Episode Transcript
Transcripts are displayed as originally observed. Some content, including advertisements may have changed.
Use Ctrl + F to search
0:00
brain fog, insomnia, moodiness,
0:02
achy joints, weight gain.
0:05
Maybe you're thinking they're all just part of getting
0:07
older, or that's what your doctor tells
0:09
you. But Middy Health understands that
0:11
for women over 40, they
0:13
can all be connected. Hormonal changes
0:16
that happen during perimenopause and menopause
0:18
are at the root of dozens
0:20
of symptoms women experience, not just
0:22
hot flashes. and
0:30
FDA approved. Plus, they're covered
0:32
by insurance. A convenient telehealth
0:34
visit with a midi clinician can be
0:36
your first step to getting personalized care.
0:39
They'll tailor a treatment plan for your symptoms
0:41
and health history, so you can get back
0:43
to feeling great. 91%
0:46
of Middy patients get relief from
0:48
symptoms within just two months. When
0:50
your body changes, your care should too.
0:53
Book your virtual visit today
0:55
at joinmiddy.com. That's
0:58
joinmiddy.com. Hello,
1:12
everyone, and welcome to Talk Nerdy.
1:14
Today is Monday, March 25th, 2024,
1:19
and I'm the host of the show, Cara Santa
1:22
Maria. And as always, before we
1:24
dive into this week's episode, I do want
1:26
to thank those of you who make Talk
1:28
Nerdy possible. Remember, Talk Nerdy
1:30
is and will always be 100%
1:33
free to download, and that's because I rely
1:35
on the support of listeners just
1:38
like you. This week's
1:40
top patrons for the show, that's
1:42
those who pledge the top
1:45
dollar amounts to keep the
1:47
show going, include Daniel Lang,
1:49
David J.E. Smith, Mary Neva,
1:52
Brian Holden, David Compton, Gabrielle
1:54
F. Jaramillo, Joe Wilkinson, Pasquale
1:57
Gelati, Riva Keith, and Ulrika
1:59
H. The man. If.
2:01
You would like to pledge your
2:03
support on an episodic and basis.
2:06
All you have to do is
2:08
visit petrionand.com/talk Nerdy where you can
2:10
learn more about how to do
2:13
so. All. right? Let's dive
2:15
into this week's episode. So.
2:18
I. Had the opportunity to speak
2:20
with a doctor din Yellow. Ruth
2:22
See is a roboticist and professor
2:25
of Electrical Engineering and Computer Science
2:27
at Mit. She is the Director
2:29
of Computer Science and Artificial Intelligence
2:32
Laboratory there, and she's also a
2:34
member of the National Academy of
2:36
Engineering, the American Academy of Arts
2:39
and Sciences. and she's a Mutt
2:41
Macarthur Genius Fellow. See.
2:43
Coal wrote with Gregory
2:46
Mon a brand new
2:48
book called of the
2:50
Heart and That ship
2:52
our bright future with
2:54
robots So without any
2:57
further ado here see
2:59
his his officer then
3:01
jealous. Will Daniela? thank you
3:03
so much for joining me today. Thank.
3:06
You car! I'm so excited to spend the
3:08
time with you. As a
3:10
you've got a Bahram new book called.
3:13
The. Hearts and That ship
3:15
our bright future with robots
3:18
and you of course our
3:20
of roboticist and an electrical
3:23
engineer and a computer science
3:25
etc etc. I'm. Talk
3:28
to me a little bit before we
3:30
get into the book and sort of
3:32
our relationship with robots and robotics. Touch
3:35
me a little bit about your. Past
3:37
and your journey. How does
3:39
one become? A roboticists, a
3:42
computer scientist, an engineer. All
3:44
three together. Well
3:46
tara. I grew up in Romania.
3:49
At the time of great austerity,
3:52
And in my spare time I read a
3:54
lot of books. And. I reflected
3:56
on my fantasize about superpowers
3:58
sense far away the places.
4:01
But. I never imagined back then that
4:03
I would be one day working at
4:05
M I T. Building
4:08
robots and thinking about. A
4:10
bright future with machines. My
4:12
family emigrated to the United
4:14
States around that time I
4:17
was finishing high school. And
4:19
I was an undergraduate at the University of
4:21
Iowa. Where. I focused on
4:24
computer science, mathematics, and astronomy.
4:26
And. I often two trips to Chicago
4:29
where I enjoyed the lake, the cigar
4:31
same glass at the museum of our
4:33
it's the Music Life. At
4:36
the foods that it was really
4:38
exciting. Not toward
4:40
the end of my studies I had
4:42
an encounter that change the course of
4:44
my professional life. One
4:47
of the great minds in
4:49
computer science. At Computer
4:51
Scientists John Hop crossed. That
4:53
visited the university and gave
4:56
a distinguished talk. Now.
5:00
He's talk was I was
5:02
extraordinary in his talk. She
5:04
basically explains that computer science
5:06
will solved and he was
5:09
time for the grand applications.
5:11
You. Can imagine as an undergraduate
5:13
student studying computer science. I.
5:15
Found that a little bit unsettling.
5:19
Read a chance to speak with
5:21
him afterwards and what he what
5:23
he said was that a lot
5:25
of the problems that the founders
5:27
of the Fields post. Initially,
5:30
Was. Solve. They have solutions.
5:33
But that was a time of great opportunity.
5:36
And the great opportunity was
5:39
to build computing that interacted
5:41
with a physical world and
5:43
with people. To.
5:45
Think that for granted now, but
5:48
in the late nineteen eighties when
5:50
this conversation took place. Computation.
5:53
Was primarily for the digital world.
5:57
Oh I was very page. And.
6:00
I was excited to join
6:02
the john at Cornell University
6:04
to work on one of
6:06
these grand applications which was
6:09
robotics. And. Job It
6:11
was an extraordinary time. Because.
6:14
I'm we had. We.
6:16
Had the aspiration to solve these
6:18
these big problems will have to.
6:20
Didn't have the machines that could.
6:23
Translate. Our ideas are
6:25
into into solutions into fact.
6:29
And. And so. Simply.
6:32
Developing computational. Techniques.
6:35
Didn't work because the rogue robots
6:37
we had back then was a
6:39
bit big and bulky and am
6:42
had a very limited the repertoire.
6:44
Of doing things. And
6:46
so that's why am I? I sort of
6:49
realize that in order for us to make
6:51
real progress in the field of robotics, We.
6:53
Have to think about two aspects of the
6:56
field. We have to think about the body
6:58
of the machine and the brain of a
7:00
machine. Because. The machine. Will.
7:02
Only be able to do what the
7:05
bodies capable of doing. But.
7:07
Then the most local needs a
7:09
brain because without the brain without
7:11
the controllers and the algorithms without.
7:13
At the A I brain at the
7:15
machine will be an inert mechanism. And
7:18
so know that for robotics to
7:21
achieve it's potential, we really needed
7:23
to think about. The.
