The incredible inventions of intuitive AI | Maurice Conti

Machine generated English transcript…

Translator: Eman Shahen Checker: Sarah Shahid How many of you are creative, engineer, designer, entrepreneur, artist, or do you really have a great imagination? Do you raise your hands? (Cheers) Most of you have some news for us as creators over the next 20 years, a lot will change in the way we do our work unlike in the last 2,000 years. In fact, I think we’re at the dawn of a new era in human history right now, there are four major historical eras. They are categorized according to the way we work.

The era of hunting and gathering lasted for millions of years, then the agricultural era lasted for thousands of years, and the industrial age lasted for a few centuries. Currently, the information age has lasted for a few decades and today we are on the threshold of our next great era as human races. Welcome to the era of development. In this new era, your human capabilities will develop thanks to computer systems that help you Thinking, automated systems that help you implement, and a digital nervous system that connects you to the world beyond your natural senses, to begin with cognitive development. Who among you is a robot with scalable capabilities? (Laughter).

I’m going to make the argument that we’re really scalable. Imagine you’re at a party, and someone asks you a question that you don’t know the answer to. If you have one of these, in a matter of seconds, you can figure out the answer, but that’s just a rough start until I see you’re just an unconscious machine. Actually, for the last three and a half million years, The tools we had were completely unconscious. They did what we told them to do and nothing more. Our first tools cut where we hit them, the engraving cuts where the artist shoots. Even our most advanced tools do nothing without our specific guidance.

In fact, to this date, and this is something that has frustrated me, we’ve always been constrained by the need to manually push our will into our gadgets like, manually, literally using our hands, even with computers but I’m a lot like Scotty in Star Trek (Laughter) I want to have a conversation with a computer I want to I say, Computer, let’s build a car, and the computer shows me a car and I say, No, it’s more fast looking, non-German pop, and the computer gives me a choice. (Laughter) That might be a little bit of a conversation, maybe less than many of us think, but right now We’re working on it.

Tools are making the leap from being unconscious to being productive. They use computer-generated design tools and algorithms to shape geometry. They come up with new self-designs. All you need are your goals and instructions. I’ll give you an example in this drone airframe case. All you have to do is tell it something like, that It has four columns, which you want to be as lightweight as possible, and you want them to be aerodynamic efficient. What the computer does is explore the full scope of the solution:

Every possibility is going to meet your standards — millions of them. This requires big computers but it’s up to us for designs that we couldn’t, ourselves, envision and the computers do it all on their own — nobody drew anything and started right from scratch and by the way, it’s no coincidence that it looks like (Laughter) That’s because algorithms are designed to work in the same way as an exciting evolution. We’re starting to see this technology manifest in the real world. We’ve been working with Airbus for several years on this aircraft concept of the future. It’s not yet seen, but recently, we’ve used intelligent design Synthetic produced to come with this computer-generated hologram cabin partition.

It’s more powerful than the real one and weighs half the real weight, and it will fly in the Airbus A320 later this year so computers can produce now; She can offer her own solutions to our specific problems but it’s not so intuitive she still has to start from scratch every time, because she never learns unlike Maggie (Laughter) Maggie is actually smarter than our more advanced design tools What do I mean by that? If her owner picks up the halter, Maggie knows pretty sure it’s time to go for a walk and how did she learn? Well, whenever her owner picks up the halter, they go for a walk and Maggie does three things: she had to pay attention, she had to remember what had happened.

And she had to memorize a path in her head and set it, and what’s interesting, that’s exactly what computer scientists have been trying to get AI to do for the last 60 years or so, back in 1952, they made this computer that makes a great tic-tac-toe sound and then after 45 A year, in 1997, Deep Blue defeated Kasparov at chess in 2011, Watson defeated these two humans at Gabardi, which is something more difficult for a computer to play than chess.

In fact, Watson needed to use logic to beat his human opponents, and then a few weeks ago, AlphaGo’s Deep Mind defeated the best human in Go, which is the hardest game we’ve ever had. In Go, there are more changes possible than atoms in the universe. So to win, alphago had to develop intuition and in fact, at some points, alphago programmers didn’t understand.

As AlphaGo was doing what it did and things were moving really fast. I mean, think about — in the human lifespan, computers have gone from being a child’s play to being the head of strategic thinking. What’s basically happening is that computers go from being like Spock to being more Kirk-like. (Laughter) Right? From pure logic to intuition, will you cross this bridge? Most of you say, oh, hell no!

(Laughter) And you decide that in less than a second you know kind of that this bridge isn’t secure and that’s exactly the axiom that our complex cognitive systems are beginning to develop now. Soon, you’re going to be able to literally show something that you’ve made, or you’ve designed, to a computer, and it’s going to look at it. And he says, “Sorry, dude, this will never work. You have to try again, or you might ask him if people are going to like your next song, or your next flavor of ice cream, or what’s more important than that, working with a computer to solve a problem we’ve never had before.” For example, climate change We don’t do well on our own.

