Kind: captions
Language: en
welcome to MIT course 6 s 0 9 9
artificial general intelligence today we
have Ray Kurzweil he is one of the
world's leading inventors thinkers and
futurists with a 30-year track record of
accurate predictions called the Restless
genius by The Wall Street Journal and
the ultimate thinking machine by Forbes
magazine he was selected as one of the
top entrepreneurs by Inc magazine which
described him as the rightful heir to
Thomas Edison PBS selected him as one of
the 16 revolutionaries who made America
Ray was the principal investigator of
the first ccd flatbed scanner the first
omni font optical character recognition
the first point to speech reading
machines for the blind the first
text-to-speech synthesizer the first
music synthesizer capable of creating
the grand piano and other orchestral
instruments and the first commercially
marketed large vocabulary speech
recognition among his many honors he
received a Grammy Award for outstanding
achievements in music technology he's
the recipient of the National Medal of
Technology was inducted into the
National Inventors Hall of Fame holds 21
honorary doctorates and honors from
three u.s. presidents Ray has written
five national best-selling books
including the New York Times bestsellers
The Singularity is near from 2005 and
how to create a mind from 2012 he is
co-founder and Chancellor of singularity
University and a director of engineering
at Google heading up a team developing
machine intelligence and natural
language understanding
please give ray a warm welcome
[Applause]
[Music]
it's good to be back I've been in this
lecture hall many times and walked the
infinite Carter I came here as an
undergraduate in 1965 within a year of
my being here they started a new major
called computer science it did not get
its own course number that's 6 1 even
biotechnology recently got its own
course number but how many of you are CS
majors ok how many of you do work in
deep learning how many of you have heard
of deep learning here I came here first
in 1952 when I was 14 I became excited
about artificial intelligence it had
only gotten its name six years earlier
the 1956 Dartmouth conference by Marvin
Minsky and John McCarthy so I wrote
Minsky a letter there was no email back
then and he invited me up he spent all
day with me as if he had nothing else to
do he was a consummate educator
I then and the AI field had already
bifurcated into two warring camps the
symbolic school which Minsk II was
associated with and the connectionist
school was not widely known in fact I
think it's still not widely known that
Minsk II actually invented the neural
net in 1953 but he had become negative
about it largely because there was a lot
of hype that these giant branes could
solve any problem
so the first popular neural nets the
perceptron was being promulgated by
Frank Rosenblatt at Cornell so Minsky
set out what are you going now and
saying I said to see Rosenblatt at core
now is that don't bother doing that and
I went there and Rosenblatt was touting
the perceptron that it ultimately would
be able to solve any problem so I
brought some printed letters that had
the camera and it did a perfect job of
recognizing them as long as they were
courier ten different types I didn't
work at all and he said but don't worry
we can take the output of the perceptron
or feed it as the input to another
perceptron and take the output of that
and feed it to a third layer and as we
add more layers it'll get smarter and
smarter and generalize and so that's
interesting if you even tried that well
no but it's high on our research agenda
things did not move quite as quickly
back then as they do now he died nine
years later never having tried that idea
turns out to be remarkably prescient I
mean he never tried multi-layer neural
nets and all the excitement we see now
about deep learning comes from a
combination of two things
both many layer neural Nets and the law
of accelerating returns which I'll get
to a little bit later which is basically
the exponential growth of computing so
that we can run these massive nets and
handle massive amounts of data it would
be decades before that idea was tried
several decades later three level neural
nets were tried there were a little bit
better they could deal with multiple
type styles still weren't very flexible
that's not hard to add other layers it's
a very straightforward concept there was
a math problem the disappearing gradient
or the exploding gradient which I'm sure
many of you are familiar with basically
you need to take maximum advantage of
the range of values in the gradients and
not let them explode or disappear and
lose the resolution that's a fairly
straightforward mathematical
transformation with that insight we
could now go 200 layer neural nets and
that's behind sort of all the fantastic
gains that we've seen recently
alphago trained on every online game and
then became a fair go player it then
trained itself by playing itself and
soared past the best human alphago zero
started with no human input at all
within hours of iteration sort Pascal
phago also soared past the best just
programs they had another innovation