7:26
Body of the machine. And the
7:28
brain of the machine. And that's what I've been
7:30
doing ever since. You
7:32
know, I love that I think about so I'm
7:35
I'm a child of the eighties, so I just
7:37
turned forty this past. Year. Was born in
7:39
Nineteen Eighty Three. And when I
7:41
think about my childhood here in the
7:43
United States, I am sort of one
7:46
of the i'm an elder Milenio. You
7:48
could say I'm one of the sort
7:50
of last of the generation. Where are
7:52
my two boys growing up? Were definitely
7:54
not all mechanical. There were a lot
7:57
of sort of electric toy things that
7:59
will bow. Eerie powered. But
8:01
I did not really have computer
8:04
chips in my toys and I
8:06
have always. I grew up in
8:08
a world where there was this
8:11
distinction between the mechanical. Or
8:13
the electrified m, the
8:15
computational or the. You
8:17
know, like the I phone in
8:19
your pocket or the laptop sitting
8:21
in front of you. I always
8:24
saw like a big distinction between
8:26
those two, but children growing up
8:28
now probably don't even have that.
8:30
That. View
8:32
or that. understanding. Because computational
8:34
power is in. So.
8:37
Many things and I use toys
8:40
as an example because you can
8:42
very much see the evolution of
8:44
toys through these different era and
8:47
see that there quite a. Quite.
8:51
A kind of time capsule of
8:53
the technologies that were that we
8:55
were working within a commercial way.
8:58
and so I'm super curious. You.
9:00
Know from from your perspective.
9:02
Is do you think that that is
9:05
a good way to to look at
9:07
the these changes you know it or
9:09
our our children's toys. Even a good
9:11
sort of metric for that. It's
9:14
a very interesting point of view
9:16
param so as I think about
9:18
it. We observe that
9:20
when I was a kid I
9:22
was his by the porridge. By.
9:25
How that because magically. Abroad
9:28
colors in photos to life right
9:30
this before your eyes. And.
9:33
Now my daughter's live in the
9:35
world where. They. Have in
9:37
some sense the equivalent of
9:39
a digital polarizing the form
9:41
of a small smartphone and
9:43
they touch the smartphone. And
9:45
they get images on screens. To.
9:48
Respond to the pipe. it's a very
9:50
different world. When. I wanna do is
9:52
I want. That. Image to come
9:54
out of the phone and grow
9:56
into a physical structure. physical thing
9:59
that can. Interact with us
10:01
and help us and really transform how
10:03
we view the world. Yeah.
10:07
Sort of this. it's almost like
10:09
this by mode or this this
10:11
from know how to put it
10:14
into words. But we went from
10:16
the. Chemicals the mechanical
10:18
into this a booming era of
10:20
the computational. And now we're in
10:23
this world where there's a deep
10:25
synthesis between the two, where the
10:28
computational is no longer in the
10:30
ether. It's no longer us, just
10:32
on the internet. It's no longer
10:35
just behind the screen, but it's
10:37
becoming an everyday part. Of our
10:39
experience it's becoming integrated into places
10:41
we didn't ever had of imagine
10:43
it would be. Computation.
10:46
Is indeed everywhere is
10:48
in so many devices
10:51
and objects we interact
10:53
with Bots am. There.
10:56
Is a really interesting phenomenon.
10:58
Ah, that is happening right
11:00
under our eyes today. And.
11:03
This is. This. Is the
11:05
Ai phenomenon? So.
11:07
I is is growing
11:09
in capabilities right under
11:11
our eyes. We. We
11:14
are watching these capabilities
11:16
expand. And we
11:18
know a lot about Ai today. But.
11:21
You see most of the the
11:23
extraordinary Ai solutions we have today.
11:26
That. Exhibit: So much intelligence, so
11:28
much computational power are still
11:30
in the digital world. We.
11:33
Also have a physical world
11:35
where we have. Robots.
11:38
On the manufacturing sore. And
11:40
these robots are masterpieces of engineering
11:42
that kinda so much more than
11:45
people can. And. Yet
11:47
am the robots on the
11:49
manufacturing with floor or not
11:51
that intelligence? In. Fact: They
11:54
have to operate in a. In.
11:56
A rigid environments they perform
11:58
sake stop and they have
12:00
to be isolated some people
12:03
because they are. Big.
12:05
And bulky and dangerous to be
12:07
around. What? I would
12:09
like to see ah is emerging of
12:12
the two worlds. Emerging of
12:14
the the extraordinarily intelligence
12:16
digital A I world.
12:19
With. A. Potentially. Magical
12:21
World of physical machines.
12:24
And. I want all of this to
12:26
be intelligence. I wanted to create physical
12:29
intelligence. And. This is actually coming
12:31
to life today. And
12:33
some of the capabilities we
12:35
have are highlighted in the
12:37
book. Or. Maybe you
12:40
can help us sort of go
12:42
back to a more foundational conversation
12:44
because I think we we want
12:46
to make sure we're always working
12:48
from the same operational definitions. And
12:50
I think sometimes when we have
12:52
conversations we take we take things
12:54
for granted and we use words
12:56
without fully understanding what they mean.
12:59
So when we talked about of
13:01
robots, what are we actually talking
13:03
about? Is it just a machine?
13:05
How do we define what a
13:07
robot is as opposed to. You
13:09
know, is my vacuum a robot?
13:12
Or is my vacuum. A machine
13:14
or. Are they won in the same. Is.
13:16
It's a roomba it's the robots. If.
13:18
It's a manual vacuum cleaner. It's
13:21
not a robot. Basically, you can
13:23
think of robotics as putting computation
13:25
in motion. And you can think
13:27
of a robot as a programmable
13:30
mechanical device. That. Can. Exert.
13:33
Forces. By. Taking
13:35
input from the world, reasoning about
13:37
that, inputs to computation, and then
13:39
deciding what to do next. Okay
13:42
so when I think about going back to
13:45
sort of the children's toy example when I
13:47
grew up by I had a toy that
13:49
I loved called had Iraq Spend. So this
13:51
was a a teddy bear that had a
13:53
tape tape player and it's belly and it's
13:55
mouse. Would move along with the with the music
13:57
or with the. Stories of A Selling This was a
13:59
purely. Mechanical device like
14:01
kids today might have a
14:04
similar toy that reacts to
14:06
their voice or that you
14:09
know can take inputs. From
14:11
the real world. And make decisions
14:13
about what to. Do next.
14:15
So that's more of a of
14:18
a robot. Like there are robots
14:20
as children's toys. Yes, There.
14:22
Are some robots? Ah, as
14:25
two boys, Ah, you have
14:27
to think about how much
14:30
autonomy and how much computation
14:32
is processed by that toy.
14:35
And. Co op your toy from
14:38
your childhood. Was a
14:40
sophisticated automaton. Because the
14:42
same thing, no matter what. A
14:46
robot. Would. Have
14:48
to adapt to the situation
14:50
or that it finds itself
14:53
in. Like
14:55
imagine imagine walking into the
14:58
dentist's office. And. Imagine
15:00
being greeted by. And.
15:02
Little humanoids at toy.
15:05
A fad. Recognized as you
15:07
walk through the door. Greets
15:09
you and then asks you. Have
15:12
you lost your teeth recently? That.
15:15
Would be a kind of a robotic
15:17
toy. Because that's.
15:20
That. Instance that devise. Is.
