 The organizers of the ISAL Summer School asked me to respond to these three questions in 10 minutes. Max, that is a tough length. I made a little show and tell, we'll see how it goes. What sparked your interest in artificial life research and how did you get started in this field? You know, it's always been clear to me ever since I was a teeny kid that my mission in life was to make something do something by itself because that is inherently cool. And as I grew up, the power source has changed from, you know, gravity and wind to rubber bands and batteries and black powder and so on. And likewise, my ambitions to make more complex things, do more complex stuff grew as well. And it wasn't until high school when I was very lucky at the time to encounter one of these, a computer, that it was so easy to make this machine do something by itself and it would do it exactly the same over and over and over again. It was incredible. And it carried me along through high school, through undergrad and graduate school. And it became part of, you know, there was this, you know, I think every thinking person has to, you know, confront the big questions at some point, you know, what's it all about? What does it all mean? Why am I here? And, you know, it came to me that my job was to produce a theory of Dave. And, you know, if a theory of Dave was also a theory of Sally or a theory of Joe, then that's great. But if it turned out that I was weird in some way, well then, you know, I'm going to pursue the way that I am and other folks can pursue their own ways. And the computational perspective informed the theory of Dave a lot. But by the time I was graduating from finishing my PhD, Christopher Langton had started this artificial life workshop, the whole idea of, you know, we could build living systems out of artificial stuff and study them. And, you know, I saw the announcement for the first artificial life workshop in 1987, but I was pounding away on my dissertation and I let it go by I still regret not having gone to the first one. But I went to the second one. And I was working with Michael Littman, who was a student at the time, and we had done some work that we wrote up and we sent in to this thing. It got accepted as a poster. Now, I didn't really understand all that was. So I decided, you know, I don't want to be stuck talking to a poster this whole time. So I made a video and I figured I could just leave the video on a TV on a stand and people could watch the thing and then I could go around and look at all the other posters and so forth. But when I got to the artificial life workshop, shortly after it began, apparently they had had some cancellations and our paper had been on the bubble. The last, you know, one of the ones that was closest to becoming a talk. So they asked if I wanted to have it upgraded my poster to a talk. And I said, okay, I was the very last day of the workshop. I don't know if it was the very last talk, but it was pretty late. And, you know, so I basically got up and said, well, you know, I wasn't expecting to give a talk, so I made this video and I pressed play and I sat down and then I took questions afterwards. It was about 15, 16 minutes. Here is a tiny little taste, how I got started in artificial life. I'm Dave Ackley. Let me show you something. I'd like to present the results of some work that Michael Littman and I have done. How can an organism learn during its lifetime, given only its own death as feedback? Here's the data we obtained. Of course, if you took a stable population of natural guys like this and stuck in a couple of those hand designed guys, they'd take over in a couple of generations, wipe these guys out. I guess that's life. I guess that's life. And, you know, did a bunch more models like that, a bunch of work with Michael Littman and just took off from there. Number two, can you explain one of your most exciting or surprising findings in artificial life in a way that a beginner could understand? I'm not completely sure. Originally, when I was going to answer this, I was going to go back and do another one from the early days, the evolution of communication stuff, because that had some cool stuff in it as well. But instead, what I thought I would do is leap 30 years into the present, more or less, and try to show a tiny little demo. We'll see how it goes. Here we are. Okay, this is the code behind the paper that I'm going to be talking about at the artificial life conference real soon now. And I just wanted to let it start up a little bit. So this is the initial start seed, a single atom that spreads out and starts to unfold. And it makes this diamond shape, which is all doing all this complicated stuff. And you can see there's this little bit of a loop here, which is actually going to become the equivalent of the digital DNA for this creature. But what I want to do right now is just pith this, get rid of that code. And yeah, and you notice it actually takes the rest of the code takes care of itself, just to show you the underlying behavior of the diamond itself, which is the diamond is a circle measured in Manhattan distance. So these are these are really kind of spherical cells loosely. And you see the thing grows. And it moves around. And you know, that was sort of the beginning, the table six, how can I get a big object to move? Now, can you see this, this red jagged area there? That is the size of the neighborhood in this cellular automata. It's all local communications. We asked that guy in the middle that being labeled hg. So what do you want to do? And it can look at all of this stuff in that little red lines, and decide what to do, it can change itself, it can change the stuff that's in there or both whatever it likes. But then it's done and it goes on and some other event happens someplace else and you can see them flashing around. That is the whole gist of it. But we can build up and build up to, let me get some wall here. Okay. So you see the thing moving around, but you know, like we can make an obstacle, right? And it moves away, the thing moves from the obstacle, we can make another obstacle, you know, and so on. And the thing reacts. So even though these things are made of these tiny, tiny little windows, and there's no global communication, there's nobody in charge. There's just this complex set of rules, which, you know, I wrote in this particular case, only to see the thing actually is like growing now as well, and so on. So this, I think, is a lot of fun. To me, it's satisfying, you know, 30 years after doing the interactions between evolution and learning. I mean, that was fun too. But this takes it to a whole other level because I did not program the organism directly. Now, we're programming the chemistry, we're programming the physics underneath, and then observing this stuff. And so for example, you know, I don't know if this is going to work. This is probably a mistake, but what the heck. You know, I can come here and erase a bunch of this. And it regrows, and there's a and it regrows and it successfully moves again, because it grew from a single atom to begin with, it had to deal with the vagaries of being running into obstacles and so on and so forth. So I think this is this is very exciting, and whether it makes any sense, well, it does, you get it, right? You can make these things, and then you step back and they do stuff by themselves. So, all right, you spoke physics, artificial chemistry, that's what we're talking about. Number three, what is one big unanswered question in artificial light that you think could be important in the future? Now, this is going to seem like a stretch, and maybe it is, but it's a my sincere answer. I think one big unanswered question is how can artificial life science and engineering help save democracy and make a more just and robust society? And, you know, to me, the big picture, all of this stuff, making something do something by itself, is all about studying how machines work. And very quickly, you move past a single machine that's completely deterministic, and you get collections of machines, and you get redundancy, and you know, even the people that do the domino setups that look like it's a single step by step by step, when you talk to the folks who do the really big ones, they have all these tricks, like they have alternate paths that the things are running down. So in case there's a screw up on one path, the whole thing doesn't get lost, and so on. And all of that comes to the front when you take this idea of indefinite scalability seriously, that we're going to build machines that you can make them bigger and bigger and bigger and bigger, instead of the idea that you have a CPU and a RAM, and that's it, it's fixed. You know, once you run out of space there, then start over. So I believe A-Life can be the poster child for this alternate approach. Instead of top down, maintain control, flatten everything out, cut out the middleman, all efficiency, no robustness, bottom up says encourage agents that learn skills and then put those agents together to contribute their skills and collaborations, and that is going to be the basis of a more just and robust society. We'll see.