basically you need to evaluate the
quality of the board at each point they
used another hundred layer neural nets
to do that evaluation so there's still a
problem in the field which is there's a
motto that life begins at a billion
examples
one of the reasons I'm at Google is we
have a billion examples for examples of
pictures of dogs and cats that are
labeled so you got a picture of a cat
and it says cat and then you can learn
from it and you need a lot of them
alphago trained on a million online
moves that's how many we had of master
games and that only created a sort of
fair go player a good amateur could
defeated so they worked around that in
the case of go by basically generating
an infinite amount of data by having the
system play itself had a chat with
Denver's house office you know what kind
of situations can you do that with you
have to have some way of simulating the
world so go or chess are even though go
is considered a difficult game it's
a-you know the definition of it can
exist on one page so you can simulate it
that applies to math I mean amass axioms
are can be contained on a page or two
it's not very complicated it gets more
difficult when you have real-life
situations like biology so we have
biological simulators but the simulators
on perfect so learning from the
simulators will only be as good as the
simulators that's actually the key to
being able to do deep learning on
biology
autonomous vehicles you need real-life
data so the way mo systems have gone
three and a half million miles
that's good that's enough data to then
create a very good simulator so the
simulator is really quite realistic
because they had a lot of real-world
experience and the they've got a billion
miles in the simulator but we don't
always have that opportunity to either
create the data or have the data around
humans can learn from a small number of
examples your significant other your
professor your boss your investor can
tell you something once or twice and you
might actually learn from that some
humans have been reported to do that
and that's kind of the remaining
advantage of humans now there's actually
no back propagation in the human brain
it doesn't use deep learning it uses a
different architecture that same year in
1962 I wrote a paper how I thought the
human brain worked there was actually
very little neuroscience to go on there
was one neuroscientist Vernon mount
Castle that had something relevant to
say which as he did I mean there was a
the common wisdom at the time and
there's still a lot of neuroscience that
says say this that we have all these
different regions of the brain they do
different things they must be different
there's v1 in the back of the head where
the optic nerve spills into that can
tell that that's a curved line that
that's a straight line does these simple
feature extractions on visual images
it's actually a large part of the
neocortex does the fusiform gyrus up
here which can recognize faces we know
that because if it gets knocked out
through injury or stroke people can't
recognize faces they will learn it again
with a different region of the neocortex
is the famous frontal cortex which does
language in poetry and music so these
must work on different principles he did
autopsies on the neocortex and all these
different regions and found they all
looked the same they had the same
repeating pattern same interconnections
he said neocortex is neocortex so I had
that hint otherwise I can actually
observe human brains in action which I
did from time to time and there's a lot
of hints that you can get that way for
example if I ask you to recite the
alphabet you actually don't do it from A
to Z you do it as a sequence of
sequences ABCD efg hijk so we learn
things that secret forward sequences of
sequences forward because if I ask you
to recite the alphabet backwards you
can't do it unless you learn that as a
new sequence so these are all
interesting hints I wrote a paper that I
that the neocortex is organized as a
hierarchy of modules in each module can
learn a simple pattern and that's how I
got to meet President Johnson and that
initiated a half-century of thinking
about this issue I came to MIT to study
with Marvin Minsky actually came for two
reasons one the Minsky became my mentor
which was a mentorship that lasted for
over 50 years the fact that MIT was so
advanced it actually had a computer
which the other colleges I considered
didn't have it was an IBM 7090 for 32 K
of 36 bit words so it's 150 K of course
storage to microsecond cycle time two
cycles for instructions or a quarter of
a myth and that thousands of students
and professors shared that one machine
in 2012 I wrote a book about this thesis
is now actually an explosion of
neuroscience evidence to support it the
European brain reverse engineering
project has identified a repeating
module about a hundred neurons it's
repeated three hundred million times
it's about 30 billion neurons in the
neocortex the neocortex is the outer
layer of the brain that's part where we
do our thinking and they can see in each
module axons coming in from another
module and then the output acts the
single output accent of that
Jil goes as the input to another module