15:22
Able to identify when you
15:24
come in. Is. Able to
15:26
come to you. It's able
15:28
to engage in an interaction
15:31
with you. And is
15:33
able to adapt as interaction.
15:36
To. You or to me or
15:38
anybody else. Walking. In the
15:40
in the space. Yeah,
15:43
you know, I think it's interesting
15:45
as a sort of lay consumer
15:47
to think about. where is that
15:50
dividing line? Like where is that
15:52
thresholds you know a lot of
15:54
us are use do the science
15:56
fiction. Versions or the Boston
15:59
Dynamics the versions of these humanoid
16:01
or these dog like or bear
16:03
like robots. We can instantly identify
16:06
the thats of robots or you
16:08
know, like you mentioned the roomba
16:10
We kind of all know that's
16:13
a robot vacuum or these large.
16:15
Manufacturing robots, That are
16:17
working on the line in let's
16:19
say, a car manufacturers. But where
16:22
is that sort of minimum threshold
16:24
for something to go from being.
16:26
Mechanical. Or purely
16:29
computational to into d kind of
16:31
definition of a robot. For.
16:34
The elements that we need.
16:36
Our perception. The. Census
16:38
of Robots. The brain.
16:41
The computation that the robots.
16:45
Executes based on perception
16:47
values. And then the
16:49
actual areas that have the ability
16:51
to do something in this in
16:53
the world. So these other three
16:55
important components that define a robot.
16:57
But I want to make a
17:00
comment about. About
17:02
what you said with respect
17:04
to humanoids and robot arms
17:06
and I want to observe
17:08
that. Most. Of the robots
17:10
that we have today. Are
17:12
inspired by the human form.
17:15
Would. Have been manipulators. That
17:17
look like arms. We have
17:19
humanoids. We also have robots.
17:22
That. Looks like boxes on wheels
17:24
like cars. And
17:26
will I think this is a very
17:28
limited view of what a robot could
17:30
be in our future. Our.
17:33
Bills, Environment. And.
17:35
Our. Natural. Environments
17:37
time in. So many more.
17:39
Safe. We. Have
17:41
fish said we have total said
17:44
we have cheaper saves. We have
17:46
to say we have table sage.
17:49
Imagine if we have the ability.
17:51
To. To take. Any
17:54
kind of shape in in the physical
17:56
world, in the built environment, or in
17:58
the natural environment. And create
18:01
a machine. Inspired.
18:03
By that shape. A. Machine that
18:05
could be optimized. To. Help
18:07
you. Do certain tests. To.
18:09
Help you may be and fits
18:12
right when you have to pay
18:14
attention or help you relax. When.
18:17
You can take it easier. So
18:19
this is part of what I
18:21
want to do out with my
18:23
work and this is an idea
18:26
I tried to convey in the
18:28
book. That's why we are. Really?
18:32
Am. Constrains:
18:34
In what we typically
18:36
associates with a robotic
18:38
devised. I think it's
18:40
important to expand our horizons and
18:42
think about all kinds of things
18:45
in our world. that could be
18:47
robots that could do things for
18:49
us and week. And and also
18:51
we should think about all the
18:53
materials that we used in order
18:55
to build this machine. Traditional.
18:59
Robots are built out
19:01
of metals, them and
19:03
has a plastics and
19:05
this leads to quite
19:07
as rigid. And
19:09
dangerous. Man. Mechanisms to
19:12
be around. But. In the
19:14
metal, the natural world, we have
19:16
so many more material, so why
19:18
not consider making machines out of
19:20
anything? And. In our lab
19:22
we have made robots out of paper
19:25
and plastic and silicone. We even made
19:27
a robot out the foods. And
19:30
so it's sad it's It's
19:32
really a almost liberating to
19:34
think about. All. These
19:36
different kinds of tools that we
19:38
can we can bring into our
19:40
lives in order to support us
19:42
with physical work. Yeah,
19:45
sort of this evolution that
19:47
we've we've been seeing unfolds
19:49
between. like you mentioned these
19:51
these rigid hard parts and
19:53
these more like soft robots
19:55
using these really innovative i'm
19:57
materials from from different mature
19:59
the scientists and also this
20:01
idea of. Bio.
20:04
Mimicry of of sort of looking. At
20:06
the natural world and seeing
20:08
how things move and change.
20:10
And and borrowing from. That it
20:12
reminds me of the. When
20:15
you look at sort of the history
20:17
of technology, you often see that we
20:19
are constrained by our own imaginations and
20:21
I think about the earliest cars right?
20:24
Like I think about like the Model
20:26
T and some of the very early
20:28
I'm experimentation with cars or you watch
20:30
a television show like Picky Blinders and
20:32
you're seeing them driving these these coaches
20:34
and they just look like a carriages
20:37
because that's what we always used to
20:39
before they basically took the horse away
20:41
and put it in the engine. but
20:43
that the shape and the functionality. Of
20:46
these early cars with based on what
20:48
we knew before and I think technology
20:50
often evolve that way we're so constrained
20:52
by well, that's how it worked for.
20:54
I feel the way we can make
20:56
it work instead of thinking well, no,
20:59
now we're using a radically new approach.
21:01
We don't have to be. Constrained by
21:03
the way it's always looked and acted to
21:05
us. What
21:07
exactly? Exactly. We don't have to
21:09
be constrained. Ah, we
21:11
can be very imaginative.
21:14
And. We can imagine. Turning.
21:16
Anything in so robot. Weekend.
21:19
For an orphanage. Signs, robots. Are
21:21
we can Now we can make robots
21:23
that flynn like fish. Are
21:26
we can make robots that seemed
21:28
like turtles? We can make robots
21:30
that run. Ah, Like cheats
21:32
us. And are there are
21:34
many. Extraordinary uses
21:37
of these specialized machines.
21:39
Me in some sense.
21:42
The. Idea of building a machine
21:44
in our own image. That.
21:46
Is Mars and obedience?
21:49
That. Really goes back to
21:51
antiquity. And. We have built.
21:54
So. Many technological marvels.
21:56
Throughout. History. Or
21:58
what makes this machine. Today.
22:02
Is the fact that they have. These bones.
22:04
That. Allow these machines to
22:06
understand the world. To.
22:09
Respond to it's to understand the
22:11
people around them. And. Adapt
22:13
to what they need doing.
22:16
And the idea of making a
22:18
robot as a humanoid. Has.
22:21
An additional and. In.
22:24
An additional aspect to it. Namely.
22:27
That's. A. Robot
22:29
that's built in our own image. Can.
22:32
Be seen as a kind of a
22:34
universal machine. That would be
22:36
able to build a lot of
22:39
different types of tasks or unfortunately,
22:42
Our hardware. Is not
22:44
quite good enough to deliver
22:46
these universal robots yet? And
22:49
our software is not quite good
22:51
enough to have these machines. And.
22:54
Executes tasked with grace and agility
22:56
that we are out we we
22:59
are you so that we seen
23:01
the natural world that the robots
23:03
the humanoid robots at look a
23:06
little bit mechanistic in how they
23:08
move in the world. We
23:10
would like to have machines that
23:13
are much much more. Ads are
23:15
much more graceful in how they
23:17
interact with the world and with
23:20
people. And so if
23:22
you have seen the Boston Dynamics
23:24
machines those muffins sense of these
23:26
are ads as. The
23:28
But Those machines are pre programmed to do
23:30
the tasks that they do. A
23:33
don't have quite an image.
