‍

In this episode of The Hive Podcast, host Nathalie Nahai interviews Matt Celuszak, founder of Element Human, a company using AI to measure and understand human behavior.

Matt, having spent the last decade working alongside top experts in technology and psychology to build machine learning that recognizes body language using webcams, works at the intersection of people and technology, pioneering a new class of Human Data that enables tech, AI and its users to be more emotionally aware and empathetic.

With clients including the BBC, Amazon, Whalar and Omnicom, Element Human helps organizations measure the conscious (behavioral) and subconscious (emotions) drivers of human behavior within their ecosystems, using everyday devices paired with advanced intelligence to better understand key stakeholders, improve their experience, and drive more socially sustainable business.

The conversation explores both the potential benefits and challenges of AI, emphasizing the need for responsible development and use of technology while maintaining human values and sustainability.

Key topics discussed include:

• The current global psyche: A mix of fear and excitement about AI and technology, with people questioning their identities and roles.
• AI and creativity: How AI tools are democratizing creativity and knowledge, allowing anyone to create content easily.
• Ethical concerns: The need for regulation, copyright issues, and the challenge of distinguishing between AI-generated and human-created content.
• Energy ethics: The significant energy costs associated with AI and the need for sustainable approaches to technological innovation.
• Human-AI collaboration: The importance of maintaining human creativity, ethics, and decision-making in conjunction with AI capabilities.
• Emotional awareness: Matt's work on building an "emotional layer for the internet" to bring people back into the feedback loop of technology.
• Personal wellbeing: The importance of connecting with nature and surrounding oneself with supportive people to maintain balance in a tech-driven world.

‍

Full Transcript: 

[00:00:00] Nathalie Nahai: Hello and welcome to The Hive Podcast, the series that inquires into our relationship with one another, with technology, and with the living world.

Join me, Natalie Nahai, and some wonderful guests as we explore how we might reimagine humanity in the face of accelerating technological advancement, ecological disruption, and systemic change.

For more information on today's episode please visit NatalieNahai. com forward slash The Hive Podcast. And for additional books and resources, check out nathalienaheye. com forward slash resources. Thanks for listening. And I hope you enjoy the show.

Today, I speak with Matt Selusak, founder of Element Human, a customer understanding platform that uses AI to measure and make sense of the drivers of human behavior. Having spent the last decade working alongside top experts in technology and psychology to build machine learning that recognizes body language using webcams, Matt works at the intersection of people and technology, pioneering a new class of human data that enables tech, AI and its users to be more emotionally aware and empathetic.

With clients including the BBC, Amazon, Waylar and Omnicom, Amazon. Elements Human helps organizations measure the conscious and subconscious drivers of human behavior within their ecosystems, using everyday devices paired with advanced intelligence to better understand key stakeholders, improve their experience, and drive more socially sustainable businesses.

While I am now Behavioral Science Advisor to his company, I first met Matt many years ago when he was starting out, and alongside his desire to transform the media His interest in applying technology to help create meaningful change across critical issues in education, healthcare, mental health, and children's wellbeing is something I really appreciate, especially in an industry that is not without its challenges.

From exploring human centered design and our understanding of emotions, to unpacking the shortfalls and benefits of machine learning in human data, and the impact of regulation and privacy when building artificial intelligence systems, Matt's areas of expertise touch on many of the themes I'm looking to explore in this season.

And I'm excited to share this conversation with you.

Matt, what a pleasure to be in conversation with you. How are you doing there in, I hope, sunny London?

[00:02:40] Matt Celuszak: Uh, marginally sunny saves that they've thrown a winter back at us again for the moment, but, um, great. And thank you very much. We're very pleased to be here. Really appreciate the time.

[00:02:50] Nathalie Nahai: Well, it's nice to get you, um, on the record, so to speak, some of your insights and brilliance, uh, and not just in our conversation.

So I'm going to kick off by asking you the question that I always like to open with, which is what do you think is going on in the global human psyche right now?

[00:03:05] Matt Celuszak: I think there's probably, um, a lot of in equal measure, probably fear and excitement. Um, I, I would say that with the advent of technology moving at a pace, even though it's being opened up to the public and to anyone who can use it, it still favors those who feel comfortable with diving into the new, um, without.

Making sure that we bring along everybody with us. And so I think there's a lot of people worried about their jobs. Uh, however, I also think that there's a lot of people who've already had technology kind of take over some of their jobs, uh, they're seeing now an equalizing force come in where it will, um, it will put everybody back on the same foot.

So I personally think it's very exciting, but I think that on the global scale, You've just got a lot of lack of, [00:04:00] uh, uh, identity for people right now. And a lot of people going through that questioning about identity. And we're seeing it within our organization. We're seeing it within our industry.

[00:04:09] Nathalie Nahai: So in some ways, kind of like a technologically heralded dark night of the soul, like what does it mean for us?

A

[00:04:15] Matt Celuszak: little bit. Maybe it's because I've gone down the path of watching Vikings for far too long, but I It is a little bit like the you know it's like what happened to England about establishing these regions and where do I play and what is my role and Actually, I can come from rags to riches. I can become somebody in this new society And the tools are really opening up and it's anyone's game now.

Um, and, and those who have the openness to say the rule book of yesterday can get thrown out, could really take advantage of that. Um, and those that are, are scared about it, I would just encourage everyone to just go look, try it, see it, see where it goes. And, and, and it's amazing the journey it can take you on.

[00:05:03] Nathalie Nahai: It's a funny thing, there's several things that are coming up in my mind, but one of the things that's most clear and in London a couple of months ago I was visiting one of my favourite families, my cousin and his wife and their kids, and the second eldest, I don't know exactly how old he is now, maybe 10 or 11, he's very sharp, very witty, whips out, chat GPT, And then gets him, gets him, gets it.

