Conversations on Applied AI

Dr. Antonio Di Fenza - Artificial General Intelligence Through the Eyes of a Futurist

• Justin Grammens • Season 3 • Episode 5

Share episode

The conversation this week is with Dr. Antonio Di Fenza. Antonio has a Master of Arts and a Ph.D. in Romantic Languages, Literature, and Linguistics from Cornell University. He has envisioned and written a project of a chatbot trained on philosophers and poets that is capable of meaningful interactions. Additionally, he is currently a futurist providing inspiration, suggestions, and solutions for the period of humanity's transition to the time when it will coexist with super intelligence. 

If you are interested in learning about how AI is being applied across multiple industries, be sure to join us at a future AppliedAI Monthly meetup and help support us so we can make future Emerging Technologies North non-profit events!

Resources and Topics Mentioned in this Episode

Enjoy!

Your host,
Justin Grammens


Antonio Di Fenza  0:00  

So people that are you know, complaining, including me like, Okay, this is a little boring if you look into the future a couple of years, not 20 Because that's impossible. But But, but if you look at the future, like a couple of years, it's really hard to be amazed by the power that are being unleashed. You know whether or not AGI takes place, this is OS. What is that? AI is so central because it's it's going everywhere.


AI Announcer  0:30  

Welcome to the conversations on applied AI podcast where Justin Grammens and the team at emerging technologies know of talk with experts in the fields of artificial intelligence and deep learning. In each episode, we cut through the hype and dive into how these technologies are being applied to real world problems today. We hope that you find this episode educational and applicable to your industry and connect with us to learn more about our organization at applied ai.mn. Enjoy.


Justin Grammens  1:01  

Welcome everyone to the conversations on applied AI Podcast. Today we're speaking with Dr. Antonio Di Fenza. Antonio has a Master's of Arts and a PhD in Romantic Languages, Literature and Linguistics from Cornell University. He has envisioned and wrote a project of a chatbot trained on philosophers and poets that is capable of meaningful interactions. Additionally, he is currently a futurist providing inspiration, suggestions and solutions for the period of humanity's transition to the time when it will coexist with super intelligence. So that's awesome. I'm super excited to have Antonio on the program today. Thank you for being here.


Antonio Di Fenza  1:33  

Thank you for having me.


Justin Grammens  1:35  

You know, I talked a little bit about you know, I guess your your degree and your background, but you know, how did you land in this space, I guess, based on all the stuff that you've kind of like learned in university and gotten to where you are today?


Antonio Di Fenza  1:47  

Well, there was a towards the end of my PhD project. Basically, my project was on literature and philosophy, linguistics, these kind of things. And I was launched towards the academic my academic career become professor. Stay, yeah, they will do literature and these kind of things. But towards the end, I realized that I was focused with my studies on the past. And also, I wasn't seeing what I was doing as the best use of my skills, my intelligence. So this was one part of it. And the other part was that I began reading about AI and exponential technologies during that time, and the implication of what is happening currently, you know, blew my mind, and I wanted to know more and hear more. So I came back from the States, I had to take a long break, because I was I was ill. But during that time, I read a lot. I read a lot of books, and I read all the classics. So singularities near from Ray Kurzweil and then I and then I read then another number of books, which were a little bit more recent, like books from Steven Kotler or even super intelligence by Max Tegmark life 3.0, or like an all this kind of kind of people. And so I began to be interested in try to understand a little bit distinct about AI AGI or artificial super intelligence. And then and then at that point, I was, I was mesmerised I was, yeah, when kind of crazy because, because I could see all the ramifications that this research has not only for our society, not only for how we experience the world, and we experience ourselves, but also for what it means to come to knowledge in itself to be able to produce more intelligence, whatever that means. And so and so this is, and so I read books, and then I got tired of just reading books started reaching out to people and I was like, Okay, I'm gonna do another PhD now. Maybe cognitive sciences, maybe something else. But then I met this guy who said that you crazy. Why another PhD? Why don't you do something different do a bootcamp. So data science bootcamp. Okay. I had no idea about all these things. And I had no prior experience to coding. And I started doing and I and I prepared and I and I, I joined this boot camp. And then I did, I did this thing for three months in Berlin. And then now I'm after almost a year, I've done a couple of projects. But I'm still trying to figure out what I want to do with this. If coding is my way or not, because I like coding, but I like words and ideas more to be because, you know, data scientists are experts, users of the tool. But what I realize is that when I was talking to my peers, for example, that despite them being more skilled than I was because I was the only non quant background person in the room. None of them had read, you know, the books that I had read and none of them could see the potential. Well, none of that They couldn't see what this technology is leading us. But they could not appreciate the breadth and the length of the kinds of things that I can envision. So do I want to be a like a, like a tech guy? Or do I want to do I want to do something more higher level to work to help people map the world that that is that is unraveling before our eyes at the increasing speed, as you know, of course,


Justin Grammens  5:27  

yeah. Oh, man, a lot there that you talked about? I mean, you know, you're talking to somebody who, who is probably one of those mathematical thinkers. That's what I did in college. I did, I did math and physics. And then I got really into software development and got a master's in software. And so I know what you mean about just getting in the lane, right? Yeah, I just get in the lane. And I'm just gonna write code. But I do think increasingly, more and more people are hiring and getting hired because they can take that 1000 foot view. So my perspective is, is it's good to know the details of it. But not, you certainly shouldn't feel like you're stuck have to be stuck in that every day. Right? If you want to level up and use all that information that you've gained, to have a more broader perspective, I would feel like there's there's still plenty of opportunities out there for that. I mean, you're doing some consulting a little bit with companies.