23:35
They look amazing how they
23:37
do such extraordinary. Our
23:40
behaviors and pass. I love
23:42
how they dance that to
23:44
on the Rolling Stones. A
23:48
sister. My favorites are i'm bad
23:50
at but still those. It
23:52
took a lot of engineers to fine
23:54
tune all the parameters and get those
23:56
robots to do what they're doing. If.
23:59
You take. A robot that has been
24:01
pre programmed like that and you put
24:04
the so but in a different environment
24:06
and asked his robots to do a
24:08
decent task. And the robots
24:10
will have trouble. Or the
24:12
engineers will have to come in and and start
24:14
all over. With. Tasking and programming
24:16
the robots. And so
24:18
an alternative to this idea
24:20
of a universal robot is
24:22
to think about designing machines
24:24
that. Are. Specialized and
24:27
optimized for a small a set
24:29
of tasks. Then you
24:31
get a smart device. That.
24:34
Is able to do really well.
24:36
A smaller set of tasks. Not
24:39
all the tasks. And
24:41
so that's an alternative view for why
24:44
these machines might be helpful to us.
24:47
Because Why we don't quite
24:49
have this universal humanoids? Rosie.
24:53
Ah, we actually can have. A
24:55
lot of other specialized machines that
24:58
already do so much work for
25:00
us. We. Have robots
25:02
that help with backing him factory
25:05
side by side with people. Ah,
25:07
we have robots that monitor facilities.
25:10
We. Have robots that assessed. Ah,
25:12
Doctors and Hospitals. We have
25:15
robots that are exploring Mars
25:17
or exploring the depths of
25:19
the ocean. We.
25:21
Have even robots that know
25:23
courthouse. Is a
25:25
specialized machines that will do so much more
25:28
for us in the future. Yeah,
25:30
you know, and I think that
25:32
is the you make. Such An
25:34
important points that I've often grappled
25:37
with is this idea that somehow.
25:39
A superior robot is one that's
25:42
made in our image Or that
25:44
somehow because quote unquote, we're at
25:46
the top of the food chain
25:48
or weeks you know. it's such
25:51
an anthropocentric view that v the
25:53
best or the most attuned or
25:55
the most effective robots would be
25:57
those that look like us. That
26:00
assumes that we are the most
26:02
effective A we are the best
26:04
at certain task, but there's a
26:06
hubris and that that ignores exactly
26:08
the point that you made that
26:10
maybe the human form is not
26:12
the most adapted to milking a
26:14
cat our M C V, You
26:16
know there can be a much
26:18
more efficient and effective and even
26:20
towel friendly robot that milk the
26:22
cow that feel safer and more
26:24
comfortable and more welcoming them than
26:26
human hands for example, Exactly.
26:30
And. These. Specialized.
26:32
Robots These robots designs with
26:34
specific tasks in mind. Can.
26:37
Do so much for all. They
26:40
can take on physical tasks
26:42
and so that they can
26:45
save us invaluable time. But.
26:47
They can also enhance
26:49
our ability to perform
26:52
words and to get
26:54
entertains was unparalleled accuracy
26:56
for instance, Robots.
26:59
Can magnify our vision. They.
27:01
Can allow us to see with
27:03
clarity beyond the national capacity of
27:05
the human eye. Robots.
27:08
Can also. Help us
27:10
extend our reach. Try
27:13
to fight the throne of that
27:16
would go and investigate. And.
27:18
Or inspect the far away bridge or
27:20
the robots on Mars that are advancing
27:22
science. They. Can amplify
27:25
our strength like the
27:27
wearable exoskeleton. That. Are
27:29
beginning to get deployed. In.
27:32
Construction and for
27:34
physical rehabilitation. They.
27:36
Can refine our precision.
27:39
And so you see robots
27:41
You can imagine. robots. Taking.
27:43
On so many shapes the clothes you were
27:46
could be robots. The
27:48
I'm. Ah the drones
27:51
are are in these robots that
27:53
you can imagine them as kind
27:55
of mobile eyes that we throw
27:57
in the distance. To. Give us.
28:00
I saw. An. Extra
28:02
length in how we can perceive the
28:05
world. Out we have all
28:07
these the sense. Possibilities.
28:10
And am What I I'm
28:12
excited about and inspired about
28:15
is this idea that with
28:17
machines we can amplify a
28:19
human. Capabilities we
28:22
can transcend. Our
28:24
human limitations. And
28:26
we we can do so by
28:28
using machines to expand our reach
28:31
to give us strength and precision.
28:34
To. Help us defy gravity
28:36
like Ironman, By. Lifting up
28:38
in the third dimension. And.
28:40
To see really small things. And.
28:43
Even. Even to achieve the
28:45
kind of magical. Attributes.
28:48
That we typically associate with
28:50
story books and Six Sun
28:52
and Movies. And
28:55
and co the these these. Remarkable.
28:58
a machine. We.
29:01
Are not just augmenting our
29:03
physical capabilities were also redefining
29:05
the very boundaries of the
29:07
human potential. Hey. Hey there!
29:10
Did you know Kroger always gives you savings
29:12
and rewards on top of our lower than
29:14
low prices? And when you download the
29:16
Kroger app, you'll enjoy over $500 in savings
29:18
every week with digital coupons. And don't forget
29:21
fuel points to help you save up to
29:23
$1 per gallon at the pump. Want
29:25
to save even more? With a boost membership,
29:28
you'll get double fuel points and free delivery.
29:30
So shop and save big at Kroger today!
29:32
Kroger. Kroger, fresh for everyone. Savings
29:34
may vary by state. Restrictions apply.
29:36
See site for details. He
29:39
lived to see what's the weirdest
29:41
voice you've gotten Lucky Lucky. In
29:44
line as a deli against ah, my
29:46
dentist's office more than once actually do
29:48
I have to say. Yes, you do. In
29:51
the car before my kids C t
29:53
A meeting really has. Excuse me What's
29:55
the weirdest places you've gotten lucky? I never
29:57
win intel. Well there you have a duty.
30:00
What he anywhere playing at Lucky Landslides
30:02
know I com play for free right
30:04
now. Are you feeling lucky? Know businesses
30:06
are not easy clusters was. You.
30:11
Know you think about if we were
30:13
to survey a hundred. Years ago.
30:16
Citizens of our planet About.
30:18
The Future of robots. And
30:21
you might see a lot of.
30:24
Anxiety. A lot of fear.
30:26
You might also see a
30:28
lot of hope and a
30:30
lot of excitement. I think
30:32
that said day, we're living
30:34
in this really transitional period
30:36
where there's still so much
30:38
yet to discover and to
30:41
achieve. But there's also be
30:43
com a normalization of the
30:45
concept of robotics that people
30:47
are sort of familiar and
30:49
comfortable. Living among machines is
30:51
in a even Intelligent Machines.
30:53
Because we all have computers in
30:56
our back pocket, because we have
30:58
smart versions of things in our
31:00
homes, I think that there's a
31:03
comfort that has evolved yet. There's.
31:05
Still, this. Ah,
31:07
I'm. Disquiet. This
31:10
sort of. This
31:13
discomfort around this idea
31:15
of have. Too many
31:17
machines or machines becoming too intelligent or
31:19
being able to band together. this, you
31:22
know? Let's call it that,
31:24
The Terminator Fear and I hear
31:26
so much optimism in your voice
31:28
and in your writing when we
31:30
speak. Tell me a little bit
31:32
about some of your ethical concerns
31:35
or the places where you do
31:37
find caution in your optimism. So.
31:41
Let me say, I had usually.
31:44
When I tell people what I do,
31:46
I get two types of the accent.
31:49
Assessor. Some. People get
31:52
nervous and they miracle of
31:54
kindness. And the app in
31:56
the robots will take over their jobs.
31:59
And yeah, The group gets. Was excited
32:01
and asks me how much more
32:03
they can offload to machine. Can.
32:06
The machine for my laundry. When.
32:08
Will my car be totally self driving?
32:11
From. I personally believe that everyone
32:13
stands to benefit from a I,
32:16
but it's important to understand the
32:18
fears of that first group at
32:20
the Fears About Robots. And.
32:22
Coming to get us. And
32:25
it's also important to provide
32:27
solutions and perspective on how
32:29
to see things differently. And.
32:31
This threat to the understanding that robots
32:34
are tools. They. Are
32:36
incredibly powerful, Tools. But.
32:39
Like any other tool. They're.
32:41
Not inherently good or bad.
32:44
They. Are what we choose to do
32:46
with them and I believe we can
32:49
choose to do incredible things. And
32:51
this is. This is my
32:54
vision. This is my aspiration
32:56
for a future with robots.
32:58
Mine is a vision of
33:00
a world in which bespoke
33:02
machines assists people with both
33:04
basic and complex tasks, and
33:06
that honestly, the future is
33:09
beginning to mature in my
33:11
lab and in the lives
33:13
of my friends and. And
33:16
colleagues at other universities
33:18
and institutions. Around
33:21
the world and add some for
33:23
for sinking companies. That
33:26
crime. That. Alongside
33:29
the optimistic vision, we also
33:31
have to consider scenarios where
33:34
the integration of machines into
33:36
society introduces complex, technical, social
33:39
and ethical dilemma and add
33:41
the substance. Now.
33:44
One. Reason I
33:46
am are extraordinarily optimistic.
33:49
ah is because knowing
33:51
how. Machines. Words
33:53
I can tell you. That. There
33:55
are limits to what these machines can do. And
33:58
these limits or defy. And by the body
34:01
of the machine and the brain of the
34:03
machine. A robot
34:05
is limited by it's says the
34:07
core body which consists of the
34:10
save the sensor, the actuators, the
34:12
computers, data sources, storage power, And
34:15
the robot is also eliminate that
34:18
limited by it's been by the
34:20
software part that consists of all
34:22
the low level controllers, the algorithms,
34:25
and the ai systems. That.
34:27
Guy's it's most. Let
34:29
me give you an example. A self
34:31
driving car will not come in
34:34
your home so vacuum your carpet
34:36
and a robotic vacuum cleaner will
34:38
not be to access. So
34:40
we already have limits. To
34:44
watch these machines can do and
34:46
these limits are consolable by people.
34:49
Because. We designed them. We
34:51
build them as tools now.
34:53
of course we introduce these
34:55
machines to help us with
34:58
works. One. That's a
35:00
concern is. Are we putting
35:02
people out of them? Words. Are
35:04
the machine taking over the
35:06
jobs? And
35:09
the way I think about it. Is.
35:11
That. Jobs. Are very
35:13
complex, And. Jobs usually
35:16
consists of multiple tasks. In
35:19
a typical job, we. Am
35:21
we do predictable physical
35:24
work with the unpredictable
35:26
physical work with a
35:28
data tasks we apply
35:30
expertise. We. Interact with
35:32
with others, with stakeholders
35:34
and colleagues. Into
35:36
this, machines are primarily
35:39
useful for. Predictable.
35:41
Physical Work. They're. Not
35:43
so good at unpredictable physical work.
35:45
This is why it's much easier
35:47
to send a robot to Mars
35:50
then to get a robot to
35:52
clear your dining table after your
35:54
former. Squirrel Am
35:57
so you'll see it. There are
35:59
limits. To walk machines can do.
36:02
Birch. Are, but the
36:04
capabilities are gradually getting
36:06
better. And as the
36:09
capabilities are getting better, I believe
36:11
that. They. Can take on
36:13
that the kind of the low level
36:15
or tasks that we have on our
36:17
plates no matter what we do. And
36:21
that means that we can focus
36:23
more time on the more interesting,
36:26
more can't cognitive tasks. The.
36:29
Other aspects to jobs is that
36:31
there are many job for which
36:33
we do not have enough people.
36:36
To carry on those jobs. It.
36:39
Means that now we can fulfill the
36:41
past That. Otherwise would be in
36:43
a queue would be waiting a long
36:45
time for these tasks. And let me
36:48
give you an example. And.
36:50
I collaborate with and.
36:53
Who is a very young. Forwards.
36:55
That and they're thinking
36:57
doctor who specializes in
36:59
physical therapy. And
37:02
in working with a physical therapist.
37:05
I have learned that in fact,
37:08
The physical therapists, Spends.
37:11
A lot of time not applying
37:13
their expertise. They spent about a
37:15
third of that time. Going
37:18
to the hospital room. To. Pick
37:20
as a patient and say the patience
37:22
to the gym. They spend about a
37:25
third of the time. Working.
37:27
with a patient in the gym. And
37:29
then they spend another third of the time.
37:32
Taking. The pace and back to the
37:34
hospital room. And. For
37:36
now, imagine. If we had
37:38
a robotic wilshere. That. The
37:40
patients could use to travel autonomously
37:43
to the gym. That patients
37:45
could spend. Three times as much time.
37:48
With a doctor. And I would
37:50
be very beneficial for the patience. And
37:52
the doctor would have. Much. More
37:55
interesting ways. Of
37:57
applying expertise for longer period.
38:00
The time. You. Know
38:02
it's interesting because I think that
38:04
one of the things that. I.
38:06
Struggle with for coming
38:08
from a psychology perspective
38:11
is the integration of
38:13
these. Ideas. In
38:15
a global sort of
38:18
capitalist. Climate So you
38:20
know you mentioned before at that
38:22
these are tools, right? And I
38:25
think it's such an important point
38:27
to make that. I. Think
38:29
many people's fears about
38:32
the capabilities of robots
38:34
or less. I.
38:36
Would say they're They're not so
38:39
much fears about the robots themselves,
38:41
on their fears about the darker
38:43
side of humanity and how humanity
38:45
might choose to use these tools.
38:47
And so when we think about
38:50
something like taking over a simple
38:52
job that sounds great in theory
38:54
I would love it if the
38:56
sort of low skilled. Or like
38:58
low. Paying or like the
39:00
aspects of our jobs that
39:03
are the most tedious were
39:05
taken over by robots, but
39:08
it also requires I think
39:10
a sort of societal level
39:13
change that values. Certain.
39:15
Parts of the workforce. In a
39:17
way that I'm not sure that we
39:19
do as. A society and
39:21
as the governing body and I
39:24
think to even piggyback off of
39:26
that. When we talk
39:28
about things like a i one
39:30
of the big take a ways
39:32
I noticed at the beginning of
39:34
this sort of a majority. Or
39:37
Dolly Revolution. Was that
39:39
we think of a as this incredible
39:41
tool to help us? You. Know
39:43
I don't know. Do lit sources
39:45
to write or dissertation? swords? Or
39:48
help us with the sort of
39:50
tasks that are frustrating. but what
39:52
we end up seeing in social
39:55
media or out in front of
39:57
our eyes is a I making.
40:00
Poetry and Making Art. And
40:02
I don't wanna live in a world where
40:04
the robots are taking over the art and
40:06
poetry. I want to live in a world
40:08
where the robots are cleaning my house so
40:11
i have more times the make art and
40:13
poetry but the only way that gonna work
40:15
is if we value those things. In.
40:17
A society and I can afford to live
40:19
that way. And so it's almost like this
40:22
is one piece of a much larger puzzle
40:24
that I feel we. May
40:26
not be spending enough time talking
40:28
about. I'm curious how you feel
40:30
about that. Yeah. You
40:32
make some very, very important points.
40:34
Am I too want to say
40:37
something about jobs? I do want
40:39
to add something to our ah,
40:41
our discussion on jobs. I want
40:43
to. Remind
40:45
you that economists estimate
40:48
that over sixty percent
40:50
of jobs perform today.
40:53
When. Non Existence The for ninety. Moon.
40:56
Interests And so echoing that I
40:58
see the rise of robots as.
41:01
Promising. To create all kinds of new
41:04
jobs. Robot. Operator.
41:07
Robots and rates. Data analysts.
41:10
User experience Designers You can imagine
41:12
a lot of new jobs. Some.
41:15
Created Jobs. Some. Technical
41:17
jobs. Some. As they
41:19
call jobs. That do not
41:21
exist yet. Son. Not
41:24
worried that aren't humanity is not
41:26
going to figure out what to
41:28
do because we're in the thing.
41:31
These extraordinary tools. And
41:33
the other thing I want to say
41:35
that were much better. At. Thinking
41:37
about what will go away.
41:40
Then. Imagining what will com. Yeah,
41:43
we definitely have a bias there a
41:45
way of, for example, So let me give
41:47
you a concrete example. In.
41:50
Around the year is two thousand,
41:52
which I remember very well. Out
41:55
there was a computer revolution as
41:57
a digital revolution. And
42:00
we were talking. About all sorts of
42:02
things, But. We're not talking
42:04
about social media. We, We're not
42:06
talking about smartphones. We're not talking
42:09
about cloud computing. And.
42:11
These three social. Media, Cloud
42:13
Computing and thus the
42:15
smartphone. App came into
42:17
being in two thousand and seven. And
42:21
these new technologies have enabled
42:23
so many new jobs. For.
42:25
Creators as well as for
42:28
geeks. Right going.
42:30
Back to your question about a
42:32
I. How do
42:35
we think about Ai? Because
42:37
yeah, we can. Now we
42:39
can not imagine a I
42:41
powering the brains of the
42:44
robots, but we can also
42:46
think about all the extraordinary
42:48
capabilities that the generative ai
42:51
and foundational and models have
42:53
unleashed. And. When.
42:55
I think about that. I
42:58
can see multiple the friends.
43:01
At. Suggest Harrys for Humanity and
43:03
I know exactly how it's
43:05
going to all play out.
43:07
But there are many, many
43:09
different ways in which ah,
43:12
we can go. And
43:14
so let me give you some are
43:16
some examples. In.
43:19
Today's world when we work of out
43:21
when we work was a I We
43:23
essentially build larger and larger. Machine.
43:26
Learning models. And we
43:28
put our increasingly more data
43:30
into these models. Out
43:32
so that these models become a
43:35
very capable so that they are
43:37
sounds and amaze us in all
43:39
kinds of. Digital
43:41
tasks. And.
43:44
This is a possible way forward. But.
43:48
There's something really interesting about
43:51
this approach. Is you
43:53
believe that the future is all
43:55
about building these large models and
43:57
putting data ah into these models?
44:01
We. Have. We.
44:03
Have a danger wheat we
44:05
have. We have
44:07
the possibility of creating these huge
44:10
machine so we do not fully
44:12
understand. The. Large
44:14
A models are black boxes
44:16
is is really impossible to
44:19
explain how they reach their
44:21
decisions. So we
44:23
may get to the point.
44:26
Where. We we have
44:28
extraordinary machines. That and
44:30
that's that. develop. An.
44:32
Extraordinary capabilities. But
44:35
the risk here. Is that?
44:38
We. Will not be able to explain
44:41
how these machines work. And
44:43
this is kind of like
44:45
killed pilots flying sophisticated aircraft
44:48
by instruments alone. Without.
44:50
Understanding the principles of flight,
44:52
And. Without knowing how to navigate
44:54
in the absence of the instruments. And
44:58
so to me, working with
45:00
intelligent machines requires that we
45:02
understand what. They do. So
45:06
that's impossible for it.
45:08
And. And this could
45:10
be. this could be true. But.
45:13
What if we're wrong? What
45:16
if we have a different kind
45:18
of future? Where are
45:21
these projects and will prove
45:23
flaws in practice? And.
45:25
Essentially will have become
45:28
increasingly reliant. On
45:30
machines that we do not understand
45:32
and we have no idea how
45:34
to fix when these machines date
45:36
And frankly, this is my nightmare.
45:39
I don't think about.
45:42
The. Hollywood vision of suddenly same
45:44
chin machines and that decides to
45:46
extinguish off because I don't see
45:48
that as a possibility. But I
45:51
do worry that we will end
45:53
up with a massive complex system
45:55
would depend on and we can't
45:57
comprehend. And then we will.
46:00
Have. Enormous. Amounts
46:02
of discarded technologies and
46:04
electronic waste. That
46:06
will not be a good stuff for it. Not.
46:09
What I like to see as we
46:11
move forward. Is. An. Intelligent
46:14
Machines as smart as tools
46:16
for people. I would like
46:18
to see a future in
46:20
which robots and A I
46:22
Solutions. Are. Trustworthy, Their.
46:25
Explainable, they can be soda five.
46:27
Hour their well understood than
46:29
their capabilities are well understood.
46:32
And are these robots and they
46:35
I solution. Will. Then assist
46:37
us safely. And. Was
46:39
in a way that allows us
46:41
to trust them. And they
46:43
will assist us with a lot of
46:45
physical tasks. A lot of cognitive tasks.
46:48
They will lift humanity as a whole
46:50
to a higher standard of living. And
46:53
perhaps they give out. They will even give
46:55
us superpowers. Like the heroes
46:57
from our childhood books. You.