That's interesting, I gendered it. Gets it to start creating a script in comedy about a specific theme in the style of his favourite comedian. And then he just, the prompting, I was so surprised by the quality and insight in the way that he prompted this AI. that then he kind of recounted or gave voice to the script and it was brilliant.

I mean it was side splittingly funny and this is a 10 11 year old boy who just has this natural unbridled curiosity and wit and ease and facility with this particular AI platform. And it just made me realize, God, you know, I'm kind of walking in, into this pool, barely daring to put in a toe, even though this is kind of my stomping ground, or at least it should be.

And there's someone who's, you know, a quarter of my age, who's just embracing it wholeheartedly. So I'd like to ask when you're talking about Ways in which we can, and this is kind of one of the themes that we're exploring in this season, it's how to reimagine humanity in the face of, in part, this accelerating technological advancement.

What are some of the ways in which people can embrace or engage with or start to understand AI in their lives, if they're not in your sort of world?

[00:06:35] Matt Celuszak: Yeah, I think it's, it's a really good question. I'm not sure anybody actually has a very good answer for that yet. Um, however, I asked a very similar question cause like you, I was sitting there going, I should know all about this stuff and I feel like I'm already behind the curve because of my cohort and they're all technologically advanced and they all seem to get it.

Um, but what's very interesting about it is it truly is anyone's game. And. When you change the shape and size of the problem, before you had to be a developer and a coder to be able to kind of understand and build things in here, you can, as you said, prompt, I just, I want, I can command and, and humans are very good at commanding and so.

So you could just command a machine to do it. And, and I was sitting there talking to a CMO of a, of a very big brand. And I was just, and they're very into web three and they love the future of technology. And, and I just said to them, I said, like, I said, I really need to get my head wrapped around mid journey.

And they're like, you don't, I was like, what do you mean? I don't. And she's like, well, all you have to do is take all the onboarding mid journey documents that tell you how to use it. Take it over to chat. Say, learn mid journey for me. Upload the documents, they learn it for you, then say, great, I want to create this in mid journey.

What are the prompts that I need to create? And, and it just immediately, I, I don't have to tool up. I don't have to be an expert in any of this stuff. And [00:08:00] I think that to me was the eye opening moment to going, of course an 11 year old could do this. Uh, uh, if you could command, you could do this and say, I don't know X, so machine, please go learn it for me.

And then do this. So where I, I think this is really exciting is a turning around then to, um, uh, any area where you have done any sort of creation of your own. And I can't wait to see it. Like if you start to absorb all of your own content and data and how you write on all that type of stuff, say, learn my style great now in my style, send a shareholder update.

I mean, how awesome is this?

[00:08:39] Nathalie Nahai: So, okay. I can imagine, and I'm going to play devil's advocate here, that it would be super easy for people who are. I'm scrupulous and have very little integrity or, and also probably very little time to be like, sod it. I'm just going to do it. Pass it off as my own. And then the question of when it becomes impossible to tell apart AI generated, well prompted content in the style of from someone who's actually done it and put in the hours.

But then maybe also that, you reframe, which is I'm making that comment on the assumption that hours put into something. confer at value. And yet at the same time, when you do research, like writing a book or what have you, part of the whole process is digesting the information, filtering it through your mind, so you then have a better sense of understanding of the issues at hand.

So it's that question, like how can we maybe leverage the AI and its ability to help us understand or utilize things rapidly that are complex or maybe at a scale that we can't comprehend, while not diminishing our to engage in the world and Grapple with complex thoughts and synthesize information and communicate.

And do you know what I'm kind of getting at? Like we do have digestive process still intact.

[00:09:51] Matt Celuszak: So, so I think it's, um, just trying to think of a really good example where it's one of those, it's one of those scenarios where, and I think you touched on this with saying reframing from an hours based model, hours were just, I mean, time is the only commodity we have in life, right?

Like. That is it all and it's, and it's the only unknown commodity that we have in life. But if you value your time, then that has an economic value, but it also has a personal value, uh, uh, uh, an emotional value, psychological value. So, and an identity value. So there's all kinds of value that comes in from time itself.

Uh, but time has a driver to differentiate yourself in the world of expertise. That potentially is no longer needed anymore in a, in a few different ways. Um, however, I think this is where it will separate from those who truly understand their subjects versus those that don't. I'm actually less concerned about whether I could tell if this came from an AI or if this came from a human being, if it sparks me to act or change a behavior in a way that benefits me, I think that that's great.

If it sparks me to act and change behavior in a way that hurts me, I think we have to be taught to be more. critical about the information coming in. So that that's an education problem, but where it leaves those who understand, truly understand the subject matter versus those who at a surface level understand it is the same way that overnight everybody became a, you know, absolute opinion expert on chat GPT and they don't, I mean, I've been working in machine learning for nine years and I don't understand fully how it works.

Like, like, and I'm sure people could turn around and they'll say, Oh, well, I could explain it to you in 10 minutes. I'm sure you could. And you can ask chat TPT to explain it to me in 10 minutes, but I don't truly, if I was, if I was asked from day one, here are the tools, go rebuild it. I couldn't do that. Um, but that doesn't stop me from being able to leverage the capability and then advance where I am good, which is, you know, understanding this world of human data and biometric data.[00:12:00]

And so, and in fact, one of the big things, one of the big challenges we have in human data, biometric data is articulating the value proposition in the industries that we work in. And so this is where chat GPT I found is actually very useful. I will write my. Absolute kind of war and peace on what human data is, and it will say, great, but a shareholder has 15 minutes.