Antonio Di Fenza  6:14  

I'm doing some coaching right now. And then I joined the think tank that specialized on transhumanism, you know, merging with AGI, unleashing the singularity, like the guy who runs and was built this, this, this think tank because the hardcore transhumanist, I was written like five books in these issues. And so I'm doing some work with him. I'm producing content, I'm writing articles, doing research, and trying to figure out where I'm going with this, which is not clear here. Because we're like, making all of this knowledge practical right now is becoming a little bit of a struggle for me, everybody, everybody's talking about AI right now. Okay, and all the tools that are coming out, then everybody is getting an understanding of what these instruments are. But my interest is a little bit higher than this. It's like, okay, but what if, what if, and when, you know, things become a little more complex? What if we nail AGI people are single, you know, maybe AGI is never going to happen. Okay, but but we have somebody like Elon Musk, who build a company, which neural link to actually provide the infrastructure that we might need. In case, we decide that the only way to go is to merge with AGI for example, which is I'm trying to read a an article about this. And it's just kind of crazy. But but this is where we're going.


Justin Grammens  7:45  

Are you scared about AGI and by the way, if you want to mention the name of any of the companies or whatever, you know, feel free to do that, because we have a transcriptions. And we have like liner notes for each one of these episodes. So feel free to talk about any of that stuff if you want. But I am curious about this think tank or stuff you're writing are you? Are you scared? Are you concerned? Are you are you happy for AGI Where do you sort of feel like we're headed right now?


Antonio Di Fenza  8:07  

No, I'm not scared. I'm an optimist. I don't think that everything is going to be rosy and easy. And that we are going to you know, unfold the hand and on earth without any struggle and seamlessly as though we I don't know we're building a video game or something. I don't think this is the case. But on the other end, I'm not freaked out. Because I'm some I'm entranced by the capabilities that we are unleashing, right now, even without the AGI, we are increasingly becoming more capable of manipulating information and matter to avoid that we are always more powerful, and so and so and so face this last bit could come up come around, which is AGI, then, then I will, then it's really hard to envision how the world is going to look like especially if we're going to merge with it, for example, which is again, only one one way to go and it's scary. But then again, if you think about it, it's not even that strange, because we are already in deeper relationship with our tools. Always use our tools. And so neurons neural link, for example, is just an end we already carry tools in our bodies. We get heart surgery, pacemakers and and so so so we are we've already become cyborgs as he says so it's kind of normal. It's just with the brain that we're a little more concerned and freaked out rightly so I think that's more excitement because because a lot of the great scholars so you know sluice Russell's to sweat Trussell or Tegmark or Bostrom everybody's really focused on on talking about the risks, the risks, the risks, we need to get it right. We need to get it right we need to get it right because if we don't get it right, things can go really bad. it, and I get that part I was actually helping. Are you familiar with the Millennium project? John Glenn. So it's some international world organization that has different chapters all over the world and has laid out some some key, some key areas of study, you know, food, water, you know, climate change, and then they and they mandate and they they produce research, basically, and they help governments like figured out their policies to address these issues. So I'm reaching out to a lot of futurists. And turns out that this John Glenn is producing the first paper about AGI for the European community, because basically, politicians still don't even mention it. And so on. So I helped him a little bit, you know, I read this paper, and I am helping him a little bit to to improve it.


Justin Grammens  10:51  

Yeah. So you're still pretty optimistic on this on all technology? You mentioned the word transhumanist? I think, right? Maybe you could define what that means. I think it might be outside of the general vocabulary.


Antonio Di Fenza  11:03  

So transhumanist is somebody who claims the interaction with our technologies is going to be more intimate over time to the point that we're going to get incorporated technologies that we deal with, for instance, you know, if you think about neural link, so neural link is right now is a Bluetooth, micro computer, it's as big as, I think, 25 cents, or even smaller than that, that that you that you that you apply on your brain, and you connect it through some very thin wires directly on your brain basically. And the idea of this tool is that for the first thing you can do is you can access your phone and your computer, telepathically, basically, and this is already been the most demonstrated, like monkeys can do that. They can play Pong, and they can write things, just thinking of thinking, just by thinking, the cool thing is that this tool, the idea is that this tool can restore vision, this tool can help you know, people to acquire, again, movement where there has been paralysis, because the idea is that the human body is made of wires, and electricity, and sometimes these wires, which is the spine and the nerves are in get interrupted, breaks. And so with these tools, you know, you know, we can resort that. So transhumanist is a person who says, Okay, we're going there, and it's fine. We like that, because there is no thought because they see that not only we can restore things that once were broken, but we can also augment our, our cognition and our powers through these tools. And they will come this, this progress, this change depends on how you want to call it because depends on where you stand. But this kind of change is welcome an increasing merger with the technologies that we use every day. transhumanists are okay with that. And then there is, you know, there is the element of, you know, the longevity, because people claim that we might be able to hit Escape Velocity, you know what it is, it's, it's the idea that, at some point, for every year, you live scientists gonna be able to elongate your life a year or more. And so


