I always liked everything that gravitates
around science fiction
whether it is
interstellar travel
beautiful city where life seems more
simple comfort more accessible to
everyone from the little robots that we
help in the work that we
make life easier who are capable
to interact as we humans I
did you find that nice phone you know
probably elite in android the
cult series star trek android who
understood everything but humor
I find it good then just like
that I’d like to know with a show of hands
how many people think that in a
near future
there will actually be robots that
are able to understand the language
human and interact as humans
not worse not fixed that good but when
even skeptics
I would like to show you a little
video that maybe will you do
change your mind and I’m talking about
skeptics
it’s done singstar good as you
see there is still a
improvement there is a little humor
still what you see actually
it’s not a prototype in a
laboratory is not a toy it’s not
just a moving doll that’s
actually a real robot is a
real company trying to sell
that and that’s what you’ll find
in a very short time in
information kiosks at the airport
in shopping centers so we are not
really not far from having this
kind of technology then there is more than
15 years old now I had started a
doctorate in what’s called the
automatic language processing
natural for the purpose of understanding
human conversations so
automatic so we speak we talk but
how can we allow
the machine to understand what they say
then once we understand that is
what the machine can do a return
then what I did well I had
so did my doctorate on it
sure my parents hood is easy
but what art are you going to do
with that can find you a job with that
but I still managed to
work a few years in the
research in different universities
then a few years later I
worked on crime
worry door is it’s the center of
computer research montreal
I was not yet made delinquent
girls my parents was super happy
I had a job I had a few pay
there everything was beautiful and working
at crime I had the opportunity really
to meet several companies from
voices actually what we could do with
artificial intelligence for
meet the real needs of the
life and that’s how I got
said ben actually i think i have
really everything to try me too from
confront everything I’ve learned to
solve real problems
and that’s how I left my
company with my associate charles
olive tree both passionate
artificial intelligence ben on
started working to try to
solve problems including the one that
today we do in the company
which actually allows to provide a
solution for businesses for with
to help them make their customers
happier so if we were able
today to understand everything that
were telling customers via their texting
had said to their emails on
facebook on twitter and we were
able to synthesize all of that and say
finally to the company down and
not look at what was customer and not
are not happy with this product on such
aspect there are things to improve
you imagine how much and we
so that it is less known consumer
would be happy that’s what we try to
solve with with our product
so that’s for sure that what I did
like any good entrepreneur I
started reading tons and
tons of books on how to create a
business
the difficult parts the traps to
avoid all kinds of things but
in hindsight
a few years now I got
realized that in reality with when
there is shown a product that contains
artificial intelligence ben there are
as a degree of complexity
additional that makes the company
becomes a bit more perilous
more complex there are other traps
typical to artificial intelligence
that you have to try to avoid and it’s a
I do not want to talk to you
so when we go to the
marketing of this type of
technology that we want to avoid
but let’s start at the beginning with
beginning what is
artificial intelligence but
artificial intelligence is the
process and that’s the ability of
reproduce human reasoning this
cognitive process which which is included
in a decision in our predictions
so finally we how do we
think about it and what does that mean
it means that there is something
who’s going who’s calling
to knowledge and that’s it a little
the complexity we will find with
artificial intelligence us as
that being human has plenty of
knowledge we use full of
knowledge we have acquired from our
lived our studies but it’s done
in time so how do we
can allow the machine to do
that quickly and have the whole hall
knowledge actually needed to
decision making
that’s what makes that
it’s difficult that’s what makes this
technology still relatively
more complex than technology
standard I will go so much
stubbornness forgiveness to the technologies of
information
so to answer seven accomplices
complexity there is a solution that this
solution is learning
automatic
So what’s the learning
automatic maria the itc an algorithm
who has the ability to learn a
general scheme a global scheme to
from a lot a lot a lot of
data that describe this schema
there but individually so
if we take for example the
facial recognition
so now there are plenty of systems
who are able to recognize a
person generally by
relation to an animal a dog or a cat
how come it actually is
will give the algorithm thousands
even millions of people picture
then this algorithm he will learn to
recognize globo traits that make
that they are distinctive with respect to
the animal and that’s really what makes
that the voice algorithm learn to
recognize the face of a human being
but obviously when we talk about
a lot of data we will say ok
but we will go get them where these
ball data we are lucky since
a decade or so about where you
hear a lot about the term big
data so what are these
millions of data about everything and
whatever
and then thanks to the advent or where the
face the ease of access to the
storage and calculation capacity
now it is available almost
to anyone any
company can have access
to his computing abilities and the
ability to store a lot of
data you know the companies
like amazon google microsoft every
genres and all these companies there
today
they offer this type of infrastructure to
shots simply derisory and
that it really helped to indulge in a
great boost to research in
artificial intelligence because
suddenly people could pick up
data you want could
improve the algorithms and so