So, uh, so, so then it'll take that and it will push it down. And I might have to go through a couple of iterations, but I have a sparring partner there to get the words into the right way. So my question to you is, is that AI generated or is that Matz Luzak generated? And then the final point is, if it gets the point across to the reader in the right way, doesn't matter.

So I think the real question here is not whether us versus AI, it's artificial intelligence is like a tool. We've been digging for gold with our hands and we finally been given like a laser pointing guiding system with a shovel and we could get that exactly where the gold seam is and get that gold out right away.

And now it's all about the jewelers who can craft it into something new and something better. And so I think I, for me, I think it's less about the Do I need to know if it's authentic and more about, um, I mean, I do, there's the element of the bad actors and we'll get into that in a second. But in, in terms of using this in a benevolent way, I, I I'll use it.

I mean, just like, uh, if I could go faster for point A to point B, then walking, I can move to a horse. I moved to a car. I mean, it's just an evolution thing. I think it's going to help us. I mean, we are going to be able to solve so much scientifically just by being able to let, The machine read the back data and inform a scientist in the field who says, Oh, I can connect the dots here, but even inform non scientists say, could we connect the dots here and raise the right questions that actually need to be raised.

So I think what it does is you've touched on it at the beginning, the hours. What it does is it frees up those hours for us to actually focus on the real, the stuff that's unanswered. Versus having to go and review the stuff that's already been answered. And, and it helps summarize that in a way so that I as somebody could take the, the body of knowledge, extend that body of knowledge, and massively and then have a much more guided question moving forward and say, Oh, I actually like three or four hypotheses that I was going to ask.

I no longer even have to ask that body of work's been done.

[00:14:26] Nathalie Nahai: It's curious. I was listening to, um, an episode of the Great Simplification the other day about fossil fuel use and its impact on, uh, ways in which you've been able to accelerate innovation, uh, in the world. in particular in the world of agriculture and infrastructure.

Anyway, and one of the things that they said was that for every barrel of oil, it equated to, I don't know how many months worth of human physical labor. And it seems to me that this is almost a parallel in terms of cognitive labor, the sense of, you know, there's, there's something very curious about those two revolutions, if you like the industrial revolution and now the technological revolution.

So one of the things I know you and I have talked about in previous conversations, I'd love to dive into in a bit more detail is how AI and all of this potential connects with our ability to be more creative and social. Like, what are the applications that most excite you? you know, roam into the realm of business as well.

But what are the key things that you think AI will augment in a positive way in our societies and culture moving forward?

[00:15:31] Matt Celuszak: Yeah. So I think, I think first we have to, um, Just for a definition check, artificial intelligence has been around for like over 40 years, so, uh, it, this isn't the first time we've had assistance from computational power, and, and, uh, what's happening is that we're now able to do a lot more with a lot less energy, but because we can do a lot more, we're expanding that amount and that energy.

Equation is still there and it's still heavily on my mind, particularly when you give everybody shovels, well, then the [00:16:00] mountain disappears. And so you have to be careful about the damage that you're doing. And that is the kind of part of the unknown questions when it comes to creativity, I think what I love about artificial intelligence and it comes back to, um, I'm part of this founders group and, and we do these away sessions and, and it's about kind of get out of your head type of stuff.

And one of the founders did an activity with, I said, You know, who here could draw and like, I think two people put up their hand and he said, I bet you in an hour, every one of you can draw. And, and it was, it was an amazing thing because it was just totally untapped skill that I thought that I never had is that I couldn't draw.

And it turns out everybody can draw. When you, when you dedicate your time to certain areas, you tend to play to your strengths. And the question is that then you feel that you cannot create in the area. That you might enjoy quite a bit, but you can't create there. And for me, again, I love business. I love business strategy.

So what's really cool about, and what we're talking about is this wave of generative AI and, and kind of exceptionally broad learning at exceptional speed is, is that we can start to apply this in ways where. It's creativity beyond visual audio, the area that was reserved for the artists. But we can actually look at business artists.

We can look at strategists actually saying, how do I get strategy into a tactical implementation? I'm looking at business leaders today. I mean, One of the greatest skills about a management team that you need to look for today is someone who can actually work and read a spreadsheet. I think that skill will be gone in about two years.

I think that's going to get cracked and we'll be able to empower empathetic leaders, people who've always had the gut intuition around. If I make these people moves, we're going to get a kind of force multiplier effect on our business, except they couldn't make the economic argument, so it never got approved.

And, and you're just like, well, You don't need to know the economics anymore because economics are just model theories and, and, and we could, that could get learned. We don't need to do that. And so I think there's a, when it comes to creativity, I think a couple of things, um, anyone now could create something.

And because. People are typically going to create and in either domain they're very curious in or a domain that they feel strong in our prompts are going to get really strong. So I think what, what, what's going to happen there is you're going to have the experts creating in, in their area that, that are quite strong.

Um, so we're going to see stronger and stronger prompts. So for those who want, who aren't an expert in that area can actually look over the experts who are probably hopefully sharing and some will and say, look at what I could do. I can now build a business plan. Two minutes I did. I literally did that yesterday.

[00:18:40] Nathalie Nahai: And how was it?

[00:18:41] Matt Celuszak: It was amazing. It literally wrote my business plan. And I said, you know, in the market size, and can we get more numbers in there and reference points? And it did. And I was like, and it's, it lines up with half the numbers I already have. And I was like, Whoa, like this actually, and it's a good, strong narrative.

I mean, we could. It's just absolutely nuts what you could do with this stuff. Um, but, but I think the other aspect here is, is that there is no good feedback loop for the creator. So currently that to me is still the sticking point and the friction point is you sit there and go. I've just created something and I, I noticed it while I was creating the business plan.