We can certainly use the help that we can get. That’s what I’m talking about. Technology enhances our cognitive abilities. So we can imagine and design things that weren’t within our reach as simple, undeveloped humans. So what about this crazy new thing that we’re going to invent and design? I think the era of human evolution is as much about the physical world as it is about the intellectual virtual world. So how will technology evolve us? In the physical world, robotic systems. Well, there’s certainly a fear that computer systems will take away human jobs, and that’s true in some sectors, but I’m more interested in this idea that humans and robots working together will evolve from each other, and start taking on a new field.

This is our applied research lab in San Francisco, where one of our areas of interest is advanced robotics, and in particular, human-robot collaboration, and this Bishop, our experimental robot, we set up to help a construction person do repetitive tasks — tasks like drilling socket holes or switches. Lighting in plaster wall. (Laughter) Bishop’s human partner can tell him what to do in plain English and with simple gestures, kind of like talking to a dog, and then Bishop does those instructions very tightly. We use humans for what they’re good at: awareness, perception, decision making, and use What robots are good at:.

Discipline and Repetition And here’s another excellent project that Bishop is working on. The goal of this project, which we call the Hive, is to prototype the experience of humans and computers and robots all working together to solve a very complex design problem. Humans represent the labor as they walk around the construction site, and they form the bamboo — which Because it’s a non-formal material, it’s very difficult for robots to handle and then robots twist these fibers, which was almost impossible for humans to do and then we had the AI ​​controlling everything.

He was dictating what humans should do, and robots as well, tracing the path of thousands of individual components. The interesting thing is, building this pavilion wouldn’t have been possible without humans and robots and AI developing each other. Well, I’m going to share with you another project that’s a little crazy. We’re working with an artist in residence. In Amsterdam, Joris Larmann and his team at MX3D are to productively design and automatically print the world’s first autonomously built bridge. So, Joris and AI are designing this thing now, as we speak,.

In Amsterdam and when they’re done, we’ll give a go and the robots will start drawing 3D drawings on stainless steel and then they’ll continue to draw without human intervention, until the bridge is finished So if computers are going to improve our ability to imagine and design new things, robotic systems will help us build and make things New that we weren’t able to do before, but what about our ability to perceive and control these things? What about a nervous system for the things we make? Our nervous system, the human nervous system, tells us about the things that are going on around us.

But the nervous system of the things we make is primitive at best, for example, a car doesn’t tell the city’s Department of Public Works that it crashed into a ditch at the corner of Broadway and Emerson, a building doesn’t tell its designers whether the people inside like it or not, and a game designer doesn’t know if it’s Play with it — how, where, and whether it’s fun or not. Look, I’m sure the designers imagined this way of life for Barbie.

And they’re designing it. (Laughter) But what if Barbie actually seemed to be alone? (Laughter) If designers knew what was going on in the real world with their designs — the road, the building, Barbie — they would have used that knowledge to create a better user experience. What’s missing is a nervous system that connects us to all the things we design and make and use. What if you had that kind? From the information that comes to you from the things that you make in the real world with all of these things that we make, we’ve spent a tremendous amount of money and energy — in fact, last year, about 2 trillion dollars — getting people to buy the things that we made.

But if you have this relationship between the things that you design and make after they’re in the outside world, after they’re sold or released or whatever, we can actually change that, and go from getting people to buy our stuff, to making the things that people want in the first place. The good news is that we We’re working on a digital nervous system that connects us to the things we design. We’re working on a project with some guys in Los Angeles called Bandito Brothers.

And their team, and one of the things that these guys do is they design crazy cars that, of course, do crazy things. These guys are crazy — (Laughter) Best of all, and what we do with them is take the chassis of a classic race car and give it a nervous system, and so it’s equipped with dozens of sensors, and we’ve got a driver with He drove it, took it out into the desert, drove it for a week, and the car’s nervous system picked up everything that was going on with the car. We collected 4 billion data points; All of the forces that were subjected to it.

And then we did something crazy. We took all that information, and we got it into a productive AI design we called Dreamcatcher. So what happens when you give a nervous system to a design machine, and you ask it to design you a finished chassis? You’ll get this. This is something that no human could ever design except for one human who designed it, but it was a human who evolved through the productive design of AI.

And a digital nervous system and a robot could make something like this, so if this is the future, the advanced age and we’re going to evolve cognitively, physically and cognitively, what would that look like? What would this wonderland look like? I think we’ll see a world where we’ll go from manufactured things to cultivated things, and where we’ll go from built things to growing ones, we’ll go from being loners to being connected, and we’ll move away from extraction to accepting assembly. Our world will change dramatically. We will have a world that is more diverse, more connected.

More dynamic, more complex, more adaptive, and of course, more beautiful the shape of future things will be different from the ones we’ve seen before. Why? Because what’s going to shape those things is this new partnership between technology and nature and humanity that, to me, is a future worth looking forward to. Thank you, all of you.


Comments

Leave a Reply

Your email address will not be published. Required fields are marked *