spend more quickly having a
better maturity is passed to the
marketing and so that’s what
I want a little elaborate on it
so beyond the complexity I
would say scientific intelligence
artificial
when we go to the level of
marketing via a new
complexity another layer that
add up to that and then 1
there are about three elements that
I ‘m going that I would like I would like
speak to you then the first is
clearly avoid having a tweet
technology
in other words, as a scientist
we can very quickly leave on a
ambitious project that solves the
problem a universal problem and one
can spend his life pillon this
will not happen anywhere he’ll ever be
marketed so a first thing
it’s really focus and on an ace a
very particular problem and why
it’s important it’s because as I
said it just now when he
something that touches on reasoning
human to knowledge ben if we have
need a lot of knowledge
it’s difficult to take the case of a
system that makes dictation
automatic
if we were to develop this
system in the medical field ben he
should just learn the vocabulary
medical
but if we wanted to sell a system
that allowed to dictate for
Mister everyone
it would be the medical field that would be
finance will be the hotel would be
a multitude of areas that makes that
it’s really hard to put any
this information there in this system
is the solution to that is enough
simple is to try to find
a company that would actually be that
would live a problem in particles
and that intelligence could solve
and try to work together so
on one hand it will allow to have
information and secondly it’s fine
allow our element very
important is to recover the
feedback from our future customers
once again avoided the trip
technology is also knowing
what helps the customer and then what
is not she does not help them and that’s it
brings me to the second point the second
element is to be able to explain this
what brings artificial intelligence
why why me as a
who your customers who will buy
a technology I’m going to fetch
i’m going i’m going to want to buy your
technology because it intelligence
artificial is actually very
often what we will find in the
market is it’s technology
that they have a problem
and when we arrive with a
new technology that includes
artificial intelligence I had
already solutions
why mine will be better so
there is quite a whole a notion of
awareness why this one in
particular which contains a
artificial intelligence sees to be
better and for that one more time
there are several elements there are several
strategies one is to be able
demonstrate the contribution of this type of
technology the second element that’s
allow a little to show other cases
of use of intelligence
artificial and how that
actually improves things
today we will have systems
who are in the medical field who
are in the trade there are several
you look amazon goes back there
there are also intelligence algorithms
who will refer you to other products
who you will love because you
bought another product
particular so that’s a little do this
parallel and build on on this success
the year the third element is who and what
not the least I would say is that of
the promise we make to our customers
when we sell them beyond this type of
technology us as a human being
we are very often we work very
often in binary worried is that
it works it works it does not work
it’s one of two things
near
it’s not the right message we mean
another client but yet this is the
reality
let’s not forget we’re in the process of
reproduce human reasoning then
even us as a low human
we are wrong we are we hear the same
facts but everyone will react
we will answer a different answer
on the other and we can be wrong
so it’s normal that also as
product there is so a certain margin
of error there is some a certain
level of error is so at this moment
here it’s all about managing the tent
of the customer how can we
explain who can have
the mistake then to reassure him too
show that we have the ability
opportunity to allow him to
fix the system
so there is really it’s all a
paradigm that will change and that it takes
really take the time and do the
things for them to understand and he
put confidence in somewhere so
these are really this type of challenge I
would you say that they are things that
add up to complexity
what are we going to have when we create a
new a new business
so does that mean that the
rate of i would say two successes is
less good for companies that
use TV intelligence
artificial than other companies
clearly no I would say to you
since 2011 there is still a lot more
of companies that make intelligence
artificial actually that sells
products with a strong component that
use artificial intelligence and
with that it’s going so well for this
kind of technological also the
investors started investing
much more in this type of
technology so that’s good it’s happening
well I think the revolution of
artificial intelligence has begun
then she is here to stay for sure
that if it brings other challenges that I
will put more like challenges
socioeconomic
just to give you an idea
since 2015 there is an initiative that is
led by several countries the states
united canada then other countries in
europe to just think about
the impact
what is the impact of this revolution
of artificial intelligence is this
that we actually have robots that go
we take our place ourselves as
that being human at work
is the social notion of interacting
between people is going to be different
is that tomorrow we will be friends
would be robots so that’s all
kinds of debates really that they
people really take time to
think about it then I can you
to say that it’s serious enough for
drive the initiative this initiative
over a period of a century
so it started in 2015 it’s going
to finish in 1115 is still
still major enough and I can you
also say that I do not necessarily
the answer to that question there and
what the impact is what it means
to say that tomorrow morning we will have the
half of the unemployed population who
will have more work for people I do not
do not think personally but I think
that the response sees coming especially to
as the future and then we
as a human being and then how
we adjust convoy who sees who
let’s build this
this answer to that makes me
leave with with this question there and
then we’ll have the time just now
talk about his thanks a lot
[Applause]