47:01
Know it's it's it's such
47:04
an interesting concept. This idea
47:06
of cause and effect versus
47:08
correlation. This idea of the
47:10
statistical understanding of sort of
47:13
what comes first and amplify
47:15
something else. And I think
47:17
one of the things that
47:19
I continue to grapple with
47:22
as I am a cautious.
47:24
Techno. Optimists. Is
47:27
that? There. Are
47:29
deep social inequities and there
47:31
are supremacist forces at work
47:34
and authorities authoritarian forces at
47:36
work that really are striving
47:38
to keep people divide. It
47:40
did that are striving to.
47:43
Let's say keep for example,
47:45
poor people, people of color,
47:47
Imprisoned. Women
47:50
in a subservient role. There
47:52
are. These kind of invisible.
47:55
You. Know social forces at play,
47:57
And I think that technology can be
47:59
used as the tools to continue to
48:01
divide, but it can also be used
48:04
as a tool for it. He
48:06
gala terry and purposes to
48:08
unify. And to you know
48:10
that the sort of a rising
48:12
tide lifts all boats. And
48:15
my concern sometimes is. Admin.
48:18
Maybe it's my cynicism or
48:20
my pessimism is that. Advances
48:23
in technology especially in
48:25
a and robotics will
48:27
deep in social divides.
48:30
Whereas what I'm hearing from you
48:32
is that may be. With
48:35
these advances. We will see
48:37
more movement towards any gala
48:39
terry and future. And
48:41
it's hard to know, right? because we don't
48:43
know what it is. It the human activity.
48:46
That. Is. Affecting.
48:48
The. Or
48:50
that the human I guess interests
48:52
that are affecting the technological advancements
48:55
or as the technology itself in
48:57
turn having an effect on our
48:59
humanity and it's probably both the
49:01
you know, it's that kind of
49:03
a chicken and egg scenario. but
49:06
I'm curious about how you grapple
49:08
with those questions about sort of
49:10
the darker side of humanity and
49:12
technology being a tool for our
49:14
purposes that can be used. For
49:17
good or or deep deep evil.
49:19
Corrupt. The spread of intelligent
49:21
machines will make our lives easier.
49:24
But. Many of the roles that he
49:27
can play. Will
49:29
displace work done by humans today.
49:32
So. We really need to
49:34
anticipate and respond to the
49:36
economic inequality this could create.
49:39
At the same time we
49:41
have some moms at technical
49:44
and technological issues because especially
49:46
with I we have a
49:48
lack of interprets ability. And
49:52
dislikes of interprets ability could
49:54
lead to significant issues around.
49:56
Trust and Privacy. And
49:59
it needs. Dress these issues. We
50:01
need to develop an ethics and
50:03
legal framework for how to use
50:05
intelligent machines for the greater good.
50:09
At the same time,
50:11
the spread of misinformation
50:13
and disinformation remains a
50:15
huge concern. And
50:17
am social media platforms are
50:19
struggling with a response. as
50:23
the face get better and more
50:25
widespread. This. Problem becomes
50:27
even more urgent for national
50:30
security. And
50:32
because as we gather more
50:34
and more data feeds into
50:37
our intelligence ai systems and
50:39
not only. Is the
50:41
risks to privacy higher? But.
50:43
So are the opportunities for
50:46
authoritarian governments to leverage these
50:48
tools to curtail freedom and
50:50
democracy around the world. But
50:54
the way I see it is that
50:56
these problems aren't like the pandemic. We.
50:59
Know they're coming. And weekend
51:01
out to find solutions at
51:03
the intersection of policy, technology
51:06
and business in advance. And
51:08
so I can tell you
51:10
that we already have. Solutions.
51:13
For the bias problem, We
51:16
can look at how a machine model
51:18
is. Trains were kind of data. The
51:20
machine is trained on. And
51:23
weekend the biased or beta so that
51:25
the function of them the side of
51:27
the machine learning model is the biased.
51:30
We. Can also
51:32
starts. At lot
51:34
of marking trusted information sources.
51:37
So that when we get a
51:39
piece of information, we know whether
51:41
it comes from a trusted source
51:43
or not. I think
51:45
it's important. For. Us to
51:48
know whether we interact with machines are
51:50
people. When. We interact. And
51:53
so you see, there
51:55
are already beginnings of
51:57
technological solutions that cancelled.
52:00
Make things better. And
52:03
this is. Part. Of the
52:05
reason. Why I'm hopeful other
52:07
solutions will not be entirely.
52:10
In. The realm of technology. We
52:13
really have to think about how
52:15
the solutions in the hands of
52:17
people impact society and that means
52:20
we need to include policy we
52:22
need to include the i'm the
52:24
business side to have comprehensive discuss
52:27
sense for how to deploy these
52:29
tools safely. What needs to be
52:31
done before a tool can be
52:34
safely are put in the hands
52:36
of people. but we are making
52:39
a lot of progress. Every
52:41
day we are in at we
52:43
are understanding more about. The.
52:45
Flaws of our systems and we're
52:47
coming up with solutions to address
52:49
them. If. Anything I
52:52
want to say as
52:54
that you were asking
52:56
about whether these technologies.
52:58
Are. Limited to the elite.
53:01
And I want to highlight that
53:04
the trajectory of. Robotics.
53:07
Much. Like the trajectory of
53:09
the smartphones. Is. Really
53:11
poised to transition from
53:13
rarity and luxury so
53:15
ubiquity and accessibility. And
53:19
I see that initially
53:21
the robotic technology particularly
53:23
out with respect to
53:25
advance the timeless robots
53:27
like building a self
53:29
driving car ah remains
53:31
expensive and somewhat limited
53:33
to industrial applications and
53:35
nice markets. And
53:37
this is akin to the early
53:40
days of smartphones ah which were
53:42
very costly and considered luxury items.
53:45
But as robotic technology
53:47
advances the initial high
53:49
costs associated with with
53:52
research and development and
53:54
the novelty of the
53:56
technology will decrease because
53:58
over time manner. Acting
54:00
processes will improve, the economies
54:02
of scale will be achieved
54:04
and the cost of producing.
54:07
These machines will decrease. And
54:10
so I think of this
54:12
trend coupled with competitive pressures
54:14
will make robotic technology much
54:17
more affordable and accessible to
54:19
the public. And
54:21
I believe that this mirrors
54:23
how we have experienced a
54:25
smartphone trajectory. Where.
54:28
It was the advancements in
54:30
technology and mass production. That.
54:32
Reduce the cost and made
54:35
smartphones accessible to of the
54:37
ass population on our planet.
54:41
Here you know it's it's, it's I
54:43
feel like this is almost like a
54:45
good, a really good starting point for
54:48
us because it sort of encapsulates a
54:50
lot of the things that we've been
54:52
talking about and leaves us with a
54:55
lot to think about that I have
54:57
death I have to just ask because
54:59
one question they came up, especially when
55:02
you were talking about the sort of
55:04
former issue of ethics and I'm thinking
55:06
about things in advance, but also the
55:09
multi disciplinary aspect of technology research. Right
55:11
that we need to be talking to policy makers,
55:13
That we need to be talking to psychologists and
55:15
we need to be talking to as this is.
