Translator: Michele Gianella
Reviewer: Alessandra Tadiotto As a data scientist, I literally
spend my day predicting the future. I try and model
systems and businesses, and I use really cool tools,
like artificial intelligence or blockchain to help understand what makes humans tick. And what I found most fascinating
from the past few years of my experience and what's also been the most frustrating is that all those expressions
and all those clichés are true. It bothers me; I don't know why. The fact that the journey
is more important than the destination. The fact that to go further
and look further, you stand on the shoulders of giants. That those who don't understand
what history was are doomed to repeat it. In "The Tempest," William Shakespeare
wrote that "What's past is prologue." I think a lot of people
recognize this expression to mean that everything
that happened before was exposition, that to live in the present
is what's important. Data has shown me that that's different. To me, "What's past is prologue" means is that the future can be based
on a barometer of the past and understand the trends, signals,
and patterns of where we've been can show us what we've done before,
and where we're going to go. This presentation
is an argument for basic income. Let me show you how I get there. I start in 1696 in Britain,
as part of the glass tax. In an effort to raise revenue, the British Parliament
created a prohibitive tax on glass and window panes, and the like. It meant that glass and windows
were so expensive that, literally, lords and ladies
would take glass out of their houses and bring them up to their summer castles
for fear of theft or breaking. If you were just a person
living in the city, you lived most of your
internal life in darkness. It affected things that people
had never considered before in different ways. Architecture, interior design - your dining room didn't have
a table in the middle, you'd hit your shins in the darkness. People had a bench
by the side of the room and a large board up against the wall. When it was time to have a meeting
or have a meal together, they would take the benches,
sit everyone down, bring the board on everyone's lap. That's where they had their meeting.
That's where they had a discussion. That's where we got
the term "board meeting." (Laughter) Not only that, that guy at the end,
that guy in the chair, he was in charge, he ran the show. That's right. That guy was the chairman of the board. (Laughter) To me, it is fascinating
how arbitrary choices from the past can affect the future centuries later in ways people can't understand
or think about at all. And we are coming to one
of those critical junctures and choice points again soon. I'm going to skip ahead
a couple of decades to the British countryside. The Church of England,
in an attempt to reorganize themselves, split up the country into various plots
and created parsonages. Each one of these sections had a reverend. His job was to help guide the flock
and all the things that reverends do. These people were given
a strong education. They were given enough resources to understand where
their next meal was coming from, and they were sent on their way. This was arguably one of the first times
a society had a class of people who were academic, well read,
and didn't really have that much to do. What did these reverends do? They changed the world. The Reverend Edmund Cartwright
created the power loom and brought up part of the Industrial
Revolution as we know it today. Automation, improving in process,
happened because of that. The Reverend John MacKenzie Bacon was the father
of modern aerial photography. He showed us the first pictures
of cities from above, revolutionized urban planning
and city design. The Reverend William Buckland wrote the essay
on the principle of population. You know this already
what he learned back then. Cities only grow
when there's an excess of food more than is needed
to sustain the population. The Reverend William Greenwald
was the father of modern archaeology. Before him, people
would just go to dig sites and grab stuff because it looked cool
and brought it home. He was one of the first people to suggest that we apply proper scientific rigor
and scientific method, and revolutionized
the field of archaeology. It wasn't just the reverends, too, it was their families,
their children, their spouses. When people knew
there was a roof over their head and a meal coming to not worry about,
their creativity flourished. I bet you know at least half a dozen
names on this list of people who were either children
or spouses of these reverends - Hook, Wren, Bronte,
Tennyson, Carroll, Jane Austen. These people were so profound in their ability
to push arts and science further that even centuries later,
we still consume their art and the ways it affects us
that have never been considered before. We as a society, we as a culture,
push further and go farther when we take the tools
that have been built before us and go to the next obvious step. Because those expressions are true,
those clichés are all true. Most people thought
it was actually Isaac Newton who said it, but it was Bernard de Chartres who said, "If I have seen further, it is by standing
on the shoulders of giants." The fact that every time that we want
to try to push culture further and make society go
that little bit further, that we don't need
to go back to the beginning is because we use the tools and processes
of people who were built before us. And that's how we advance
