivity boost
00:10:24
allows us to Outsource the grunt work to
00:10:27
Ai and do the most creative things and
00:10:30
derive Joy from our work now does that
00:10:32
mean it's all going to be beautiful for
00:10:35
all of us no there are going to be some
00:10:38
of us who if we don't res skill if we
00:10:41
don't focus on having skills that
00:10:44
require creativity empathy teamwork
00:10:47
leadership those kinds of skills then
00:10:51
lot of the other jobs are going away
00:10:53
including knowledge work uh Consulting
00:10:57
software development you know it's
00:10:58
coming into all of these so just to
00:11:01
remind everyone this is Eric bradow
00:11:02
professor of marketing and statistics
00:11:04
here at the Wharton School and also Vice
00:11:05
dean of analytics we're here at the
00:11:07
analytics at Wharton AI at Wharton
00:11:09
podcast series on artificial
00:11:10
intelligence we're here with our first
00:11:12
episode of our at least 10p part series
00:11:14
and this one is artificial intelligence
00:11:16
is here and I'm talking to my colleagues
00:11:18
cardic hoser and Stephano pontoni so
00:11:21
Stephano could um something cardic
00:11:23
mentioned in his last thing was about
00:11:25
humans and AI matter of fact one of the
00:11:27
things I heard you say from the
00:11:28
beginning is it's not humans or AI it's
00:11:31
humans and AI how do you really see that
00:11:34
interface going forward is it up to the
00:11:37
individual worker to decide what part of
00:11:39
his her their task to kind of Outsource
00:11:41
is it up to management um how do you see
00:11:44
it you know kind of how do you see
00:11:46
people being even willing to skill
00:11:48
themselves up in artificial intelligence
00:11:50
how do you see this I think this is a
00:11:52
bigger qu biggest question that any
00:11:55
company should be asking not just about
00:11:56
AI right now frankly I think the biggest
00:11:58
question of of all in business how do we
00:12:00
use these tools how do we learn how to
00:12:03
use them there's no template nobody
00:12:04
really knows how for example generative
00:12:06
AI is going to impact different
00:12:08
functions we're just learning about
00:12:10
these tools and these tools are still
00:12:11
getting better so what we need to do is
00:12:13
to have some uh deliberate
00:12:15
experimentation we need to build
00:12:17
processes for learning such that we have
00:12:19
individuals within the organizations
00:12:21
tasked with just understanding what this
00:12:24
can do and it's going to be a um you
00:12:26
know impact on individual it's going to
00:12:28
be impact on teams on workflows how do
00:12:30
we bring this in in a way that we just
00:12:32
maybe not simply think of re-engineering
00:12:35
a task to get the human out of the
00:12:36
picture but how do we re-engineer new
00:12:40
ways of working such that we can get the
00:12:41
most out of people the point shouldn't
00:12:43
be you know human replacement and
00:12:45
obsolesence it should be human
00:12:47
flourishing how do we take this amazing
00:12:49
technology to make our work more
00:12:52
productive more meaningful more
00:12:54
impactful and ultimately make Society
00:12:56
better so cardock let me take what what
00:12:59
Stephano said in combining with
00:13:00
something that you said earlier which
00:13:01
was about the exponential growth rate so
00:13:04
my biggest fear if I were working at a
00:13:07
company today and you please I'd love
00:13:08
your thoughts is that someone's using a
00:13:11
version of chat GPT or some large
00:13:13
language model or even predictive model
00:13:15
some Transformer model and they fit it
00:13:18
today and they say see the model can't
00:13:21
do this and then two weeks later the
00:13:24
model can do this and so companies in
00:13:27
some sense create these absolute like
00:13:29
you just mentioned you were a real
00:13:30
estate well chat GP or large language
00:13:33
models AI can't you know sell homes they
00:13:35
can't build massive predictive models
00:13:37
using satellite data yeah maybe they can
00:13:40
can't today but maybe they can tomorrow
00:13:42
how do you in some sense try to help
00:13:45
both researchers and companies move away