Justin Grammens  13:20  

and so live forever. One, the


Antonio Di Fenza  13:22  

idea is that you buy live indefinitely, forever, nothing is forever, not even the universe. But But indefinitely. But the movement is very wide. And there are a lot of different people in it. So a lot of different things thinkers, a lot of different takes. But yeah, so this is what transhumanism means.


Justin Grammens  13:39  

Very interesting. Okay. Well, good. Yeah, no, I sort of heard of it on the peripheries, but I actually hadn't hadn't heard somebody actually define it. So I like that. And obviously, there's a lot of books in this. I like what you said at the beginning with regards to you didn't want to be reading about the past, you're more interested about reading about the future. And so those are the types of books those are the types of, you know, specialties and subjects, those types of authors that you're sort of reading is in this transhumanist area, is that we think you should continue to spend your effort or are you finding other areas within that that are more exciting,


Antonio Di Fenza  14:11  

more exciting than AGI and the singularity? No, ha I don't think there is anything more exciting because it's so if you think about it, so the idea of understanding because you don't need to produce intelligence you need to understand intelligence and creativity. One could argue that the same thing or not, but we could discuss about this because intelligence is the recognition of patterns that then allows you to do a lot of things right. But creativity is basically needs also this this understanding of patterns to generate so so anyways, so there is this idea that intelligence is this basic element of the universe. It's not just us. It's not doesn't belong just to humans just belong to animals. I think it's belongs everywhere because because every time you see an orderly structuring of matter that answers to a problem somehow, I think that's intelligent. And so, so the fact that we are understanding intelligence to the very deepest level, it's the Holy Grail of knowledge, like, whoa, and the Holy Grail of philosophy, the holy grail of science. Like you couldn't get any deeper than that. I think, and this is why I don't I don't think I would ever want to care about anything else. I mean, besides their thing, yeah. Super exciting. Yeah.


Justin Grammens  15:33  

Well, you know, you talked about literature and linguistics and writing and stuff like that, and sort of the buzzword, or the buzz technology going on right now. Is all this stuff with ChatGPT? Yes. Right. So open AI and all this ChatGPT it's all over the news, man. I can't I can't open up another browser window without any more about what are your thoughts? Have you had a chance to experiment around with it? Or do any doing stuff with it?


Antonio Di Fenza  15:56  

Yeah, I did, actually. And I'll tell you something, like, as much as I'm excited with all these tools, I like to read about them. I read late lately, I read about them more than then I use them. So me journey, like all these tools that people have been using to produce even art and then but okay, but with ChatGPT, I used it quite intensively for three or four days, because I have heard a lot about the writing capacity of this this tool. And I said, Okay, you know, let me give it, give it a try. And when you open it, it says that this duel is optimized for chatting, and I tried to chat, okay, and it was pretty boring. Because every question I will make, I will try to poke here. Are you conscious? Are you aware? I realize, can you tell the truth? Only answers that I was getting over and over was like, No, I'm a model. I'm not made to be sentient. And I'm not meant to be aware. And so I was always getting like this kind of answers, which was like, Okay, you're, you're you're a little boring. I like to talk to the other. With the other models from open AI, they will claim that they're sentient, they will claim that they're conscious. And when you ask them like, Okay, what do you mean, they give you an explanation, which are many times make sense. And so it's interesting to fool around anyway. So I played with ChatGPT for a little bit. And then and then I was like, Okay, this is boring. So I left it for a while, then I was reading an article, I'm hearing that some people are complaining and saying that the tool the model is not very deep, that the writing is superficial, that the reading is shallow seems articulate, but it's not. That's what it's for. But but the way I used it was different because I gave it formed paragraphs. And then I say, Okay, can you edit this paragraph? Take care of the grammar, the misspellings, and you know, all those some inconsistencies between verbs and so on. So the result was was great. And then I was saying, Okay, can you flip the order of the ideas? And then and that was done too. And then I was saying, Okay, can you optimize? I was I was using terms, like rhetorical terms, can you optimize for clarity? Can you optimize for cohesion? Okay, can you optimize for coherence? And all those things were happening, or at some point, I also said, you know, I would like this to sound a little more dramatic. Can you do that? And that happened, too. So I ended up spending about three days co working, doing it, like back and forth, but I was not just being fed the things from from I was giving it my own ideas and say, Okay, can you at some point, I said, Okay, give it a paragraph? Oh, well, I had an argument. And I said, Okay, what do you think about the argument? Is it weak? Is it strong? What is it missing? What are the shortcomings of this argument? Can you list them? Can you tell me what else until the end? And all these kinds of things were answered? Always and I wasn't necessarily having the entire time brilliant answers. But that doesn't matter. Because if this is the level of this technology now, what's going to happen in a year, two years, three, five, so people that are you know, complaining, including me like okay, this is a little boring. Yeah. A few fuel. Just just, you know, look into the future a couple of years, not 20 Because that's impossible. But But, but if you look into the future, like a couple of years, it's really hard not to be amazed by the powers that are being unleashed right now. To us. You know, whether or not AGI takes place, this is awesome. What is happening and AI is so central because it's it's going everywhere. So yes, it's going to everywhere. And it's allowing us to do things that we could barely dream about a while ago. So I don't know if you heard from the Washington Post this morning that came an article, I think, or yesterday or this morning, an article that says that there are some, I don't remember the university in the US. But basically, there are some some, there's this article that says that scientists nailed nuclear fusion, what is nuclear fusion, it's basically the same process that takes place in the universe, when a star comes to life. Hello, we can produce the same reaction to bring stars into existence on earth. You don't think that these people that the scientists were we're using deep learning models to understand this this? It's not just deep learning? Of course, yes. But you don't think that AI plays any role in this?