as a society and as a people. It has never been easier to try and push these ideas ahead
and move forward. The ability to take this thing over here
and combine it with that thing over there and produce something new
has never been easier, faster, less expensive or less involving risk
than it is right now. I can't guarantee you that the idea of combining
this thing and that thing will work, but you'll know quickly
if it's going to fail, or if it has legs. Never was that way before. Technology usually got in the way
more than it helped. This is special. Back then, it was a problem. Example number one,
let's talk about audio recording. The ability to take a song or music
and record it to a wax disc was incredibly difficult. And if someone made a mistake,
if a microphone screwed up, or if someone dropped
something on the ground, you had to throw it away
and start all over again. Oh, you're already on hour two? Sorry, it doesn't care. Start all over. In fact, the audio engineers, too,
they had a huge problem. Let's say you're recording
a symphony in a concert hall. Let's say if the sound got all screwed up. You can't have a violin and a clarinet
near each other next to the tuba, the tuba would be too loud. The hack from the sound engineers
was to actually move the instruments around the entire theater. So you'd have the flutes nearby. You'd have the tubas in the background. The drums were so loud, they actually were down the hall,
down the stairs in a separate room. They had to create a series of mirrors so the drummers can see the conductor
to know when to play. If you were a soloist,
you were pushed on a little dolly. When it was your turn, they'd push you up to the front,
right near the microphone to hear it. When you were done,
they'd pull you right back again. And if at any time
in any one of those songs if someone, one person, made a mistake,
you threw it out, and you started again. You didn't take risks. You didn't have the time
or the opportunity. Now if I want to try something new, if I want to see what it sounds like
when a guitar gets loud, I just press the guitar knob
and turn it up a little bit. Never before have we had
the ability for technology to make our life easier
and push our creative drive further than we have right now. But it wasn't always like that. It was difficult. If you go back to the war, planes coming overhead
as part of a bombing run was a huge problem. Before radar, people didn't really have good solutions
to figure when the planes were coming. Planes are high. Planes are fast.
Planes are far away. We took the best people
in the allied resource. We took the entire knowledge
of what we have to do to figure out how to find planes, and we came up with this. Giant earphone-like tubes. (Laughter) And a soldier who's hopefully
really good at hearing stuff. (Laughter) Listening, hoping to hear their propellers. Stretching, yearning,
going that little bit further just to hear, to buy
an extra minute or two, so people can run down to safety. We don't have this problem anymore. Radar has basically solved
the problem in many respects. I can go to the store and for a couple of bucks
buy a little radar in a box. I can put it in my car, on my bike. I can put it in my hat
and do a lot of cool stuff if I wanted to. It has never been easier
to give a chance and see if it works. Will that idea work? I got no clue, but I'll know quickly
if I can move on and try something else. It wasn't always like that. Technologies often got in the way. Do you ever wonder why people
in old photos don't smile? Yeah, I see some nodding heads. I'm going to be embarrassed about this. When I was a kid, I literally thought that everyone born pre-1920
was just miserable, sad, and surly. (Laughter) Turns out this was a technical problem. To actually exist and have a camera take a picture back then
required a huge amount of light. It required a huge amount of time to sit there and let the light
expose the film. So if you're going to take
a photo of someone moving, it's going to screw up. So when they'd say "Don't move," you don't move. Because you know if you move, you're going to ruin the photo, and then you've got to start again. This was a huge
technical problem at the time. That's why people were told
to not keep their smiling face because I don't know about you, but smiling for a minute
starts to look insincere. (Laughter) Instead, they said,
"Use your resting face. If you don't have
a nice looking resting face, sorry, that's the photo that you get." (Laughter) It's not like that anymore, right? We've solved a lot
of these technical problems. Most people have a camera
in their own pocket. I take a thousand photos,
pick the best three that I like, and I throw the rest away. I don't need to know
about focus and aperture and all these types of technical terms
to take a fairly good photo. Technology has made it so that the creativity
can push us further than we have before. If you want to build different tools, if you want to bring
these things together, it's super easy. Machine learning, too. If you want to take
some cool new AI things and move to the next step
and go forward, it's fine. You want to take yourself a chatbot
or recognize emotional detection or look at text or videos
and see what's going on, you can do it for fractions of a penny. I just tried this last night.
I built something for this presentation. I went online, didn't write a bit of code, and I can make something
that looked at my face. It can tell if I'm smiling. It can tell if I'm yawning or not.