so we can see it organized as a
hierarchy it's not a physical hierarchy
it's the hierarchy comes from these
connections the neocortex is a very thin
structure it's actually one module thick
there's six layers of neurons but it
constitutes one module and we can see
that it learns in simple pattern and
various reasons I cite in the book the
pattern recognition model that's using
is basically a hidden Markov model how
many of you have worked with Markov
models okay
that's usually no hands go open I asked
that question but Markov model is not it
is learned but it's not back propagation
it can learn local features so it's very
good for speech recognition and the
speech recognition network I did in the
80s used these Markov models that became
the standard approach because it can
deal with local variations so the fact
that a vowel is stretched you can learn
that in a Markov model it doesn't learn
long distance relationships that's
handled by the hierarchy and something
we don't fully understand yet is exactly
how the neocortex creates that hierarchy
but we have figured out how it can
connect this module to this module does
it then grow I mean there's no virtual
communication or wireless communication
it's actually connection so does it grow
an axon you know from one place to
another which could be inches apart
actually they all all these connections
are there from birth like the streets
and avenues of Manhattan there's
vertical and horizontal connections so
if the it decides and how it makes that
decision it's still not fully understood
that it wants to connect this module to
this module there's already a vertical
horizontal and a vertical connection it
just activates them we can actually see
that now and I can see that happening in
real time on non-invasive brain scans
so there's a current amount of evidence
that's in fact the neocortex is a
hierarchy of modules that can learn each
module learns a simple sequential
pattern and even though the patterns we
perceived don't seem like sequences they
may seem three-dimensional or even more
complicated they are in fact represented
as sequences but the complexity comes in
with the hierarchy so the neocortex
emerged 200 million years ago with
mammals all mammals have a neocortex
it's one of the distinguishing features
of mammals these first mammals were
small they were rodents but they were
capable a new type of thinking other
non-mammalian animals had fixed
behaviors but those fixed behaviors were
very well adapted for their ecological
niche but these new mammals could invent
a new behavior so creativity and
innovation was one feature of the
neocortex so a mouse is escaping a
predator its usual escape path is
blocked it will invent a new behavior to
deal with it probably wouldn't work but
if it did work it would remember it and
would have a new behavior and that
behavior could spread virally through
the community another Mouse watching
this was with say to itself that was
really clever going around that rock I'm
gonna remember to do that and it would
have a new behavior didn't help these
early mammals that much because as I say
the non-mammalian animals were very well
adapted to their niches and nothing much
happened for a hundred and thirty five
million years but then 65 million years
ago something did happened there was a
sudden violent change to the environment
we now call it the Cretaceous extinction
event there's been debate as to whether
it was a media or an asteroid I mean a
meteor or a volcanic eruption the
asteroid or meteor hypothesis is in the
ascendancy but if you dig down to an
area of rock reflecting 65 million years
ago the
geologists will explain that it shows a
very violent sudden change to the
environment we see it all around the
globe so is a worldwide phenomenon the
reason we call it an extinction event is
that's when the dinosaurs went extinct
that's when 75% of all the animal and
plant species went extinct and that's
when mammals overtook their ecological
niche so to anthropomorphize biological
evolution said to itself this neocortex
is pretty good stuff and it began to
grow it so-now mammals got bigger their
brains got bigger at an even faster pace
taking up a larger fraction of their
body the neocortex got bigger even
faster than that and developed these
curvatures that are distinctive of a
primate brain basically to increase its
surface area but if you stretched it out
the human neocortex is still a flat
structure it's about the size of a table
napkin just as thin and it's basically
created primates which became dominance
in their ecological niche then something
else happened two million years ago
biological evolution decided to increase
the neocortex further and increase the
size of the enclosure and basically
filled up the frontal cortex with our
big skulls with more neocortex and up
until recently it was felt that as I
said that this was the frontal cortex
was different because it does these
qualitatively different things but we
now realize that it's really just
additional neocortex so remember what we
did with it we're already doing a very
good job of being primates so we put it
at the top of the neocortical hierarchy