I was sitting going, yeah, yeah, yeah. Like nodding my head as it started to write. I'm like, it's like watching somebody work. It's like when you, when you have a good fun and I'm such a nerd here, but when you have a good finance person and they're sitting there going, Oh, but what if we move the cash this way?

Would that change the cashflow? Or would What if we did this to the budget? How would that change the forecast? And they're, and they're like doing it real time. Like, yeah, yeah, that looks good. Yeah. Yeah. It's like magic. So watching people work in their strength. I mean, here, you've got an awesome SPARI part to say, so do you mean this?

And this is like your awesome copy partner when it comes to the written form and the language models, uh, mid journey. I'm, I'm starting to explore that. I can't wait to see that. Cause that's an awesome visualization partner.

[00:19:59] Nathalie Nahai: [00:20:00] What is mid journey?

[00:20:01] Matt Celuszak: So I would be the last person asked to describe it to you. I am literally just getting into it, but what it, what it does do is it starts to look at, um, generative imagery.

Uh, and so if you, again, what, what they've done to language models, they've taken into images and, and being able to describe those and connect those to language. So when somebody describes an image, they get a language and, and, but you can create like avatars, fake people, all that type. I mean, it's, it's amazing.

[00:20:29] Nathalie Nahai: It's just bonkers. I saw one of my good friend's partner has been fiddling around with, um, it was an avatar generating AI platform, which created portraits of him in different styles. And so this raises so many interesting questions. I know in the States, there's some interesting kind of lobbying happening around, you know, Copyright and what happens to artists when their work gets, quote unquote, mined without consent.

I mean, there's all sorts of interesting battles that are heading our way. When it comes to things like creativity, I can understand how a business plan or computing in like a sort of a research based or financial way Would be something to which this would apply and be absolutely brilliant and, and probably not have so many problems, but if it's someone's music or their art, things that we also connect to, or like a screenplay, let's say, I mean, I can imagine Netflix, they've probably already done this, I can imagine like a platform like Netflix or Amazon going, right, these are the kinds of shows that get the most traction.

I want you to do a new show and create something in the style of X, what do you think are going to be the battlegrounds? In terms of creating guardrails or limits that enable AI to help us in the domains where we need it, but don't end up undermining human culture and the things that maybe we think of as sacred or

[00:21:49] Matt Celuszak: well, this goes back a little bit to the point of what is human culture.

I mean, there's an entire car culture that didn't exist when there weren't cars. And so again, it's just a technological advancement. There's going to be an entire culture. And, and I think the, the obvious one, and actually there's finally an article that came out about it recently about the American smile and yeah.

And, and, and I think that's, that's bang on it. Cause it was so people are saying, oh yeah, here's a Biden and Trump hanging out. I'm like, why are their teeth so perfect all the time? And that's, and it's, it's those small little things that as a human, you kind of clock go, that's just not. normal, they're abnormally white.

Like there's somehow the machine has thought the teeth need to be perfectly white all the time and perfectly straight. And the jaw needs to be boxed. It's like, so there's, there's all these small little things that I. That I think where, where I think there's a couple of different areas where battlegrounds come in and one you are going to have a massive battleground a decision to do about original content versus generated content or the amount of generated content influence in the creation process.

I personally think that's actually the wrong place to focus. I do think that it's actually about creating tooling to help, uh, identify or either, and if you want to focus there, then verify it, I mean, just have a verification system, but what we haven't solved even for social media pre this era, what we haven't solved is moderation.

And moderation hasn't been done very well at all. And now you're exploding the content stream. So we're continuing. So first we solved the amplification problem. If we call it a problem, first, we solved the amplification element technologically, but that totally got rid of really the human impact. It was, you know, fail fast.

Fail often to build your capitalistic business, but at what cost? Well, somebody has to deal with that failure. And then at that point, so consumer. So then the, the, then amplification was done. Now creation has already started with all these creator tools, creator platforms, the explosion of the platforms will last two to three years, but here's this new Tik TOK [00:24:00] format, here's this new carousels format and all these new formats for creation.

So that's great. That empowered a bunch of people and our creators went from like, A movie industry of, you know, 300 really kind of key stakeholders to like 3000 X, what that was. Um, and anybody and their brother and sister with like 200 pounds could pretty much become an influencer if they, they, they, they do the right thing.

But there's still an element of, you still need the buy in of your audience. And, and again, even the amplification that just gets the word out there. It still doesn't keep the, the, the human in the feedback loop. And then I think what the generative stuff is, again, it's about putting the human in the feedback loop.

And so we continue to build these technologies that are very much individual out, rather than a recursive relationship of saying, okay, it goes out and it comes back and oh, that was actually really shit and I've caused damage here. And therefore I'm actually regulated based off of that. That capability.

So I think there's that one. I think the, the second one is related to that is you touched on it a little bit is regulation, uh, which is largely driven by kind of the ethics and moral set out by your society, which largely sits within your current governance systems. I think we are on the precipice of a big geopolitical refactor of the world as it is.

Um, it was already in play and I am not sure. I get why OpenAI did what they did, but

[00:25:34] Nathalie Nahai: What do you mean when you say this?

[00:25:35] Matt Celuszak: I get that, uh, I get that OpenAI, you know, I mean, they've totally redefined and created a whole new category that big tech hadn't really covered, and frankly Google hadn't really got there, of being able to enable anybody to actually really have a real sensible conversation with the internet, almost as a whole community.

And, and open up that capability to the world and put it in the hands of everyone. But what But I think it was a little bit naive, uh, put out there on the, with the governance systems and saying, Oh, well, yeah, but U. S. thinks that we could use this for good, but you've got, there's a lot of unsolved bad actor elements out there.

And frankly, regulation just has not, we need to move to regulation by design, not regulation after the fact world. And my view on this is particularly with a little bit of an East versus West. geopolitical environment starting to kick up. We have to be really careful here. Um, and because these tools are great, but then, you know, then when big tech and all the Western companies go, we're going to put a hiatus on this for six months so that we can get regulation order.

It's the right thing to do is the regulations, the right thing, but putting a hiatus. That just gives anybody who doesn't agree with your value system six months to catch up.

[00:26:53] Nathalie Nahai: It's that whole race to the bottom again.

[00:26:55] Matt Celuszak: It's that race to the bottom. And, and we have to really, really carefully here. So I think your battleground, I think we're going to see geopolitical battlegrounds on this one.

I think we are going to see fundamental ethical, moral identities challenged in this one. And then kind of at the copyright level. I mean, unfortunately, the traditional copyright system, the way it actually shook out was putting all the power into the hands of very few people. I mean, look at record labels.

I mean, that the music industry is prime example. Video industry is not too far behind or video movie and TV industry is not too far behind. So, uh, yeah, I mean, I'm all about democratizing this. I think. It shouldn't be in the hands of few. I think creativity shouldn't be limited. And I think frankly, everybody should be able to create anything they want to go and create.

And the market will, but, but as long as the market or the audiences have a voice. And so I think we need to start putting people back in the feedback loop.

[00:27:55] Nathalie Nahai: You just had a crazy thought with that. So like, and I was thinking about some of the key players [00:28:00] that advance. technological innovation. So if we're thinking about, and this is one of the reasons why I have this show marked explicit, because sometimes you move into this territory, thinking about pornography, for instance, and the darker side.

So I'm talking about now the consensual, and I do believe that can be ethical pornography. I don't imagine it's particularly common, but I have met people who work in that area. Um, but then there's a whole rest of the world of pornography where people get exploited. And I'm, I'm thinking on the positive side that if it becomes possible to create.

avatars that are lifelike, like for instance, now in a non pornographic context, the Levi models or other models that have been generated. Yes, of course, some people are gonna be put out of work, but I wonder if there's also the opportunity to create an entire or replace an entire exploitative ecosystem with sexism.

AI generated content that then removes pressure. And this is, again, this is quite a layered conversation to have. Maybe it merits this entire season, but that there's a potential to alleviate human suffering by replacing industries that at their heart are generally speaking, in most cases, exploitative.

And so while we're thinking about creativity and business cases and innovation and culture and ethics, there's some really key tangible examples to my mind where. Disruption could be absolutely extraordinary on, on various different levels. I kind of, I don't mean to derail you.

[00:29:28] Matt Celuszak: I mean, I really like this cause so a couple of things kind of come to mind here.

First of all, in order to tackle kind of the, the human data problem that we had to tackle, we ended up having to work with people who knew how to stream really high quality data streams. In real time, uh, which led us to the people who built the underlying video streaming ecosystem for MindGeek, who are the owners of Pornhub and MindGeek own about 80 percent of the pornography, online pornography industry, um, also based out of Montreal.

And they are, uh, also backed by Canadian ethical fund. Really? So it's a fascinating, fascinating, uh, very complicated thing. Well, when you become the 80 percent owner of the industry, you also become the number one enforcement regulator. So, and, and similar things are happening with gambling, right? So it's very, uh, I find, I find this, the, the, the, the area of pornography I find interesting because of exactly as you said, there's sex workers and there's people that have been exploited for sex.

And, uh, the demand is huge. Um, obviously, uh, because all one has to do is look at the amount of data transferred on the web and how much of that's pornography and you can just tell that the demand is huge. But I think what it is, it's people exploring kind of an area that were taboo topics, but they had no place to safely explore.

And in order to get that exploration done, there had to be content that was produced, uh, for that, which means you had to have real people doing it. So I certainly take the point. And I think that this is a very interesting area that if you could actually, and, and, and you started to see this democratize a little bit with like things like OnlyFans.

And stuff like that. So, so you start to see, see these areas, but I think what's fascinating about, um, uh, both the industry and how that industry is playing out is first of all, we're finally opening up the conversation. So like. I mean, the, the sex toy industry and, um, hedonism resorts and stuff like that's all kind of even on, on the upswing and it's more openly talked about, but even just the, and what's really, really great to see as a, as an offshoot of this is just the focus on women's health.

Yeah. Just the fundamental focus on women's health and how much it was ignored. I mean, my, my, my mom, uh, worked at, [00:32:00] uh, and, and helped set up the birth control clinic in our hometown back in the eighties and nineties. Um, and, and every day she'd come home, like talking about some of the cases that she has and.

And it was fascinating growing up in that environment because for us, like, sex was a pretty, I mean, we could talk about sex very openly and you could pretty much tell me anything to do with sex and I probably wouldn't be surprised and I've probably heard it before and probably know exactly which resource to go check out if it's got to do with something negative.

That's so useful. Yeah, like it's a, it's a very interesting area. And it's a very human, biological thing. I mean, it's. It is fundamentally what makes us human in, in a very big way and how we identify ourselves. So I think in a world of, of generative artificial intelligence, there is a huge opportunity here to, and again, this is where you have your actual sex professionals who love doing it.

They like the work, they make a lot of money through the work, uh, and, and they only have to do certain things in order to be able to get, and, and open this up to an imaginative sphere that obviously. Might not be intenable, um, or whatever. And, and I think now whether it feeds the wrong behaviors and traits, again, you got to let the audience decide that and, and the market decide that and your governance decide that, and this, and this is where that runs into the regulation side, but depending on where you live.

And how they approach those topics, I think it's up to them. I don't think governments or anyone should be expected to change their approach to pornography. And I think every country should be free to exercise that approach. That's their society. They're, they're in those positions to govern that way. Um, but more importantly that those who feel that they want to work in, in that industry also shouldn't be.

be subject to being exploited beyond the bounds of, uh, of, of the area that they feel comfortable. And I think what the generative stuff allows us to do is you could take a snippet of that and then go into the realm of where it might be uncomfortable for that particular individual. But it actually could still work again.

I think you'd probably want approval. So for me, the other battleground is I didn't sign off on this. I didn't approve this. Somebody's like all these fake images about Trump and Biden going around. And so it's like, you know, but none of them approved it. And, and I think that would be the, for me, that would be the authenticity challenge, and we're going to have to get really smart all the way down to our education system of saying.

Question everything.

[00:34:34] Nathalie Nahai: So I think from that perspective with the authenticity, there is a question around how do you prepare people to discern what's authentic and what's not. And when, when we're saying authentic, I'm guessing here that we mean something that is what it claims to be. So if it's like an AI generated learning aid and it says I'm an AI generated learning aid, look at me.

I'm great. And it's really useful. Great. But not to pretend to be something that you're not. So there's something around, um, being able to. Uh, vouch for something's provenance, really, how do you imagine that's going to take off? Is there a way to legislate for that? Because I think in terms of human being, being dupable,

[00:35:13] Matt Celuszak: we

[00:35:13] Nathalie Nahai: are so easily duped as humans and you know, nevermind with all of the things like confirmation bias and the rest of it.

The kind of content that we consume, look at the pandemic, there was so much misinformation in political campaigns in recent years. It's just so easy, even in my family, for people to send articles and them not be referenced properly and no one's actually bothered to check before sending them on. Like, it's just human nature not to be constantly vigilant for the authenticity of the content that we're consuming or sharing.

What can we do about that? Do you think

[00:35:41] Matt Celuszak: I mean, that's a massive topic, but if I could solve that, I'd be a very wealthy person right now. Yeah, right. Wealthy and benevolent. Um, so I think rather than the kind of going to the solution, really exploit the problem a little bit. One is, I think authenticity is two sided.

So one [00:36:00] is it's somebody either trying to pose as something else or somebody or something trying to pose as something else. Right. Uh, so that's intentional. Uh, and then the other side is whether people buy into it. I think, uh, first of all, society, we have to be taught that everything is an opinion now and nothing is fact anymore.

Um, and, and actually that's not an unhealthy, uh, I think we're very quick to say it's a fact because five people said so in my circle group and it's like, well, it's not. Um, and. And by, by removing the escalation into this is concrete, it's like, no, I just have a strong opinion and everybody's allowed their strong opinion.

That's frankly allowed. And some people hold them loosely and some people hold them strongly. And honestly, it doesn't really matter. I think it, I think it's going to help us evolve to become. And again, I think this is your natural evolution. We've created technology. That's now going to push humanity to rather than being right by being the knowledge expert.

No longer going to be the knowledge expert in the room. Uh, but those that are going to be, I think we'll start to, or so what we're going to start to see is then the shift of saying, great, now that we have this knowledge, what's next. And it's going to shift us from a planning oriented society to a growth oriented society on the benevolent side.

Um, and what I mean by that is that before I was the best person to write the plan because I knew the most. And so I was the person who could probably see around the corners and write a plan that had the highest likelihood of succeeding. Whereas, uh, now it's, well, you can consume everything that I know in about 10 minutes, arguably down the path.

Um, and not that you'd want to, um, but, but you could consume everything that I know in about 10 minutes. Uh, but now that facilitates us to have a, you're caught up. So now we can have the conversation about, okay, what next? What, what else do we want to try with our time, time being the valuable thing? What else do we want to do with our time?

And, and I think that if we can take down the value of, of, of fact versus opinion and just go, everything's now an opinion, like even in science, it's an opinion, even though it may have been proven time and time again, let's just call it an opinion. I mean, the greatest one that we came across was there's over a hundred years of psychology saying that there are universal expressions and emotions.

We're not seeing that in the data. We've never seen that in the data. And so we found a better way to measure that and say, like, actually when, or, and, and it may just be that the way you can measure versus the theory are just never going to align. But in the world of application, where we do want to know when mental wellness is an issue, we want to be able to track our psychological state.

It's really important that you understand that there's a difference there and you're able to have a constructive conversation around that. And so I think that that's the first bit. The second bit, actually, I think the answer is already sitting in front of us in terms of copyright and authenticity is we've already developed a system of being able to use blockchains to keep provenance.

It's already there. It's disconnected at the moment, but it won't take long that once the regulators say, well, if 30 percent of this or more is AI generative, then it's open net, whatever, if it's less than that, then we can call it authentic or, or proprietary capability. So, so I think both of those, but the other thing that I think we're going to see, and this is again, in the IP world, intellectual property.

Okay. Well, it's your property kind of just doesn't exist anymore. Like, and, and the system's so backlogged anyway, that to get to enforceability, I mean, gosh, the amount of time it takes to get a [00:40:00] tech patent through the American courts all the way through to enforceability, like you've lost 10 years. I mean, it just, that made sense in pharma, but even in pharma, you could disrupt pharma tomorrow with the base knowledge that they already had.

Like, so, so I think there's, I think we're on a wave of seeing some fundamental pillars. Get struck down and rebuilt in a different way.

[00:40:23] Nathalie Nahai: I mean, the scope for change on a really fundamental level is huge. And one of the things that, that we've touched on in the past that I'd love to include in the conversation before, before you run out of time, there's so many areas to explore is a point that you've raised on various occasions around energy ethics.

[00:40:37] Matt Celuszak: Yeah.

[00:40:38] Nathalie Nahai: Can you talk us through that? Through what energy ethics is and why it's so important because it does connect to a lot of the issues that I'm concerned about on the podcast and many people who are concerned about the future of humanity and all of life are also concerned about.

[00:40:50] Matt Celuszak: So what is it? So the cost of energy is coming down in a sense, but it's moving to renewables and all of that fun stuff.

Right. Let me start with breaking down the problem. Basically, if you take a generative AI tool and then you give it out to the masses, every time somebody puts a prompt in, it has to go into this massive machine brain.

[00:41:09] Nathalie Nahai: So every time my little cousin's kid takes out GPT, puts in a prompt, isn't this fun? Ha ha.

Great.

[00:41:14] Matt Celuszak: Yeah.

[00:41:15] Nathalie Nahai: This is what you're talking about.

[00:41:15] Matt Celuszak: Yeah. Every time they ping that server and the server has to crunch and throw stuff back, there's energy involved in keeping those servers up and running, making sure the calculations can be done and then throwing that prompt back. And of course, then you just, I mean, nailed the system.

These data centers, while they might be more and more efficient and running, you have to keep your coal fire power plants up and running to be able to fuel that. And so it still becomes a war around energy. Um, and a war around attention. So attention is about getting as much creative out there as possible, which, which both of these are actually, it's, uh, they're both escalating effects, uh, on the environmental sustainability element.

So, I mean, if you just had everybody in the world unplug for one hour a day, Can you imagine how much energy would be saved? So I think the ethical implication of artificial intelligence is that it needs to keep mining and feeding. And basically we've moved, uh, these kilojoules of kind of energy from our brain power onto a machine and then just to just crank the dial.

And the irony here is that artificial intelligence is probably going to help us solve this problem. But until we have really good understanding of renewable resourcing and full supply chain. mechanics of energy, uh, distribution. A lot of this stuff, a lot of stuff that slips through the cracks because.

One aspect of the supply chain just isn't factored into the energy equation. And so until we have a really good macro view that is well beyond the comprehension of a single individual or human being of the energy cost to have these servers run and have them hit this many times and have the creation happen as many times.

I mean, think about screens need to go up. You need to have an image to put up there, all that type of stuff until that happens. Um, we're not really going to solve an energy issue. And in fact, if anything, we're just absolutely bombarding and putting major stress on the energy ecosystem, which means it's now about maintenance and, and putting more and more power in and not actually solving the real problem, which is renewable energy.

And so I think we do have an ethical issue here of saying, I think we should throttle based off of energy. I mean, the other part is like, do you take that energy away from people being able to actually heat their homes and have a comfortable enough environment so that they can. Actually read for school and learn like, or take it away from the hospitals.

Like, I mean, there are some fundamentals that we can't get carried away. And the, and the, the challenge with the innovation community is that. Once they see a string that they can pull on, they will pull, they will yank. And so I think we just, that's where we have to be careful. Um, because the energy [00:44:00] ethics on this one, really, really, really important.

The data's heavier, it's getting bigger. The compute power, I don't know if it's keeping in line. I'd have to look into that, but I don't know if it's really keeping in line. But regardless, you have to turn these servers on, these research cores that are massively expensive. So hopefully quantum computing will help with that a little bit, but probably not yet.

Um, and hopefully artificial intelligence will be able to give us full energy consumption supply chain analysis to tell us, is lithium ion battery the best thing out there? Or, And, and what type of destruction does that do globally as well? So there's impact beyond just carbon, um, and energy consumption.

So I think there's elements there. I also don't think that we should turn our noses up at oil. I think we need to continue looking at oil, just be much more effective about how we do it right now. A lot of the loss of value. In the energy of oil is the fact that where we pull it out of the ground is miles away from where it gets refined, which is miles away from where it gets consumed.

So the transport costs in between is horrible. Like these ideas of not building a pipeline and environmentalist coming and saying, not building a pipeline for oil to get from one place to the next means you have to put it on trucks or railroads. Like, I mean, my brother works at dirt and I think this is a very good example as well.

Another metaphorical example is. You know, just because naturally a river might have a lot of copper in it, but it doesn't meet the topsoil requirements for environmental, for environmental concerns, you have to truck that out of the city, remediate it and truck it back in. Well, what's the fuel cost you just used on this?

So we're doing the same thing in technology now too is great. I have to move this stuff around. There's a cost to all of this. And I think, again, opening it up to the public. And saying everybody can now create, it's a, I agree with that ethos, but we want to do it sustainably.

[00:45:54] Nathalie Nahai: It's so tricky, isn't it?

Because it, it calls to my mind, like the, the fact that often we need, um, how do I want to put this? I've referenced this on the show before. It was this idea of phase transition and the example of the egg and the chick. eating the white of the egg and getting to the point where it's running out of food and at that point its digestive tract is strong enough to be able to eat outside food and it's strong enough physically to be able to break out of its shell but to get to that point it has to consume all the egg white and it doesn't realize what's going on until it reaches that threshold.

Obviously it's a very simple metaphor for something which is perhaps more complicated but I think it's a really interesting question and all of the issues that you're raising endemic within the answer is everything's complex everything's entangled and maybe the question has to also involve the fact that we need to innovate more in order to get the point where we can get technological solutions to help us in a real way, address some of the issues.

And that the, at the same time, although the answers might be facilitated by the tech that we're employing to help us crunch through the data, find correlations, run the assessments, make informed suggestions about how to move forward. We still have to employ the human creativity, ethic, visioning, you know, all of this kind of stewardship mentality to be able to make sense of what we're gaining from AI or what have you.

To kind of chart a path forward that there's got to be some sort of collaboration. But the thing that I always worry about is like, At what point is it too late? Like how close to that cliff edge do we get? Because historically it's really fucking close. You know, the cold war generation, that's my parents.

This is, this is not even old history. No, it's, it's right around the corner.

[00:47:36] Matt Celuszak: And we're looking at the next one. Uh, I think to answer your, your other question, the, the, the re the only thing I didn't really understand with the rollout of all of this is you've just created this wonderful ability to consume all of the knowledge around a particular subject area if you wanted to.

So learn X. Why did you not roll this out to climate change first, knowing that that's your biggest cost driver in order to enable your service? [00:48:00] So, for me, if I was sitting kind of in the seat, the hot seat of big tech right now, I would be sitting there going like, Okay. Our carbon costs are massive in order to drive this stuff.

That's the big sustainability concern and sustainability is an issue right now, globally. Like it doesn't matter how good the AI is. If we burn ourselves to death in five years, it doesn't really matter. So I think the big thing here was knowing that my servers use so much and that it's putting so much power, the first place is.

Give it to the people studying this. They know all the resources to learn, build up that body of knowledge and solve that problem before you open it up so that you're constantly building again, sustainable by design, regulated by design. That's the problem is we're going, here's a really cool piece of innovation, throw it out to them.

We're still using those old techniques of fail, fast, fail often. And it's not. I'd really love to see fail sustainably. So fail in a way that's still a rising tide. Don't fail in a way that's going to injure somebody in India. Who's now got to deal with, you know, uh, 50 odd degree summers. Uh, in Celsius or a hundred odd degrees Thomerson, you know, and, and just people are just dying because of it.

And it's like, like, let's think about how to fail sustainably versus fail fast and fail often.

[00:49:22] Nathalie Nahai: I would love to hear if anyone's listening to this and knows of any organizations, tech companies that are working in this domain, please let me know, I'd love to have them on the show. And so we're moving to the end of our conversation, which is crazy because I have so many more questions and there's so many other things that we could talk about.

But one of the things I would like to ask is. Because you are a deep thinker and the number of conversations you've had, which have roamed all sorts of depths. I'm curious to ask, when you dive into the darker areas of life, where you get hit with overwhelm, how do you orient yourself towards life and wholeness on difficult days?

[00:49:55] Matt Celuszak: Uh, there's two aspects to me that Always re center me and both of them are entirely non technical. Um, the first is, uh, fresh air, I need to get back to the land. So last weekend I went lambing and I went to pulled a little lamb out of a ewe and it was like, and, and then kind of clearing its nose of, uh, of, of the liquid and all of a sudden, like the first breath, there's just nothing more satisfying, like you just, okay.

Life is still here. That's the first one. So get outside. And the second one is really people. Just Identity and knowing who you are and grounding around that is just surround yourself with people who Challenge you but always challenge you with your best interests at heart Um, so, and, and once you could trust that somebody is going to challenge you with your best interests.

And I mean, we, we have this, uh, uh, it's, it's great because they fundamentally help you get back to the path that you were meant to be on. So you're not sitting there two years later with regrets saying, Oh, well, I was just doing this for somebody else. No, you always want to be every hour is for you. So choose your hours wisely.

And, and, and for me, it's get outside and get back to people. So. Like, this conversation provides a lot of value for me. But yeah, I think, I think that's, that's it. Nothing technical at all. I actually hate being in front of a screen.

[00:51:22] Nathalie Nahai: Analog and wonderful. Um, so I realized we've not really talked too much about Element Human, but I do want people to be able to know where to find you and find out more.

Uh, so if people want to learn more about your work, where are the best places to go?

[00:51:35] Matt Celuszak: Uh, sure. I mean, our marketing is atrocious cause we haven't done any. Uh, so the best way is probably to give me a call. Um, uh, or I think Matt. At element human. com. Uh, we do have a website element human. com. Uh, and then, but I think the best place is LinkedIn.

We're starting to ramp that up and we're doing some really cool work. So you will be able to see some news come out over the next couple [00:52:00] of months. You'll also see presentations coming on at the can, uh, and just at a really high level, our, our goal is to build the emotional layer for the internet. So how do we.

Bring people back into the loop so that we can self regulate these systems that are coming online and self regulate our amplification and creativity.

[00:52:18] Nathalie Nahai: Brilliant.

[00:52:18] Matt Celuszak: So how do we make the world emotionally aware again?

[00:52:21] Nathalie Nahai: Emotionally. I think a lot of us could do that, but the thing is getting people who might not realize they need it most.

To do it.

[00:52:26] Matt Celuszak: Yeah. The key with all this stuff is always just embedding it into the ecosystem that they're already operating in. Uh, and so, so empathizing, empowering. Yeah. And then engaging.

[00:52:37] Nathalie Nahai: Well, Matt, it's been a pleasure. I think we're going to have to get you back on to talk more about some of these subjects, but for now, thank you so much.

It was such a treat.

[00:52:44] Matt Celuszak: Thank you, Natalie. Real pleasure to be here. Thank you very much for having me.

[00:52:54] Nathalie Nahai: Thank you for listening to The Hive Podcast with me, Natalie Nahai. If you've enjoyed the show, please do pop over to iTunes, Spotify, or wherever it is that you listen and leave a rating and a review. It really does mean the world to me to read your support, and it keeps me going to create more seasons.

Especially as this is a self funded project into which we pour hours of work creating, recording, and producing each episode. To find out more about my work, you can sign up to my newsletter at natalinahai. com, explore additional books and resources at natalinahai. com forward slash resources, and you can reach me on Instagram and LinkedIn at natalinahai.

My thanks to Kerosene for producing. Thank you for listening, and I look forward to sharing more with you in the next [00:56:00] [01:00:00] [01:04:00] episode.

‍