It can tell when I'm blinking. All without writing a single bit of code. If you have an idea where you're taking
this piece and this piece and mashing them together, you can do it now easier
and faster than ever before. Let's say you want to build
a camera-tool-thingiemajob that if someone takes a picture
of someone blinking, it deletes the photo automatically. You don't need to know
how cameras work. You just take the camera
thing someone built. You don't need to know how neural networks
recognize if you're blinking, you just go grab that software
and put it together. Could you go to market with that idea? Probably not. But you'll know right away
if you've built something that has required you to know
if it has momentum or not, whether or not, you call in an expert
to make it world class. These are the opportunities we have now
to go further and push harder. How good are these tools? They're incredible. Just last year, Microsoft released a tool
for real-time language translation. So picture those guys at the UN with the little earphones
talking in real-time really quietly. Average human error rate for that
is about six percent. That's that little dot
over there in the corner. Machines are now as good as humans at translating language
in real time, at scale, 24/7. Just so you think
I'm not throwing you a red herring, let's talk about image classification. The problem of taking a photo or video
and identifying all the things in it is actually surprisingly hard. Humans will get it right
only about 93 times out of 100. Error, about seven percent. Just last year, Microsoft released a tool,
a machine learning thing with AI, that got it down
to three-and-a-half percent. As of right now, machines
are twice as good as humans at identifying images and objects
and videos and pictures. IBM beat them, too.
They're down to three points this year. Not only that, look at that chart. That's a 10-times drop in error
in only five years. I can't wait to see
what the next five years brings us. Now there's two schools of thought here. Yes, automation is coming. Yes, a lot of jobs will be made redundant. Now, the first school of thought says, "Well, Jim, I don't churn
my own butter anyways." You're not wrong. The progress goes forward and pushes us further
than we've been before. The second school of thought
argues that that's true, but is a most fortunate opportunity
that we have as a people to be able to look in advance and recognize that massive
change is coming. Almost never before have a people
looked in advance and said: That's coming, and we can have
a discussion about what we want to do. We can have a debate
and a cultured conversation about what happens when lots of people
don't have their jobs anymore. I would suggest that for these people,
if we give them an education, if we gave them enough income to make sure that they have
a roof over their head, and if they have enough food on the table, that we can create a class of people who are academic, well read,
and really don't have that much to do. We've seen before in history
that these people can change the world in ways that are still affecting us
centuries from now. This is the opportunity we have. This is the discussion
that we need to have in the next five years,
10 years, maybe 20. It has never been easier for us to take this thing and that thing
and mash it together, and try a new idea to see if it has legs. Let me give you a very quick example. This is a Sphero.
Anyone seen a Sphero before? It's a fun little toy. I got one for my kid for Christmas. It lets you actually use your phone
and connect things together and play. I got one right here.
It's a cute little toy. Totally made for kids for Christmas. And it's great because
it lets my kids learn how to code without them even knowing
that they're learning how to code. Here's what I'm going to do. Without writing any code at all,
what I'm going to do is take a picture of my face
with my cellphone. Then I'm going to connect it to a world-class,
Microsoft deep learning network to look at my face. We're going to tell
if it knows if I'm smiling or not. I'm going to change
the color of this device, based on whether
I'm smiling or frowning. Remember, we talked
about pictures and smiling. Let's see how this works. (Laughter) Okay, there we go. There we go, and there we go yellow. Woo-hoo! Now, the important part
I want to get across to you is I did this writing
ridiculously small amounts of code. Someone already built the tool
that connects to the cloud. Someone already built
the world-class network up there that can look at someone's face, recognize their age, their gender,
and if they're smiling for fractions of a penny. Someone out there was kind enough
to take this little Sphero tool and build a library I can connect into and not worry about
all the stuff about it. But in the abstract, think about
what I just did there. For fractions of a penny,
and maybe a day's worth of my time, I connected to a world-class,
deep learning network that recognized my face. I then used it to change
a wireless IoT device using Bluetooth, just like that. It has never been easier for us
to take tools, to take toys, to play with them,
to see what idea comes further, to see how we can push ourselves further. I look forward to the next five years
because I can't see what's coming. I look forward to hearing
from each and every one of you who has a brilliant idea to take this thing over here,
and that thing there, and combine them together,
and to push us as a people further. Thank you very much. (Applause)