and we increased the size of the
hierarchy it was maybe 20% more
neocortex but it doubled it tripled the
number of levels because as you go up
the hierarchy it's kind of like a
pyramid there's fewer and fewer modules
and that was the enabling factor for us
to invent language and art music every
human culture we've ever discovered has
music no primary culture really has
music there's debate about that but it's
really true
invention technology technology required
another evolutionary adaptation which is
this humble appendage here no other
animal has that if you look at a chimp
and see it looks like they have a
similar hand but the thumb is actually
down here doesn't work very well if you
watch them trying to grab a stick so we
could imagine creative solutions yeah I
could take that branch and strip off the
leaves and put a point on it and we
could actually carry out these ideas and
create tools and then use tools to
create new tools and it started a whole
nother evolutionary process of
tool-making and that all came with the
with the neocortex
so Larry Page read my book in 2012 and
liked it so I met with him in Essen for
an investment in a company I'd started
actually a couple weeks earlier to
develop those ideas commercially because
that's how I went about things as a
serial entrepreneur
and said well we'll invest but let me
give you a better idea what you do it
here at Google we have a billion
pictures of dogs and cats and we've got
a lot of other data and lots of
computers and lots of talent all of
which is true and says well I don't know
I just started this company to develop
this is well by your company and how you
got a value a company that hasn't done
anything just started a couple weeks ago
and he said we can value anything so I
took my first job five years ago and
I've been basically applying this model
this hierarchical model to understanding
language which i think really is the
holy grail of AI I think Turing was
correct in designating basically text
communication as what we now call a
turing-complete problem that requires
there's no simple NLP tricks it you can
apply to pass a valid Turing test with
an emphasis on the word valid mitch
kapor and i had a six month debate on
what the rules should be because if you
read Turing's 1950 paper he describes
this in a few paragraphs and doesn't
really describe how to go about it but
if it's a valid Turing test meaning it's
really convincing you through an
interrogation and dialogue that it's a
human that requires a full range of
human intelligence and I think that test
has to the test of time we're making
very good progress on that I mean just
last week you may have read that two
systems
asked paragraph comprehension test it's
really very impressive winning came to
Google we were trying to past these
paragraph comprehension tests we aced
the first the first grade test second
grade tests were kind of got average
performance and the third grade test had
too much inference already you had to
know some common-sense knowledge as it's
called and make implications of things
that were in different parts of the
paragraph and there's too much inference
and it really didn't didn't work so this
is now adult level it's just slightly
surpassed average human performance but
we've seen that once something an AI
does something it average human levels
it doesn't take long for it to soar past
average human levels I think it'll take
longer in language and it did in some
simple games like go but it's actually
very impressive that it surpasses now
average human performance used at LST M
long short temporal memory but if you
look at the adult test in order to
answer these questions it has to put
together inferences and implications of
several different things in the
paragraph with some common sense
knowledge is not explicitly stated so
that's I think a pretty impressive
milestone so I I've been developing I've
got a team of about 45 people and we've
been developing this hierarchical model
we don't use Markov models because we
can use deep learning for each module
and so we create an embedding for each
word and we create an embedding for each
sentence this is we have a I can talk
about it because we have a published
paper on it it can take into
consideration context
if you use smart reply on G confused
email on your phone you'll see it gives
you three suggestions for responses
that's called Smart reply there are
simple suggestions but it has to
actually understand perhaps a
complicated email and the quality of the
suggestions is really quite good quite
on point that's for my team using this
kind of hierarchical model so instead of
Markov models that uses embeddings
because we can use back propagation we
might as well use it but I think what's
missing from deep learning is this
hierarchical aspect of understanding
because the world is hierarchical that's
why evolution developed a hierarchical
brain structure to understand the
natural hierarchy in the world
and there are several problems with big
deep neural nets one is the fact that
you really do need a billion examples
and we don't sometimes we can generate
them it's in the case of NGO or if we
have a really good simulator as in the
case of autonomous vehicles not quite
the case yet in biology very often you
don't have a billion example if you
suddenly have billions of examples of
language but they're not annotated and
how would you annotate it anyway with
more language that we can't understand
in the first place so that's kind of a
chicken and an egg problem so I believe
this hierarchical structures needed
another criticism of deep neural Nets
they don't explain themselves very well
it's a big black box that gives you
pretty remarkable answers I mean in the
case of these games demos described it's
playing in both go and chess is almost
an alien intelligence because we do
things that were shocking to you and
experts like sacrificing a queen and a
bishop at the same time or in close
succession which shocked everybody but
then went on to win or early in a go
game putting a piece at the corner of
the board which is kind of crazy to most
experts because you really want to start
controlling territory and yet it on
reflection that was the brilliant move
that enabled it to win that game but it
doesn't really explain how it does these
things so if yeah if you have a
hierarchy it's much better at explaining
it because you could look at the content
of the of the modules in the hierarchy
and they'll explain what they're doing
and just and on the first application of
applying this to health and medicine
this will get into high gear and we're
going to really see us break out at the
linear extension to longevity that we've
experienced I believe we're only about a
decade away from longevity escape
velocity we're adding more time than is
going by not just the infant life
expectancy but to your remaining life
expectancy I think if someone is
diligent they can be there already I
think I've
at longevity escape velocity now a word
on what life expectancy means it used to
be assumed that not much would happen so
whatever your life expectancy is with or
without scientific progress it really
didn't matter now it matters a lot so
life expectancy really means you know
how long would you live what's the in
terms of a statistical likelihood if
there were not continued scientific
progress but that's a very inaccurate
assumption that scientific progress
is extremely rapid I mean just as an AI
in biotech there are advances now every
week is quite stunning
now you can have a computed life
expectancies let's say 30 years 50 years
70 years from now you can still be hit
by the proverbial bus tomorrow we're
working on that with self-driving
vehicles but we'll get we'll get to a
point I think if you're diligent you can
be there now in terms of basically
advancing your own statistical life
expectancy
at least to keep pace with the passage
of time I think it would be there for
most of the population at least if
they're diligent within about a decade
so if we can hang in there we may get to
see the remarkable century ahead thank
you very much no question please raise
your hand we'll get your mic hi
so you mentioned both neural neural
network models and symbolic models and I
was wondering how far have you been
thinking about combining these two
approaches creating a symbiosis between
neural models and symbolic ones I don't
think we want to use symbolic models as
they've been used how many are familiar
with the psych project
that was a very diligent effort in Texas
to define all of common-sense reasoning
and it kind of collapsed on itself and
became impossible to debug because you
fix one thing and it break three other
things that complexity ceiling has
become typical of of trying to define
things through logical rules now it does
seem that humans can understand logical
rules we have logical rules written down
for things like law and game playing and
so on but you can actually define a
connectionist system to have such a high
reliability on a certain type of action
that it looks like it's a symbolic rule
even though it's represented in a
connectionist way and connection systems
can both capture the soft edges because
many things in life are not sharply
defined they can also generate
exceptions so you you don't want to
sacrifice your queen in chess accept
certain situations that might be a good
idea so you can capture that kind of
complexity so we do want to be able to
learn from accumulated human wisdom that
looks like it's symbolic but I think
we'll do it with a connection system but
again I'm think the connection systems
should develop a sense of hierarchy and
not just be one big massive neural net
so I understand how we want you know use
the neocortex to extract useful stuff
and commercialize that but I'm wondering
how you know our middle brain and organs
that are below the neocortex will be
useful for you know turn that into what
you want to do something well the
cerebellum is an interesting case in
point it actually has more neurons than
the neocortex and it's used to
govern most of our behavior some things
if you write a signature that's actually
controlled by the cerebellum so a simple
sequence is stored in the cerebellum but
there's not many reasoning to it it's
basically a script and most of our
movement now has actually been migrated
from the center vellum to the neocortex
cerebellum is still there some people
the entire cerebellum is destroyed
through disease they still function
fairly normally their movement might be
a little erratic as our movements is
largely controlled by the neocortex but
some of the subtlety is a kind of
pre-programmed script and so they'll
look a little clumsy but they're
actually function okay a lot of other
areas of the brain control autonomic
functions like breathing and but our
thinking really is is controlled by the
neocortex in terms of mastering
intelligence I think the neocortex is
the brain region we want to study I'm
curious what you think might happen
after the singularity is reached in
terms of this exponential growth of
information yes do you think it will
continue or will there be a whole
paradigm shift what do you predict well
in the singularities near I talked about
the atomic limits based on molecular
computing as we understand it and it can
actually go well past 2045 and actually
go to trillions of trillions of times
greater computational capacity than we
have today
so I don't see that's stopping anytime
soon and we'll go you know way beyond
what we can imagine and it becomes an
interesting discussion what the impact
on human civilization will be so take it
may be slightly more mundane issue that
comes up as a kind of eliminates most
jobs or
jobs a point I make is it's not the
first time in human history you've done
that how many jobs circa 1900 exist
today and that was the feeling of the
Luddites which was an actual society
that formed in 1800 the automation of
the textile industry in England they
looked at all these jobs going away and
felt that employment is going to be just
limited to an elite indeed those jobs
didn't go away but new jobs were created
so if I were oppression Futures to 1900
I would say well 38% of you work on
farms and 25% work in factories it's 2/3
of the working force but I predict by
2015 115 years from now it's going to be
2% on farms and 9% factories and
everybody would go oh my God we're gonna
be out of work and I said well don't
worry for all these jobs we eliminate
through automation we're gonna invent
new jobs and say oh really what new jobs
and I'd say well I don't know we haven't
invented them yet that's the political
problem we could see jobs very clearly
going away fairly soon like driving a
car or truck and the new jobs haven't
been invented I mean just look at the
last five or six years as many a lot of
the increase in employment has been
through mobile app related types of ways
of making money that just weren't
contemplated even six years ago if I
really prescient I would say well you're
gonna get jobs creating mobile apps and
websites and doing data analytics and
self-driving cars cars what's a car and
nobody would have any idea what I'm
talking about now the new job
some people say yeah we created new jobs
but it's not as many actually we've gone
from 24 million jobs in nineteen hundred
242 million jobs today for 30 percent of
the population to forty five percent of
the population the new jobs pay eleven
times as much in constant dollars and
they're more interesting and as I talk
to people starting out their career now
they really want a career that gives
them some
life definition and purpose and
gratification we're moving up Maslow's
hierarchy hundred years ago you were
happy if you had a back-breaking job to
put food on your family's table so and
we couldn't do these new jobs without
enhancing our intelligence so we've been
doing that well for most of the last 100
years through education we've expanded
to K through 12 and constant dollars
tenfold
we've gone from 38,000 college students
in 1870 to 15 million today more
recently we have brain extenders and not
yet connected directly in our brain but
they're very close at hand when I was
here that my tía to take my bicycle
across campus to get to the computer and
show an ID to get in the building now we
carry them well you know in our in our
pockets and on our belts
they're going to go inside our bodies
and brains I think that's a notic really
important distinction but so we're
basically going to be continuing to
enhance our capability through merging
with AI and that's the I think ultimate
answer to the kind of dystopian view we
see in futures movies where it's the AI
versus a brave band of humans for
control of humanity we don't have one or
two a eyes in the world today we have
several billion three billion
smartphones and last count will be six
billion in just a couple of years
according to the projections so we're
already deeply integrated with this and
I think that's going to continue and
it's gonna continue to do things that
you can't even imagine today just as we
are doing today things we couldn't
imagine you know even twenty years ago
you showed many graphs that goes through
exponential growth but I haven't seen
one that isn't so I would be very
interested in hearing you haven't seen
that what that is not exponential so
tell me about regions that you've
investigated that have not seen
exponential growth and why do you think
that's the case well
price performance and capacity of
information technology invariably
follows a exponential when it impacts
human society it can be linear so for
example the growth of democracy has been
linear but still pretty steady you can
count the number of democracies on the
fingers of one hand a century ago two
centuries ago you can count the number
of democracies in the world on the
fingers of one finger now there are
dozens of them that this and it's become
kind of a consensus that that's how we
should be governed
so the and I attributed all this to the
growth and information technology
communication in particular for
progression of social cultural
institutions but information technology
because it ultimately depends on a
vanishingly small energy and material
requirement grows exponentially and will
for a long time there's recently a
criticism that well test scores have
it's actually a remarkably straight
linear progression so humans think it's
like twenty eight hundred and it just
sort passed out in 1997 with the blue
and it's kept going and remarkably
straight and saying well this is linear
not exponential but the chess score is a
logarithmic measurement so it really is
exponential progression so if you're
lhasa furs like to think a lot about the
meaning of things especially in the 20th
century so for instance Martin Heidegger
gave a couple of speeches and lectures
on the relationship of human society to
technology and he particularly
distinguished between the mode of
thinking which is calculating thinking
and a mode of thinking which is
reflective thinking or meditative
thinking and he posed this question what
is the the meaning and purpose of
technological development and he
couldn't find an answer he he
recommended to remain open to what he
called and he called this an openness to
the mystery I wonder whether you have
any thoughts on this is there is there a
meaning of purpose to technological
equipment and and is there a way for a
human success access that meaning well
we started using technology to shore up
weaknesses and our own capabilities so
physically I mean who here could build
this building so we've leveraged the
power of our muscles with machines
and we're in fact very bad at doing
things that you know the simplest
computers can do like factor numbers or
even just multiply two eight digit
numbers computers can do that trivially
we can't do it so we originally started
using computers to make up for that
weakness I think the essence of what
I've been writing about is to master the
unique strengths of humanity creating
loving expressions in poetry and music
and the kinds of things we associate
with the better qualities of humanity
with machines that's the to promise of
AI that we're not there yet but we're
making pretty stunning progress just in
the last year there's so many milestones
that are really significant including in
language and but I think of technology
as an expression of humanity it's part
of who we are and the human species is
already a biological technological
civilization and it's part of who we are
an AI is it's part of humans so AI is
human and it's it's part of the
technological expression of humanity and
we use technology to extend our reach
you know I couldn't reach that fruit at
that higher branch a thousand years ago
so we invented a tool to extend our
physical reach we now extend our mental
reach we can access all of human
knowledge with a few keystrokes and
we're going to make ourselves literally
smarter by merging with AI hi
first of all honor to hear you speak
here so I first read The Singularity is
near nine years ago or so and it changed
the way I thought entirely but something
I think it caused me to over steeply
discount was tail risk in geopolitics in
systems that span the entire globe and
my concern is that there are there is
obviously the possibility of tail risk
existential level events swamp in all of
these trends that are otherwise war
proof climate proof you name it so my
question for you is what steps do you
think we can take in designing
engineered systems in designing social
and economic institutions to kind of
minimize our exposure to these tail
risks and and and survive to make it to
UM you know a beautiful mind filled
future yeah well the world was first
introduced to a human-made
existential risk when I was in
elementary school we would have these
civil defense drills to get under our
desk and put our hands behind our head
to protect this from a thermonuclear war
and it worked we made it through but
that was really the first introduction
to an existential risk and those weapons
are still there by the way and they're
still on a hair-trigger and they don't
get that much attention there's been a
lot of discussion much of which I've
been in the forefront of initiating the
existential risks of what sometimes
referred to as GN rg4 genetics which is
biotechnology and for nanotechnology and
gray goo robotics which is a
and I've been accused of being an
optimist I think you have to be an
optimist to be an entrepreneur if you
knew all the problems you were going to
encounter you'd never start any project
but I've written a lot about the
downsides I remain optimistic there are
specific paradigms and not foolproof
that we can follow to keep these
technologies safe so for example over 40
years ago some visionaries recognized
the revolutionary potential both for
promise and peril of biotechnology
neither the promise no peril was
feasible 40 years ago but they had a
conference at the Asilomar conference
center in California and to develop both
professional ethics and strategies to
keep biotechnology safe and they've been
known as the Asilomar guidelines they've
been refined through successive sell
more conferences much of that's baked
into law and it in my opinion it's
worked quite well we're now as I
mentioned getting profound benefit it's
a trickle today it'll be a flood over
the next decade and the number of people
who have been harmed either through
intentional or accidental abuse of
biotechnology so far zero actually I
take that back there was one boy who
died in gene therapy trials but 12 years
ago and there's congressional hearings
and they cancelled all research for gene
therapy for a number of years you could
do an interesting master's thesis and
demonstrate that you know 300,000 people
died as a result of that delay but you
can't name them they can't go on CNN so
we don't know who they are but it has to
do with the balancing of risk but in
large measure virtually no one has been
hurt by biotechnology now that doesn't
mean you can cross on our front list
okay we took care of that one because
the technology keeps getting more
sophisticated and Christopher's great
opportunity there's hundreds of trials
of Christopher's technologies overcome
disease but it could be abused you can
describe scenarios so we have to keep
reinventing it January we had our first
Asilomar conference on AI ethics and so
I think this is a good paradigm it's not
foolproof I think the best way we can
assure a democratic future that includes
our ideas of Liberty is to practice that
in the world today because the future
world of the singularity which is a
merger of biological non-biological
intelligence it's not going to come from
Mars I mean it's going to emerge from
our society today so if we practice
these ideals today it's going to have a
higher chance of us practicing them as
we get more enhanced with technology if
that doesn't sound like a foolproof
solution it isn't but I think that's the
best approach in terms of technological
solutions
I mean AI is the most daunting you can
imagine there are technical solutions to
biotechnology and nanotechnology there's
really no subroutine you can put in your
AI software there will assure that it
remains safe intelligence it's
inherently not controllable
there's some AI that's much smarter than
you that's out for your destruction the
best way to deal with that is not to get
in that situation in the first place if
if you are in that situation and find
some AI that will be on your side but
basically it's going to eyeb Aleve
we have been headed through technology
to event to a better reality look around
the world and people really think things
are getting worse and I think that's
because our information about what's
wrong with the world is getting
exponentially better I say oh this is
the most peaceful time on you in history
if you say what are you crazy didn't you
hear about the event yesterday and last
week and well a hundred years ago there
could be a battle that wiped out the
next village in you wouldn't even hear
about it for months
of all these graphs on education and
literacy has gone from like 10% to 90%
over a century and health wealth
poverty's declined 95% in Asia over the
last 25 years document about the World
Bank all these trends are very smoothly
getting better and everybody thinks
things are getting worse but but but
you're right like on violence that curve
could be quite disrupted there's an
existential event as I say I'm
optimistic but I think that is something
if we need to deal with that a lot of it
is not technological it's dealing with
our social cultural institutions so you
mentioned also exponential growth of
software and IDs I guess related to
software so one of the reasons for which
you said that all that information
technology costs this exponential is
because of fundamental properties of
matter and energy but in the case of
ideas why would it have to be
exponential well a lot of ideas produce
exponential gains they don't increase
performance linearly there's actually
study during the Obama administration by
his scientific advisory board on
assessing this question how much gains
on 23 classical engineering problems
were gained through hardware
improvements over the last decade and
software improvements and there's about
a thousand to one improvement it's about
doubling every year from Hardware there
was an averages of like twenty six
thousand to one through softer
improvements algorithmic improvements so
we do see both and apparently if you
come up with in advance its it doubles
the performance or multiplies it by ten
we see basically exponential growth from
each innovation
so and we certainly see that in deep
learning the architectures are getting
better
while we also have more data and more
computation and more memory to throw in
these at these algorithms
thank you for being