55:19
Like you're a professor, right? The
55:21
you have students coming into your
55:23
tier laboratory. I'm wondering. how much
55:25
are they interested in grappling with
55:27
those questions? How much does a
55:29
field bird and some like I
55:31
just want to do good science
55:33
and you know, can't that be
55:35
somebody elses job Like I can't
55:37
do everything versus how based in
55:39
do these considerations need to be
55:41
in the technology itself. Whenever
55:44
we start a new project, We.
55:47
Organize a worksop. On.
55:49
What can go wrong? And
55:52
we. Gather. Students
55:55
and researchers of this and
55:57
backgrounds. And we brain.
56:00
storm, all the different ways in which
56:02
we can imagine using the technology and
56:04
thinking about what could go wrong with
56:06
it. And
56:08
if we start from this point, we
56:12
already are a few steps ahead
56:14
from the point of view of making
56:16
technology that can be used
56:18
for the greater good rather than
56:21
technology that can empower the supervillains.
56:26
We can begin to also
56:28
teach the students that it
56:30
is important to consider the consequences
56:32
of our work. And
56:35
it's important to proceed
56:37
cautiously and remember that
56:39
ultimately we are advancing
56:42
science. We are aiming
56:44
to understand AI. We
56:47
are aiming to understand robotics. We're
56:49
aiming to understand the science and
56:52
engineering of intelligence. And
56:55
doing this work is teaching us right
56:57
now that we have
56:59
so much more to learn about
57:01
these technologies and also
57:03
about life itself. And
57:06
I want to emphasize that
57:10
we are the only species
57:12
on our planet. So
57:15
smart, so aware, so
57:18
capable of building incredible tools.
57:22
But this also comes with the
57:24
responsibility of ensuring that
57:27
our work can be
57:29
deployed for
57:32
good, that we are
57:34
responsible for everything on
57:36
our planet, for
57:38
the plants and the animals, for
57:40
the neighbors we share this planet with, and
57:42
also for the future generation. We
57:45
are responsible and we have to
57:47
proceed with our work in a
57:49
way that ensures the global good.
57:53
And I personally believe that this is possible
57:55
and I'm very excited about all
57:57
the things we can do at this point.
58:00
in our history. And
58:03
of course that is a central
58:05
message to your book and
58:08
it's a central message to this conversation.
58:10
I mean just look at the
58:12
title of your book of The Heart and the
58:14
Chip, Our Bright Future with Robots.
58:17
I have to ask Daniela before we go and I
58:19
know it's a big question, but is there anything? Obviously,
58:22
there's probably a lot that we
58:24
didn't touch on or didn't talk about, but is
58:26
there anything that you feel like you
58:29
know is important to be said before before
58:32
we say goodbye. Is there anything that
58:34
we didn't cover that is important for
58:36
the listeners to know? I
58:39
want to emphasize that while
58:41
I understand that techno-optimism
58:43
isn't very popular right
58:46
now, I believe that with enough
58:49
forethought and planning, we
58:51
can achieve a future in
58:53
which our humanity
58:57
gets elevated. And
58:59
I know that this is not going to
59:01
be easy because the transition
59:04
will demand very careful considerations
59:06
of ethical, privacy and employment
59:09
implications to ensure that the
59:11
benefits of intelligent machines are
59:14
equitably distributed. But
59:19
I do want
59:21
to highlight that the
59:24
conversation around the impact of robots
59:26
and AI on a
59:28
grand or global scale tends
59:31
to focus on doomsday scenarios.
59:34
And these are obviously important
59:36
and absolutely must be accounted
59:38
for. We must
59:40
do everything we can to steer the
59:42
future in a positive direction. But
59:46
we should also think about how
59:48
we can use these new tools,
59:50
these intelligent machines to
59:53
solve some of the largest problems we face
59:55
as a species. Because
59:58
these Intelligent
1:00:00
systems powered by advances
1:00:03
in AI and robotics
1:00:05
offer unprecedented capabilities to
1:00:07
analyze fast data sets, to
1:00:10
identify patterns, to execute tasks
1:00:12
with a precision and efficiency
1:00:15
that far surpasses human limitations.
1:00:18
And the applications can
1:00:21
involve combating climate change
1:00:23
by optimizing energy consumption
1:00:25
and reducing emissions to
1:00:28
revolutionizing healthcare through individualized
1:00:30
medicine and robotic surgery
1:00:34
to creating sustainable solutions. There
1:00:36
are so many different
1:00:39
things that we can do with what
1:00:41
we have today. One
1:00:44
important area is using
1:00:46
intelligent machines to
1:00:48
advance scientific research, to
1:00:51
automate tedious experiments, to
1:00:53
provide hypotheses, to simulate
1:00:55
complex theoretical models, really
1:00:57
to accelerate the pace
1:00:59
of discovery. And
1:01:02
I believe that by effectively
1:01:04
merging the heart and the chip
1:01:07
and guiding the development of
1:01:09
applications with wisdom, with
1:01:11
foresight and with a commitment to the
1:01:13
greater good, we could
1:01:16
solve the problems facing
1:01:18
our people and our
1:01:20
planet. And we
1:01:22
could elevate humanity at
1:01:25
the same time. So well
1:01:27
said. Thank you so much. Well,
1:01:29
everybody, the book is The Heart
1:01:31
and the Chip, Our Bright Future
1:01:33
with Robots by Dr. Daniella Roos
1:01:35
along with Gregory Mohn. Thank you
1:01:37
so much for spending your time
1:01:39
with us today. Thank
1:01:42
you very much, Cara. And
1:01:45
everybody listening, thank you for coming back
1:01:47
week after week. I'm really looking forward
1:01:49
to the next time we all get together. Daily
1:02:00
beloved we are gathered here today to... has
1:02:02
anyone seen the bride and groom? Sorry,
1:02:05
sorry, we're here. We were getting lucky in
1:02:07
the limo when we lost track of time. No,
1:02:10
Lucky Land Casino with cash prizes that add
1:02:12
up quicker than a guest registry. In
1:02:14
that case I pronounce you lucky. Flight
1:02:17
for free at luckylandsucks.com. Daily
1:02:19
bonuses are waiting. No purchase
1:02:21
necessary. Voidware prohibited by law. 18-5
1:02:23
terms and conditions apply. See website for details.
1:02:26
Hello, it is Ryan and I was on a
1:02:28
flight the other day playing one of my favorite
1:02:30
social spin slot games on chumbacacino.com. I looked over
1:02:33
the person sitting next to me and you know
1:02:35
what they were doing? They were also playing Chumba
1:02:37
Casino. Coincidence? I think not. Everybody's loving having fun
1:02:39
with it. Chumba Casino is home to hundreds of
1:02:41
casino style games that you can play for
1:02:43
free anytime, anywhere, even at 30,000
1:02:46
feet. So sign up now at chumbacacino.com
1:02:48
to claim your free welcome bonus. That's
1:02:50
chumbacacino.com and live the Chumba life. No
1:02:52
purchase necessary. D.W. report. We're prohibited by
1:02:54
law. See terms and conditions. 18-plus.
Podchaser is the ultimate destination for podcast data, search, and discovery. Learn More