[Music]
around science fiction
whether it is
interstellar travel
beautiful city where life seems more
simple comfort more accessible to
everyone from the little robots that we
help in the work that we
make life easier who are capable
to interact as we humans I
did you find that nice phone you know
probably elite in android the
cult series star trek android who
understood everything but humor
I find it good then just like
that I’d like to know with a show of hands
how many people think that in a
near future
there will actually be robots that
are able to understand the language
human and interact as humans
not worse not fixed that good but when
even skeptics
I would like to show you a little
video that maybe will you do
change your mind and I’m talking about
skeptics
it’s done singstar good as you
see there is still a
improvement there is a little humor
still what you see actually
it’s not a prototype in a
laboratory is not a toy it’s not
just a moving doll that’s
actually a real robot is a
real company trying to sell
that and that’s what you’ll find
in a very short time in
information kiosks at the airport
in shopping centers so we are not
really not far from having this
kind of technology then there is more than
15 years old now I had started a
doctorate in what’s called the
automatic language processing
natural for the purpose of understanding
human conversations so
automatic so we speak we talk but
how can we allow
the machine to understand what they say
then once we understand that is
what the machine can do a return
then what I did well I had
so did my doctorate on it
sure my parents hood is easy
but what art are you going to do
with that can find you a job with that
but I still managed to
work a few years in the
research in different universities
then a few years later I
worked on crime
worry door is it’s the center of
computer research montreal
I was not yet made delinquent
girls my parents was super happy
I had a job I had a few pay
there everything was beautiful and working
at crime I had the opportunity really
to meet several companies from
voices actually what we could do with
artificial intelligence for
meet the real needs of the
life and that’s how I got
said ben actually i think i have
really everything to try me too from
confront everything I’ve learned to
solve real problems
and that’s how I left my
company with my associate charles
olive tree both passionate
artificial intelligence ben on
started working to try to
solve problems including the one that
today we do in the company
which actually allows to provide a
solution for businesses for with
to help them make their customers
happier so if we were able
today to understand everything that
were telling customers via their texting
had said to their emails on
facebook on twitter and we were
able to synthesize all of that and say
finally to the company down and
not look at what was customer and not
are not happy with this product on such
aspect there are things to improve
you imagine how much and we
so that it is less known consumer
would be happy that’s what we try to
solve with with our product
so that’s for sure that what I did
like any good entrepreneur I
started reading tons and
tons of books on how to create a
business
the difficult parts the traps to
avoid all kinds of things but
in hindsight
a few years now I got
realized that in reality with when
there is shown a product that contains
artificial intelligence ben there are
as a degree of complexity
additional that makes the company
becomes a bit more perilous
more complex there are other traps
typical to artificial intelligence
that you have to try to avoid and it’s a
I do not want to talk to you
so when we go to the
marketing of this type of
technology that we want to avoid
but let’s start at the beginning with
beginning what is
artificial intelligence but
artificial intelligence is the
process and that’s the ability of
reproduce human reasoning this
cognitive process which which is included
in a decision in our predictions
so finally we how do we
think about it and what does that mean
it means that there is something
who’s going who’s calling
to knowledge and that’s it a little
the complexity we will find with
artificial intelligence us as
that being human has plenty of
knowledge we use full of
knowledge we have acquired from our
lived our studies but it’s done
in time so how do we
can allow the machine to do
that quickly and have the whole hall
knowledge actually needed to
decision making
that’s what makes that
it’s difficult that’s what makes this
technology still relatively
more complex than technology
standard I will go so much
stubbornness forgiveness to the technologies of
information
so to answer seven accomplices
complexity there is a solution that this
solution is learning
automatic
So what’s the learning
automatic maria the itc an algorithm
who has the ability to learn a
general scheme a global scheme to
from a lot a lot a lot of
data that describe this schema
there but individually so
if we take for example the
facial recognition
so now there are plenty of systems
who are able to recognize a
person generally by
relation to an animal a dog or a cat
how come it actually is
will give the algorithm thousands
even millions of people picture
then this algorithm he will learn to
recognize globo traits that make
that they are distinctive with respect to
the animal and that’s really what makes
that the voice algorithm learn to
recognize the face of a human being
but obviously when we talk about
a lot of data we will say ok
but we will go get them where these
ball data we are lucky since
a decade or so about where you
hear a lot about the term big
data so what are these
millions of data about everything and
whatever
and then thanks to the advent or where the
face the ease of access to the
storage and calculation capacity
now it is available almost
to anyone any
company can have access
to his computing abilities and the
ability to store a lot of
data you know the companies
like amazon google microsoft every
genres and all these companies there
today
they offer this type of infrastructure to
shots simply derisory and
that it really helped to indulge in a
great boost to research in
artificial intelligence because
suddenly people could pick up
data you want could
improve the algorithms and so
spend more quickly having a
better maturity is passed to the