00:13:47
from absolutes in a time of exponential
00:13:50
growth of these methods yeah I think our
00:13:54
brains fundamentally struggle with
00:13:56
exponential
00:13:57
change um and probably there is some
00:14:00
basis to this in you know studies people
00:14:02
have done on Neuroscience or human
00:14:04
evolution and so on but we struggle with
00:14:07
it and I see this all the time because I
00:14:09
have been part of that my work has been
00:14:11
part of that exponential change from the
00:14:12
very beginning when I started my PhD it
00:14:14
was about the internet and I can't tell
00:14:16
you the number of people who looked at
00:14:18
internet at any given point of time and
00:14:20
said nobody will buy clothing online
00:14:22
nobody will buy eyeglasses online nobody
00:14:24
would do this nobody would do that and
00:14:26
I'm like no no it's all happening just
00:14:28
wait to see what's coming and so I think
00:14:31
it's hard for people to Fathom I think
00:14:33
leadership as well as Regulators need to
00:14:36
realize what's coming understand what
00:14:38
exponential change is and start to work
00:14:41
now for you brought up previously and I
00:14:42
forgot to address it about like the
00:14:43
Hollywood writer strike now it is true
00:14:46
that today chat GPD cannot write a great
00:14:50
novel however when we work with writers
00:14:53
we are already seeing how they can
00:14:56
increase the productivity for writers
00:14:59
and how they can you know and in
00:15:01
Hollywood for example you know writers
00:15:04
are notorious because they're you know
00:15:06
writing is driven by inspiration and
00:15:08
you're expecting the draft today and
00:15:10
what's the excuse oh I'm just stuck uh
00:15:13
at this point I you know um and when I
00:15:16
get unstuck I'll write again and so you
00:15:18
can wait months and sometimes years for
00:15:21
the writer to get unstuck and now you
00:15:23
give them a brainstorming buddy and they
00:15:26
start getting unstuck and it increases
00:15:28
productivity and yes they're right in
00:15:30
fearing that at some point they're going
00:15:32
to keep interacting with the AI and keep
00:15:34
training the AI and someday the AI is
00:15:36
going to say you know what I'm going to
00:15:37
try and write the script myself and when
00:15:39
I say the AI is going to say that I mean
00:15:41
the AI is going to be good enough and
00:15:42
some uh executive is going to say why
00:15:45
deal with humans and and do that and so
00:15:47
I think we need to both recognize that
00:15:51
change is that fast and start
00:15:53
experimenting and start learning and
00:15:56
people need to start upping their game
00:15:58
and res Skilling and get really good at
00:16:00
using AI to do what they do and it's at
00:16:03
that reskilling is important stop
00:16:05
viewing this as a threat because what's
00:16:07
happening is you're standing somewhere
00:16:09
and there's a fast bullet train coming
00:16:11
at you and you're saying that train is
00:16:14
going to stop on its own no it's going
00:16:15
to run over you and the only thing you
00:16:18
can do and you have to do is get to the
00:16:20
station board the train and be part of
00:16:23
that train and help shape where it goes
00:16:25
all of us need to help shape where it
00:16:27
goes mhm yeah one example I like to give
00:16:29
is that um for I know 25 plus years I've
00:16:32
been doing statistical analysis in R and
00:16:34
of course over the last 5 to seven years
00:16:36
Python's taken a much larger role and I
00:16:38
always promised myself I was going to
00:16:39
Learn Python well I've learned python
00:16:41
now I stick it directly into chat my
00:16:43
rcode into chat GPT and I tell it to
00:16:45
convert it to Python and I'm actually a
00:16:47
damn good python programmer now because
00:16:50
chat GPT has helped me take structured R
00:16:53
code and turn it into python code that's
00:16:55
a great example and I'll give you two
00:16:57
more examples like that the head of
00:16:58
product at my company Jump Card media
00:17:01
had this idea for a script summarization
00:17:03
tool so what happens in Hollywood is the
00:17:06
vast majority of scripts written are
00:17:08
never read because every executive gets
00:17:11
so many scripts and you have no time to
00:17:13
read anything and you end up
00:17:16
prioritizing based on gut and
00:17:18
relationships he Eric's my buddy I'll
00:17:20
read his script but not this guy Stefano
00:17:22
who just sent me a script I don't know
00:17:23
him and that's how decision making Works
00:17:25
in Hollywood so the head of product
00:17:27
who's not a
00:17:29
he's actually a Wharton alumnist had
00:17:31
this idea for a great script
00:17:32
summarization tool that would summarize
00:17:35
things using the language and parland of
00:17:40
Hollywood and he had the idea to build
00:17:42
the tool but he's not a coder our
00:17:44
Engineers were too busy with other
00:17:47
efforts so he said while they're doing
00:17:48
that let me try it on chat GPD and he
00:17:51
built the
00:17:52
entire uh minimal viable product a demo
00:17:55
version of it on his own using chat GPT
00:17:59
and it's actually on our website on jump
00:18:01
cut media where our clients can try it
00:18:04
um and that's how it got built a guy
00:18:06
with no development skills I actually
00:18:07
demonstrated during this real estate
00:18:10
conference this idea that you've got you
00:18:13
post a video on YouTube You've Got
00:18:14
30,000 comments on YouTube and you want
00:18:17
to analyze those comments and figure out
00:18:18
what are people saying you want to
00:18:20
summarize it I went to chat GPD and I
00:18:23
said six steps First Step go to a
00:18:26
YouTube url I'll share download all the
00:18:28
comments Second Step do sentiment
00:18:31
analysis of that third step uh find the
00:18:34
comments which are positive and send it
00:18:37
to open Ai and give me the summary of
00:18:39
all the positive comments fourth step
00:18:41
negative comment send it open get the
00:18:43
summary fifth step tell the marketing
00:18:45
manager what you should do and give me
00:18:47
the code for all this it gave me the
00:18:48
code in the conference with all these
00:18:51
people I put it in uh Google collab ran
00:18:54
it and now we've got the summary and
00:18:57
this is me writing not a single line of
00:18:59
code uh with ch gbd it's not the most
00:19:01
complex code but this is something that
00:19:04
previously would have taken me days and
00:19:05
I would have had to invol aray and so on
00:19:07
and I can get that done and also Imagine
00:19:09
in real estate doing that about a
00:19:10
property or a developer or and you say
00:19:13
it doesn't affect real estate of course
00:19:14
it does absolutely it could it does and
00:19:17
I also showed them I uploaded four
00:19:19
photographs of my home nothing else four
00:19:21
photographs and I said I'm planning to
00:19:23
list this home for sale give me a a real
00:19:26
estate listing to post on Zillow that
00:19:29
would make people read it and get
00:19:31
excited to come and tour this house and
00:19:34
it gave a a great beautiful description
00:19:36
there's no way I could have written that
00:19:38
I challenged them how many of you could
00:19:39
have written this and everyone at the
00:19:41
end was like wow I was blown away and
00:19:43
that is something that is doable today
00:19:45
I'm not even talking about this is
00:19:46
coming soon so Stefano let me ask you um
00:19:49
I'm G to ask you and then I'll ask
00:19:50
cardic as well what's at the Leading
00:19:54
Edge of the research you're doing right
00:19:56
now so I want to ask each of you about
00:19:57
your own search and then I'll spend the
00:19:59
last few minutes that we have talking
00:20:01
about AI at Wharton and what you guys
00:20:02
are doing and hoping to accomplish so
00:20:04
let's start with your own personal
00:20:06
research like what are you doing right
00:20:08
now or another way I like to frame it is
00:20:09
if we're sitting here 5 years from now
00:20:11
and you have a bunch of published papers
00:20:12
and you've given a lot of big Podium
00:20:14
talks which I know you do what are you
00:20:16
talking about that you had worked on um
00:20:19
working on a lots of projects all in the
00:20:21
area of AI and so many exciting
00:20:23
questions because we never had a machine
00:20:24
like this a machine that can do the
00:20:26
stuff that we think is crucial to
00:20:29
defining what a human is this is
00:20:30
actually an interesting thing to
00:20:32
consider when you went back in time
00:20:34
maybe a few years and you asked what
00:20:35
makes human special people were thinking
00:20:37
you know maybe compared to other animals
00:20:40
we can think and now we ask what makes
00:20:42
human special and people see instead oh
00:20:45
we have emotions or we feel and so
00:20:47
basically now what makes us special is
00:20:48
what makes us the same as other animals
00:20:50
to some extent so you see how the world
00:20:52
is really deeply changing and I'm
00:20:54
interested in for example the impact of
00:20:57
AI for the pursuit of relational goals
00:21:00
or social goals or emotional um heavy
00:21:03
type of tasks where previously we never
00:21:06
had an option of engaging with a machine
00:21:07
but now we do and what does that mean um
00:21:10
what are potentially the benefits that
00:21:12
this technology can bring but also what
00:21:14
might be the dangers for example for
00:21:16
Consumer safety as people might interact
00:21:18
with these tools while experiencing
00:21:20
mental health issues or other problems
00:21:22
that's to me that's a very exciting and
00:21:24
important area and I just want to make a
00:21:26
point that this technology doesn't have
00:21:27
to be
00:21:28
any better than it is today for it to
00:21:31
change many many things I mean Kik was
00:21:34
saying rightly this is still increasing
00:21:36
exponentially and companies are just
00:21:37
starting experimenting with it but the
00:21:39
tools are there this is not a technology
00:21:42
around the corner is in front of ush so
00:21:45
CK what what are kind of the big open
00:21:47
issues that you're thinking about and
00:21:48
working on today yeah Eric there are two
00:21:51
aspects to my work one is slightly more
00:21:53
technical um and the other is focused
00:21:55
more on humans and societal uh
00:21:58
interaction with u with AI so on the
00:22:00
former side uh I'm spending a lot of
00:22:02
time thinking about biases in machine
00:22:03
learning models uh in particular a few
00:22:06
studies related to biases in text to
00:22:08
image models for example you go in and
00:22:10
you write a prompt uh generate an image
00:22:13
of a child studying astronomy if all 100
00:22:16
images are of a boy studying astronomy
00:22:18
then you know there's an issue and and
00:22:20
and these models do have these biases
00:22:22
just because the training data sets have
00:22:24
that so but if I get an individual image
00:22:27
how do I know is it's okay or not and so
00:22:29
we're doing some work on detecting bias
00:22:33
debiasing on automated prompt engineer
00:22:36
engineering as well so you've you state
00:22:38
what you want and we'll figure out how
00:22:39
to structure The Prompt for a machine
00:22:41
learning model to get the kind of output
00:22:43
you want so that's a bit on the
00:22:45
technical side on the human and AI side
00:22:49
uh most of my interest is around two
00:22:50
themes one is human AI collaboration so
00:22:53
if you look at any workflow in any
00:22:55
organization where AI now can touch uh
00:22:58
that workflow we do not understand today
00:23:01
what is ideally done by humans and what
00:23:03
is done by uh AI you know in terms of
00:23:06
organization design and process design
00:23:09
we understand historically like for
00:23:11
example how to structure teams uh how to
00:23:14
build Team Dynamics but if the team is
00:23:16
AI and humans you know how do we
00:23:19
structure that what should be done by
00:23:20
whom so have some work going on there
00:23:22
and the other one is around trust you
00:23:24
know AI is a huge trust problem today we
00:23:26
were just talking about the right
00:23:28
strike there's an actor strike and many
00:23:30
more issues coming up so what does it
00:23:33
take to drive uh human trust and
00:23:36
engagement with AI is another theme that
00:23:37
I'm looking at so maybe in the last few
00:23:39
minutes or so Stephano could you tell us
00:23:41
a little bit and our listeners here on
00:23:43
SiriusXM and on our podcast about AI at
00:23:45
Wharton and what you're hoping to study
00:23:49
and accomplish through a center on
00:23:50
artificial intelligence here at Wharton
00:23:52
and then we'll get card's thoughts as
00:23:53
well yeah and thank you for organizing
00:23:55
this podcast series for having us I
00:23:57
think it's a great opportunity to get
00:23:58
the word out the um initiative AI award
00:24:01
is just starting out we are you know a
00:24:04
bunch of academics working on AI
00:24:06
tackling AI from different angles for
00:24:08
the purpose of uh understanding what it
00:24:10
can do for companies how we can improve
00:24:12
decision making in companies but also
00:24:14
what are the implication for all of us
00:24:16
as workers as consumers and Society
00:24:18
broadly so we're going to try to you
00:24:21
know initiatives around education around
00:24:24
research around dissemination of
00:24:26
research findings and try to create a
00:24:29
community of people who um you know are
00:24:31
interested in these topics they're
00:24:33
asking similar questions maybe in very
00:24:35
different way and then we can learn from
00:24:37
one another and cardic what are your
00:24:39
thoughts about you know why you know
00:24:41
you've been involved with lots of
00:24:42
centers over the years what makes AI at
00:24:44
Wharton special and why are you so
00:24:45
excited to be in one of the leadership
00:24:47
positions of it yeah I think first of
00:24:50
all to me AI
00:24:53
is a one maybe not even a once a
00:24:56
generation but one several generation
00:24:58
kind of Technologies and it's going to
00:25:01
open up so many
00:25:02
questions that will not be answered
00:25:06
unless we create initiatives like ours
00:25:08
for example today computer scientists
00:25:11
are focused on creating new and better
00:25:14
models but they're focused on these on
00:25:17
assessing these models somewhat narrowly
00:25:19
in terms of accuracy of the model and so
00:25:21
on and not necessarily human impact
00:25:24
societal impact you know some of these
00:25:27
other questions at the same time
00:25:29
industry is affected by a lot of this
00:25:32
but they're trying to put the fire out
00:25:34
and they're focused on what do they need
00:25:35
to get done this week next week they're
00:25:37
very interested in the questions of
00:25:39
where will this take us three four years
00:25:41
later but they have to focus quarter by
00:25:43
quarter and I think we are uniquely
00:25:46
positioned here at Wharton in terms of
00:25:49
having both the technical chops to
00:25:51
understand those computer science models
00:25:53
and what they're doing as well as you
00:25:56
know people like Stefano and others who
00:25:58
understand the psych psychological and
00:26:00
the social science Frameworks who can
00:26:03
bring in that perspective and really
00:26:04
took a take a 5 10 15 25 year kind of
00:26:08
timeline on this and figure out what
00:26:10
does this mean for how organizations
00:26:13
need to be redesigned how does what does
00:26:15
this mean in terms of how people need to
00:26:17
be reskilled how do our own college
00:26:19
students uh need to be reskilled what
00:26:21
does this mean for regulation because
00:26:24
man Regulators are going to struggle
00:26:25
with this and while the techn teolog is
00:26:28
moving exponentially Regulators are
00:26:29
moving linearly and so they will need
00:26:31
that thought leadership as well so I
00:26:33
think we fill that Gap uniquely in terms
00:26:35
of those kinds of problems big open
00:26:37
issues that are going to hit us in 5 10
00:26:40
years but we are currently too busy
00:26:43
putting out the fires to worry about the
00:26:45
the big Avalanche coming our way well I
00:26:47
think anybody that has listened to this
00:26:49
episode will agree artificial
00:26:51
intelligence is here which is what the
00:26:53
title of this episode was um again I'm
00:26:55
Eric bradow professor of marketing and
00:26:57
statistics here at the Wharton School in
00:26:58
Vice of analytics I'd like to thank my
00:27:01
colleagues Stephano ponton and CTIC ker
00:27:03
thank you for joining us on this episode
00:27:05
thank you edic thank
00:27:11
you