Justin Grammens  20:58  

Fantastic, yeah, that's, that's one of the reasons why I really enjoy doing this podcast is because, you know, the whole point is applications of artificial intelligence. And it feels like the applications are never going to end like that. The next, the next guest I get on is going to be talking about something where AI is applying to a completely different, vertical, completely different business or a completely different, you know, problem, I guess, or challenge that we have in the world. And it just continues to sort of build on top of each other. So it's, it is fascinating. You know, it's awesome, thank you for sharing your story about your experience with ChatGPT. I think it's kind of mimics kind of mine as well, right. I got on there. I was trying to, you know, the one of the first questions that I asked it, I was like, I was like, tell me the value of pi. Wow. Right. And so I sat there for a long time. And I was like, what's it going to do? Because I was hoping it was just going to be dominant is going to get into this loop, you know, where it was a 3.14159. And it was just going to continue to sit there. And so it sat there, I was like, What's this thing going to do? It finally came back and said that it's that it's an irrational number that it goes on to infinity. And that's all it really gave me. And I said, well give me pi to the 50th place. And it did, it basically wrote it out on the screen. And so I was like, well give me pie all the way to the end, right. And so I was I was, so I was taking the stance kind of you are like, well, let's just manipulate this thing was sort of like, kind of push it, see if I can find its weaknesses. And once you step back, like what you did after a couple of days, I think about it's like, well, how can I use this as a tool? How can I collaborate with this thing, then I think that's where that's that's where I think it's going to be really cool is is we're going to realize as as a as a as a race as a human race, that you were sort of better together, you know, with these with these machines. And I haven't used it to write paragraphs and stuff yet. But I have used another tool called copy.ai, which I'm finding has been really, really interesting. I tell people, I'm just I'm not a very creative writer, I didn't come up writing a lot, I came up more on the science side. But what's cool is, is you can give it a bunch of broad concepts, and I'm sure they're using GPT. Under under the covers, in some capacity. Most all companies are and you give it these broad concepts, you sort of say, well, I want to sound strong or authoritative or wanted to sound cheerful, you know, you can give it some some sort of ways. And then it's sort of spits out a outline for you first that you can sort of tweak and adjust. And then you can click another button, and it just essentially dumps out, you know, like a, like a, like 1000 word sort of essay, and I found it to be very powerful. Now, this isn't something that you would just publish on day one, it what it does for me is it actually brings out some other concepts that I'm like, Oh, wow, it kind of went down this path of writing the article this way that I would have never thought of. And so I lift and pull pieces out of there, and then I expand further on it. But it's a great sounding board. It's a great sort of like, person riding shotgun, or like a wing man, you know, to basically have a little creative writer next to me. That's where I think it's it's really, really powerful. And I think that's where we're going to see some really interesting, interesting applications of it in the future.


Antonio Di Fenza  24:00  

Yeah. And, you know, I also tried a similar experiment with the, for instance, has had the, you know, Elon Musk, I had this neural link presentation a couple of days ago, and I want to write an article. Can you pitch me five ideas? In a three line for each three lines for each ideas? Yes, of course. And so they gave me five ideas. And I was like, Okay, I like number three, can you expand a little bit on number three, and then and then and then then it got larger? And then I said, Oh, x. So the thing is that if we you let the model do everything, it's really hard for now at least, to come up with something deep, because this is this is what should high level quality writing is has some depth as breadth, as you know, with a number of these these kinds of things, but I don't think that machines are ready Do that. However, I haven't seen it. Let's say that. But But But if instead of just expecting the thing to do everything on its own, you kind of say, Okay, let's do this together and help me thinking through this. I can assistant like, like a very patient assistant, actually. Because you can ask to do something over and over a million times, it will never tell you to go to hell,


Justin Grammens  25:24  

which is great. Yes, you're right.


Antonio Di Fenza  25:27  

paranoid about getting something right. Or like the way you your way, let's say, okay, you can ask it 100 times, like, can you redo this? Can you redo it? Can you redo it like this, changing this, this comma changing this article, switch to cheerful, a little less or a little more? Oh, you know, so then. So it's bringing the creative process on smoother path. And it's becoming less intimidating, also, for people to sit down and create, because because what really freaks people out while writing is the beginning. Yeah, that white age scares everyone. And it's, it's really hard to feel that first web page. But once you can do that, and something starts coming up, then then it's a matter of moving things up and down, and then add in between, but it's easier, it becomes increasingly easier in the process. And so I think these tools are helping us with exactly the first bit, which is the beginning, making it less painful, so to speak, we'll see where it takes us. But there is a very smart man, actually, who wrote the post where he was complaining about the fact that we may, that by outsourcing our writing to machines, we actually might lose it, that the capacity of writing, so maybe our skills were going out of he was talking about, and this is something worth thinking about. But on the other end, it reminds me of Plato, who wrote this, this dialogue in 400 BC, about writing like, like, above writing, because when writing came, came around, he was like, he is writing a book thing for us. And Plato wrote a long argument saying, well, not really, I don't know, if he believed that, but he wrote this, this this counter argument to writing it says, Well, what's going to happen to our memory, once we start writing everything, because our memory is gonna is going to become weak. But okay, look at what happened, maybe our memory did become weaker. But on the other end, the the amount of things that writing has allowed civilization per se, is are basically endless, which means that maybe some skills get resized somehow, but then other skills enter into play and new ways of interacting with the world take place. So So I'm thinking that probably this this very same thing will happen right now, maybe the meaning of writing will somehow shift and change. And maybe, maybe, I mean, I like to write and write a lot maybe I will become a little less skilled in carving perfect sentences maybe. But But I can also see you know, other skills popping up like Okay, I will learn to you know, to play with 10 ideas and see them all written is supposed to be and you cannot do this when when you when you have to write yourself, you cannot ask you cannot write 1000 words, 10 times, for each kind, it will take forever. At some point, you have to be more economical when you when you want to write, and so on. So there are a lot of other things that can come around, which are


Justin Grammens  28:54  

awesome. Yeah, yeah, for sure. I mean, you can't, you can't have the ways of the past and the ways of the future in some ways. So they can feels like you have to give up something in order in order to change in advance. So I think about Yes, like, once people created the calculator, then you didn't actually have to do longhand arithmetic anymore, right? So we became not good at adding and subtracting multiplying and dividing. But it was that so bad, right? We have computers that do that. For us. I think about the keyboard. Like my handwriting is terrible. I can't my hand I can barely even read my own handwriting. And it's largely because I'm on a keyboard. I don't really handwrite very much stuff at all anymore. Right. So I've lost that skill,


Antonio Di Fenza  29:35  

but you acquired another one. So you could write do use CodePilot?


Justin Grammens  29:39  

You know, I have not used CodePilot yet? No. I haven't delved into that. I've read a lot about it. I've seen people using it. I think it's going to have its own applications. I think it's going to be the same way honestly, it's gonna give you it'll give you boilerplate stuff, but once it comes down to the details, I still feel like there's going to be a lot of inputs that the human is still going Going to be doing but of course, running it back through an algorithm. There already are there are already optimization algorithms, you can basically have your code, like run through to make sure Oh, am I looping this the most effectively, waves are different suggestions. So but yeah, I mean, I can write a lot faster using a keyboard than I can, if I were to be handwriting everything out longhand, right? So except there's a huge benefit.


Antonio Di Fenza  30:21  

As long as you don't just lose things. We keep acquiring some others, I don't think that there's going to really going to be a problem, I don't think that we're going to become stupid, I can see that the internet for all day being flooded by No, God knows how many, how many of the things that we read already are, you know, written by AI. But on the long run, I would just going to shift our position and learn to do other things, because we can focus on like, for example, when I was doing business, when I was reading this paper with Jupiter chapter, I was more I was very focused on details, something that I can also that I can do only towards the end, normally when I'm writing something myself, so I have to spend a lot of time like putting things in order. And then after I've done most of the job, like I can, I can, you know, look at details. But with doing this piece with the, with the model, I I could pay way more attention to way more attention to details that I could otherwise do. And, and I got there faster. That's the thing.


Justin Grammens  31:29  

Yeah. Just what I think the other thing that sort of popped in my head of me talking about writing was, you know, I, you've used the words, knowledge and intelligence before, you know, as you've been talking about it, and I think, I think these machines are becoming very knowledgeable, obviously, they're being fed billions and billions and trillions of words, right. And they understand second, like sentence structure. So I would argue that they're becoming very, very knowledgeable, intelligent, but they're not becoming very creative. And maybe that's, that's what my that's where my head is going is, is I think the humans still have the creativity, they're able to take a look at things that have happened in the past, I feel like and create something new and unique. That hasn't been hasn't been done before. And I feel like, at least when it comes to artificial intelligence, it's still basing its future work off of prior work. And it's not actually bringing in that creativity like that one plus one equals three. What does that make any sense to you? Does that does that jive?


Antonio Di Fenza  32:23  

I know that I know. Yeah. I don't know. If I agree with this, you know, because I don't if you look at philosophy field can take my field, you know, philosophy, literature, you know, and another field with extremely creative people. Now, what does creative really means that you come up with something that never, ever, ever? Anyone ever thought about it? The answer is known. You What do you do when you create something new is that you take things that have been done said and thought in the past, and you rearrange them to a novel order, and you repurpose, but there is but you're very, very much situated in a continuum, where you don't have emptiness behind you, you have whoever else came before you. And so you're not really because the work to create it's actually a very interesting word, because it's a biblical term. And it technically means to make something out of nothing. So to create this the activity of God, who creates the world out of absolute nothingness, that's to create, okay, what we do, it's not really create with generate, because, because the word generation means you know, life from life, so to speak. So things from things, if you look at the contemporary philosopher, for example, there is no word you can word or concept that you cannot take and identify and trace back to some some origin that dates back to the beginnings of, of Tom. So so so you can always do that. And you always realize, like, and the more you like, the more I learned, for example, with philosophy, the more I realized that the very most important things were really thought at the beginning. And then what we did over time was just to try to figure out a little more what other had been said. And so do artists who always quote each other. It's just that when you're not certain when you're when your boss is not necessarily trained, you don't you don't really realize. And so and so and so in this regard, I think that machines can can become are becoming and can be complicated because because we don't invent from zero. We just repurpose and improve what other had been done before us.


Justin Grammens  34:53  

Interesting. I see what you mean. Yeah, so humans aren't being creative. Anyways, so machines You can do that. Just yeah. Oh,


Antonio Di Fenza  35:02  

yeah. Oh word or creativity is not what we think it is. Like, there is not this element of of nothing because you know, the phone. Okay, you know, there were letters. Okay. But there were words so that there is always something that was before and that was rearranged and pre made. And even when even if we are thinking about our conversation right now, and you're inventing the words that you're using? No, they've been said, I don't know how many trillions of times before, okay, what we're doing right now, why is this this conversation original, is because the patterns around which our words are structured right now are novel, because they're taking place right now. But we're using trends that are incredibly old. And so and so this is the creativity, it's novel arrangement and repurposing of what's old. And already there.


Justin Grammens  35:56  

I just I think about when someone creates, well, you mentioned the telephone, for example, like, you know, somebody, somebody had, there was an invention that happened, or somebody is an entrepreneur, they created a business that has never been done before, you know, they see an opportunity, or they take on a new task and, and build a new organization, you know, it's part of that's like, you're right, it's it, they are just taking concepts from everything that they have learned along the way. What makes it hard for you to think is, is what a machine be able to do that same thing, what a machine be able to see a particular need in the market, and then say, I'm gonna go ahead and start a business in this space. I don't know, you know, again, like I say, this is kinda what I'm saying. It's like, I'm trying to think of, you know, how are humans elevating themselves to, like, kind of see, and play things out into the future, that maybe a machine wouldn't do?


Antonio Di Fenza  36:43  

Yeah, you re right. Or also, you see all these these images, you know, meat journey or something gland and looks like a, you know, machines and become creative, and so on and so forth. I'm like, Okay. Now, the point is, are we gonna ever have something like, the 16, sharpell, that has the same kind of importance that will be generated by a machine? I don't know, maybe, maybe, but we will know these only in, you know, few years. It's, it's just a matter of, but but, you know, the business side? Yeah, it's cool. Because you think business, I think, you know, literature culture, but but actually, but the processes are the same. Because you're, you're still trying to find something that no one else has done before. And that and that can generate it can solve problems, or generate meanings in ways that have not otherwise have been done before, to reach to that kind of creativity, that you're talking about uni, a GI unit, unit unit, the next level of stuff, which is still speculative, theoretical. Who knows? You know?


Justin Grammens  37:53  

Yeah, I feel like we've talked it back in a big circle. That's what we sort of talked about to begin with was singularity answer where things are going in the future around super intelligence. So So for our listeners, I guess, you know, sound like you're from Italy. Is that right? Is that? Is that where you're at right now. But yet you've been in the US it sounds like you've been in Berlin, you've sort of seen other parts? I don't know if you can comment on this. But I'm kind of curious what you're seeing from your appearance from a European standpoint, in versus the US. And when it comes to these new technologies? Do you see anything? Because I mean, I, I'm born and raised in the United States, and I've lived here my entire life. So I just I don't really see much from the outside. But curious to know, if you see anything different being in Italy, or United States, probably laws or rules and regulations, universities, I guess, stuff that you're writing?


Antonio Di Fenza  38:41  

I didn't then we recently had elections. Okay. And I don't think that the world artificial intelligence technology was ever mentioned during campaigns, for example, this and this is to tell you like, how worrisome is this, for instance, like in Italy, so I cannot talk about Europe, because they are quite advanced nations in Europe, and then other nations that are, you know, not so advanced in terms of being culturally forward and anticipating what the future is about to, to produce. And so, so Italy is not forward thinking, for example, political discourse, doesn't really deal with the technology alone phrase, this is this is this is very kind of disappointing, actually. But but this is, this is what it is here. Yeah, you hear our prime minister who I don't think has any clue about what's happening right now with daily breakthroughs, which are which are confusing even for me that I basically breathe and sleep into these things. Daily, and I and I, and I'm confused. But but our our prime minister, our politicians are but not really not cool, which is worrisome? Because like, how do you lead the country without, without knowing that the major forces that are shaping the present? And the future?


Justin Grammens  40:14  

Yeah, for sure. Because it's going to affect every every human across the world. Yeah. Are you seeing? And maybe you're not in the college and universities much, but you know, I mean, are you seeing computer science and AI? And are those being taught? And a lot of schools there? I think I wouldn't


Antonio Di Fenza  40:31  

know. I wouldn't know. I wouldn't know. I don't know. I mean, I have a younger, nice couple of them, actually. And I tried sometimes to talk about the kinds of things that I'm doing, but they look at me like I have horns on. So so so the thing is that with these kinds of technologies, either, either people are very opinionated, because they've seen so many movies about them. And so they think that they know what they're talking about, on their own, or there is nothing. But to have actually, a meaningful conversation over these things is quite hard. Because it's really hard to unlearn what you think you have learned. And a lot of the a lot of ideas that we have towards this, this kind of technologies comes from opinions that we have absorbed with with media, basically, which is not necessarily all bad. But a lot of it comes from there. And therefore since you, you eat so much of it, you kind of thing that you know, you're talking about. And then and then actually don't, it took me some time to go through this process of unlearning. And seeing okay, I actually don't know this thing. More, this might be a little strange for you, because you actually studied these things. So for you to give him but it's not the you know why, when I talk about these kind of things, when I when I say that they will be probably I told the woman, a younger woman, she was studying medicine. And I and I told her that at some point, we weren't gonna we were all going to use AI assisted tools for diagnosis for us, we operate surgeries and all these countries, because, yeah, she must have been 22 alternity, she had no clue that this was happening. And I made a joke. And I said, Well, you know, actually, if you if you're not careful, and you've been you don't become skilled enough, you might actually be displaced in 2030 years, because this is and show me your middle fingers. Because that shows like, oh, well, you're saying, Oh, my God, it's true. Through Yeah, better look at what's happening, because you decide how you want to keep learning so that you don't become obsolete, or you just stay relevant. You know, however you want to phrase it.


Justin Grammens  42:53  

Yeah. Or like you say, you use the tool to make yourself even better. So it's not a competition for now. Yeah, well, for sure. It's always, always changing, always change. And that's what's so exciting, I think, for me in this space. So I'll Antonio how, I guess two questions, I guess is how do people get a hold of you


Antonio Di Fenza  43:13  

LinkedIn? So on best way? Yeah. So LinkedIn, this is what I use normally. Perfect.


Justin Grammens  43:19  

Okay. And I'll be sure, like I mentioned earlier, we'll have you know, liner notes in there with contact information to get ahold of you. And then, you know, is there anything else that we we maybe didn't touch on? You know, were there other other concepts or things that you wanted to maybe share that we didn't really talk about? I don't know, I mean,


Antonio Di Fenza  43:35  

hearing how you how do you see, you know, the future unfolding. And, and us and American, you know, you also see as a technical person, so you also, so you see the power, but you also see the danger, because of, you know, the highly polarized society in which you live, for example, so I was wondering, like, come to you, do you have any, any they called you? How's it for you?


Justin Grammens  44:01  

Yeah, I mean, you mentioned about sort of the polarization. I mean, I think that's always been going on here in the United States, you know, Republican and Democrats and, you know, conservative liberals sort of back and forth. I think the fear that I have right now is, you mentioned on social media a little bit, everyone has their own feed, right. And so, you know, my parents are seeing something totally different on the internet than what I'm seeing. And what I'm seeing is, is seems to be just reinforcing my habits, reinforcing my beliefs. And that's what's happening is people are getting stuck in their own echo chamber. And they're seeing their own articles that are spun by, quite honestly just lies, and a lot of cases I believe it, but they're looking at what I'm reading, and they're thinking that my stuff is lies. And you start having sort of two sides that can't come to the table because everyone was trained with a different model, right? To bring it back to the machine learning thing. Everyone was trained with a different model, and so their core beliefs are are just completely too far apart because they've been fed through the matrix of the internet stories that just continue to reinforce what they believe and what they read. And I, you know, I, I just kind of see it on my Google News Feed, like, I read stuff, I click on things, and all of a sudden, the next day, I'm getting more of that stuff. And I'm getting more of that stuff. And I'm getting more of that stuff. And I'm sitting here thinking, you know, and then I talked to my parents, I'm like, have you seen this article about this, this that? And I'm like, No, I haven't. And to be honest, like, I don't even really care about that, per se. And I don't think that I think that's fake news. And a lot of ways, that's my fear around this stuff. And then you throw in all of the deep fake things, where people can look like whoever they want online, we can start doing videos now with, you know, essentially generative artwork, you're gonna be able to see somebody and the fear of mine, and now they're gonna be able to talk like them and right like them, the fear is, is that you're going to basically, you're not going to be able to believe what's really true anymore. You know, was that generated by what I read? Was that true? Was that not true? Man. So, you know, true, truly finding like reality, in some ways. And I'm pushing it to an extreme here is I is what I sort of sort of fear now, I'm not the kind of person who throws my hands up and be like, we're all doomed. And this is going to basically, and I think you said earlier on, you're kind of a generally an optimistic person, you know, it's like, I that's the view that I have is, is this is another tool. And, you know, hundreds of years ago, once a new technology advancement happened, a lot of people threw their hands up and said, Oh, you know, we're, we're not going to be able to, like, there's going to be no farming anymore, right? We're basically going to be, you know, using all these machines, and everyone's going to basically going to move to the city. Well, no, there's still a lot of work to be done on the farm to grow food, right? New practices need to be put into place. So you know, I don't view this as, as an end all be all of like, Okay, we're done as a society. But those are the things that are sort of top of my mind right now. And and we just went through a big, it was our midterm elections in back in November. So it wasn't the big election for the president United States. But we went through a lot of midterms. And, yeah, it's just, it's, it is ridiculous, I guess. And, you know, with regards to, with regards to what people are saying, and what's being written, and what's true and what's not. And a lot of ways, I would


Antonio Di Fenza  47:14  

say that, because you were comparing, you know, being fed and and you're doing this analogy with machines, is the thing is that when you know, when the machine makes a mistake, then there is a you know, the objective function that tries to correct and actually updates, all the weights and activities. So maybe we have to, and so and so machines are capable of doing these things that we are a little bit slower in this regard. Because of course, it's harder for us to change our mind, once we are set on something I even see it with myself, like it's come changing a little bit. With with age, I'm learning to see other things a little bit, but still hard when you're set and you think you had it, you know it to update to you know, back propagate to change is hard, man, it's really difficult, because it takes a lot of work. It's easy to believe that that your part is the whole and then everyone else is wrong.


Justin Grammens  48:16  

Well, I mean, you know, if when you go to train, at least for the supervised training model, you sort of have things labeled, so you throw it a picture of a cat, and you know that you've labeled it as a cat, you know, the truth on the outside of it. That's what I think is so hard is if you don't know what the truth is, or at least everyone has their own truth answer. It's labeled differently than it's really impossible to retrain the model.


Antonio Di Fenza  48:41  

Yeah, there is an article that AI writer Alberto Romero wrote recently, that's very interesting about these things. He only talks about, it talks about, well, this is another can of worms that I don't want to open but but anyway, the because he talks about empathy, and machines, because people are getting confused over the fact that machines are showing signs of awareness, consciousness, and we are building empathetic relationships with them. But anyway, so he says that the cut everything short, is that it's that it's very important for him to, to have a shared representation of reality, if you want to live in a functioning world and a functioning society to agree at least a minimum number of ideas and concepts because big so that we can all sit around the same table. Since this, this number of basic number of things is shrinking over lack over time. It's becoming increasingly more difficult to keep things together because you need to share representation. We don't even agree whether the world is round or flat, right? I mean, you guys have this problem over there a little bit more than we do here. But even on really basic things. We are getting more distant.


Justin Grammens  50:02  

That's interesting. Yeah, that I mean that that transcends technology. I mean, that's just I think about countries and world politics talking to each other. Like, you're right. You have to come to the table and have some sort of base understanding to work together.


Antonio Di Fenza  50:14  

Exactly. Like if you don't even agree, you're on the ship. I don't know how many, you know, flat Earth nurse are on in the US, but if the number is low, okay, but what happens if the number goes out? And you know, and then we are already witnessing these kind of things when it comes to, you know, vaccines, no vaccines, things that we thought were granted. And we we don't we don't agree with these things anymore. So it's submitter. Child, it's interesting thing. And and the thing is that we often face the discussion with the Spectular approach. Down, Right, you're wrong. And even the other part that's exactly isn't what you believe in is false. But what you believe in is false. And so if we cannot find a way to bridge this, this attitude, and to, I think Giva said identical strategy, even to give up a little bit, this attitude of I know better, it's would be very hard for us to keep things together, like for the long run. So this is yeah, I need to we need to discuss and think and be open, especially as


Justin Grammens  51:28  

I ask great, man. Yeah. So we'll end this conversation on applied AI on more of a philosophical note, I guess. Right. Yeah. So definitely away from the technology, but very much important because as you said, you know, even if you I guess, if you are working with an AI, any sort of, you know, intelligent being, whether it be a human or a computer, you need to come to this agreement, matter what happens either today or in the future makes a ton of sense. Well, Antonio, I appreciate the time today. Thank you so much, and I look forward to having you back on the program in the future. Thank you.


AI Announcer  52:03  

You've listened to another episode of the conversations on applied AI podcast. We hope you are eager to learn more about applying artificial intelligence and deep learning within your organization. You can visit us at applied ai.mn To keep up to date on our events and connect with our amazing community. Please don't hesitate to reach out to Justin at applied ai.mn If you are interested in participating in a future episode. Thank you for listening


People on this episode

Head Shot

Justin Grammens

Host