marketing and so that’s what
I want a little elaborate on it
so beyond the complexity I
would say scientific intelligence
artificial
when we go to the level of
marketing via a new
complexity another layer that
add up to that and then 1
there are about three elements that
I ‘m going that I would like I would like
speak to you then the first is
clearly avoid having a tweet
technology
in other words, as a scientist
we can very quickly leave on a
ambitious project that solves the
problem a universal problem and one
can spend his life pillon this
will not happen anywhere he’ll ever be
marketed so a first thing
it’s really focus and on an ace a
very particular problem and why
it’s important it’s because as I
said it just now when he
something that touches on reasoning
human to knowledge ben if we have
need a lot of knowledge
it’s difficult to take the case of a
system that makes dictation
automatic
if we were to develop this
system in the medical field ben he
should just learn the vocabulary
medical
but if we wanted to sell a system
that allowed to dictate for
Mister everyone
it would be the medical field that would be
finance will be the hotel would be
a multitude of areas that makes that
it’s really hard to put any
this information there in this system
is the solution to that is enough
simple is to try to find
a company that would actually be that
would live a problem in particles
and that intelligence could solve
and try to work together so
on one hand it will allow to have
information and secondly it’s fine
allow our element very
important is to recover the
feedback from our future customers
once again avoided the trip
technology is also knowing
what helps the customer and then what
is not she does not help them and that’s it
brings me to the second point the second
element is to be able to explain this
what brings artificial intelligence
why why me as a
who your customers who will buy
a technology I’m going to fetch
i’m going i’m going to want to buy your
technology because it intelligence
artificial is actually very
often what we will find in the
market is it’s technology
that they have a problem
and when we arrive with a
new technology that includes
artificial intelligence I had
already solutions
why mine will be better so
there is quite a whole a notion of
awareness why this one in
particular which contains a
artificial intelligence sees to be
better and for that one more time
there are several elements there are several
strategies one is to be able
demonstrate the contribution of this type of
technology the second element that’s
allow a little to show other cases
of use of intelligence
artificial and how that
actually improves things
today we will have systems
who are in the medical field who
are in the trade there are several
you look amazon goes back there
there are also intelligence algorithms
who will refer you to other products
who you will love because you
bought another product
particular so that’s a little do this
parallel and build on on this success
the year the third element is who and what
not the least I would say is that of
the promise we make to our customers
when we sell them beyond this type of
technology us as a human being
we are very often we work very
often in binary worried is that
it works it works it does not work
it’s one of two things
near
it’s not the right message we mean
another client but yet this is the
reality
let’s not forget we’re in the process of
reproduce human reasoning then
even us as a low human
we are wrong we are we hear the same
facts but everyone will react
we will answer a different answer
on the other and we can be wrong
so it’s normal that also as
product there is so a certain margin
of error there is some a certain
level of error is so at this moment
here it’s all about managing the tent
of the customer how can we
explain who can have
the mistake then to reassure him too
show that we have the ability
opportunity to allow him to
fix the system
so there is really it’s all a
paradigm that will change and that it takes
really take the time and do the
things for them to understand and he
put confidence in somewhere so
these are really this type of challenge I
would you say that they are things that
add up to complexity
what are we going to have when we create a
new a new business
so does that mean that the
rate of i would say two successes is
less good for companies that
use TV intelligence
artificial than other companies
clearly no I would say to you
since 2011 there is still a lot more
of companies that make intelligence
artificial actually that sells
products with a strong component that
use artificial intelligence and
with that it’s going so well for this
kind of technological also the
investors started investing
much more in this type of
technology so that’s good it’s happening
well I think the revolution of
artificial intelligence has begun
then she is here to stay for sure
that if it brings other challenges that I
will put more like challenges
socioeconomic
just to give you an idea
since 2015 there is an initiative that is
led by several countries the states
united canada then other countries in
europe to just think about
the impact
what is the impact of this revolution
of artificial intelligence is this
that we actually have robots that go
we take our place ourselves as
that being human at work
is the social notion of interacting
between people is going to be different
is that tomorrow we will be friends
would be robots so that’s all
kinds of debates really that they
people really take time to
think about it then I can you
to say that it’s serious enough for
drive the initiative this initiative
over a period of a century
so it started in 2015 it’s going
to finish in 1115 is still
still major enough and I can you
also say that I do not necessarily
the answer to that question there and
what the impact is what it means
to say that tomorrow morning we will have the
half of the unemployed population who
will have more work for people I do not
do not think personally but I think
that the response sees coming especially to
as the future and then we
as a human being and then how
we adjust convoy who sees who
let’s build this
this answer to that makes me
leave with with this question there and
then we’ll have the time just now
talk about his thanks a lot
[Applause]
[Music]
Please follow and like us: