Loading summary
Transcript50 lines
- [00:02]
Samantha Murphy Kelly
You're listening to the Next Innovation, and I'm your host, Samantha Murphy Kelly. I'm here in Austin, Texas, for the south by Southwest annual tech conference. Enterprise Ireland has set up a meeting place for Irish innovators and leaders, appropriately called Ireland House. I've been speaking with a lot of.
- [00:20]
William Fitzgerald
Them over the last few days.
- [00:21]
Samantha Murphy Kelly
And of course, artificial intelligence is among the most highly discussed topics. Many wonder what its future looks like and what the current hype means for general intelligence. So earlier in the week, I moderated a live panel discussion to further explore just that. What does the evolution of AI look like, and can it coexist with human intelligence? Is regulation key for innovation? I sat down with William Fitzgerald, co founder of Bard Global, a technical writing and content company. Dr. Susan Levy, an ethicist and professor at the University College Dublin, and Dr. Matt Lease, a computer scientist and professor at the University of Texas. Let's take a listen.
- [01:09]
William Fitzgerald
Hi, everyone. Thank you so much for coming. We're so excited to be here at south by Southwest. Today we will be talking about AI focusing sort of on the intelligence and the human side and sort of how the tech landscape is changing. And I'm joined here by my three panelists. So excited to dig in and talk to everyone. I think let's start just by just talking about the intelligence and AI, putting the I in AI, obviously, it's right in the name. What does that mean? Why is it important? Maybe walk us through sort of what might seem obvious but might not be.
- [01:42]
Dr. Susan Levy
Yeah. No. So it's something that's so large that when we're talking about AI, we almost kind of get lost, and it's seen as a piece of technology where we figured it out and we know what actually artificial intelligence is. And, you know, the systems themselves are massively important. They're very, very smart. Are they intelligent? That is kind of up for debate. So really what we want to do is to start that conversation and look like, what is intelligence for us? What do we feel like intelligence is? What is the connection between a human, a connection between a computer system? You know, many people, you know, in my world have been using technology for years, loves computers, and they have had a connection with technology for so long. But what does that mean? What does it mean that it's different from having a connection between your partner, your child, your father versus a piece of technology? So intelligence is, to answer the question is, I don't think we've really defined it and we don't know what intelligence really is right now.
- [02:41]
William Fitzgerald
What would you say then? Is the difference between, again, human intelligence and artificial intelligence, again, again, seemingly obvious, but when you break it down, especially when these systems are being trained by humans, where. Where does that all fit in?
- [02:54]
Dr. Susan Levy
I personally think the difference is actually bias, that humans have a lot of bias. So in. If you look at a. Any kind of LLM, any large language model, it can pretty much tell you anything, any view from over here, any view from over there. But in reality, when you start using it, it's your bias that actually is introduced into that system. It's something that. What is your story? What are you thinking? What are you trying to do with that system that really, actually creates that peace? That's a big part of intelligence, at least. Of course, there's memory, your whole nervous system. How you feel things is different, and how you connect and really experience things is a big part of intelligence.
- [03:35]
Dr. Matt Lease
I think we all want to realize the benefits, the opportunities that we think AI will create for us, but we also know there's potential, Potential risks of harms. And as AI has gotten more ubiquitous and more powerful in our world today, this isn't just something that's operating in a laboratory. It's out among us. It's affecting us in our lives, in our jobs. And I found myself, for instance, testifying before the legislature here in the state as they're trying to think about AI governance and regulation and policy, and how can we unlock that potential for the economy? At the same time, we're smart and trying to think ahead about possible things that might go awry. So I think this space of looking around, what's happening in the world, what's happening at least in the United States, around different states, where we're all trying to think about what's the appropriate policy for AI governance, both at a sort of legal level, but also within organizations. You know, how can your organization unlock the power of that without the risk that you're going to have a PR nightmare, litigation or something like this, copyright infringement, et cetera.
- [04:34]
Unnamed Panelist
What I think of AI, I think of it as an aspiration. Like, and. And I think, like, if you. If you were creating a business and you, you know, the. The people in the Dartmouth conference who coined the term AI, you want them to name your business, like, what a. What a great name. I think it was called mechanical Computational thinking, we wouldn't be here talking about it, you know, but we have to look at AI. It's an aspiration. It was experimental and along the way to achieve this. Just, can we replicate human intelligence on the way to achieving that We've come up with some techniques that are really useful and we use them a lot. We can think about how the future. But like most people in this room, a lot of people interact with an AI system before. They interact with a human now. And what is it doing right now? It's telling people, here's the news you want to read, here's the product you want to read. And with generative AI, maybe it can make up the news you might want to read and make up content, super personalized content. And we can look at the social impact of that right now, as well as the potential future opportunities and risks as well.
- [05:51]
William Fitzgerald
William, you brought up earlier like biases in general. I know that you both follow sort of the ethics and a lot of the ramifications there too. When ChatGPT first launched and you know, Google's AI model that followed, there were a lot of issues around hallucinations. Obviously, you know, biases is part of that as well. Inaccuracies. A lot of the companies have mitigated some of the concerns around hallucinations. Obviously, it still continues, but as the technology evolves, where are we right now in the current state of some of the biggest systemic challenges and what does that future look like? Are a lot of these things that potentially we can work out, or will they always sort of innately be there?
- [06:36]
Dr. Matt Lease
AI is a fallible technology, just like people are fallible. No one will ever be the oracle on top of Delphi who's always correct. So we want to think about how can we design the technology to be maybe akin to the interactions you have with other people in your organization? How do you know how much you should trust them or rely on what they say? What's the kind of interactions you would want to have to clarify points, to know how much you can trust or use something? So I think this idea that we need to work on improving the accuracy of artificial intelligence is absolutely great. But we shouldn't sit around and wait for it to be perfect because it never will be. So we should design from the ground up of it being a fallible technology and think about how we can make that work effectively for us. Just like we deal with fallible people.
- [07:18]
Unnamed Panelist
Around us every day, basic AI literacy will help. And so under the EU act, if you're an organization and you use AI to make decisions in your organization, you have to provide AI literacy to your employees who are going to be using this so that they know when to trust the outcome, when not, and they can use it as a tool and not be fooled. By it. And that can help because generative models aren't built to be accurate. You know that that's by design. They're not dictionaries, you know, so if you understand what they can and can't do, then it'll prevent a lot of use cases where they're used for something they should never have been used for.
- [07:58]
Dr. Susan Levy
What we've seen for our people and the people inside is that they're not using the internal tools as much as you would think because of that lack of trust. Like it is going to be a first draft in reality. And like a lot of times you could just do it faster yourself, smash the actual draft out there instead of having to double check every single thing that the system has done. That is a reality. And like that's something I think that, you know, we've seen recently that Microsoft has cut back from their investments. We've seen that companies are realizing, okay, this is not a catch all, like this is not a situation where we can just deploy the system and it'll, you know, do all the work for us. I think that's the best. You know, especially in 2025, more and more companies are going to come back and realize we now are not going to say this is everything, the investments are going to be better. I think a lot of, you know, there's been a lot of layoffs because of this, because we didn't know what was going on for the last two years. Like, can this do what our engineers are saying that it's going to be doing? And I think a lot of, a lot of the upper management right now are saying no, they're going to adjust their investments accordingly. And like I said, it's not used especially by content people as much in companies because of that lack of trust.
- [09:07]
William Fitzgerald
Speaking of content creation, I guess what's the balance there when it comes to deploying this type of technology ethically and responsibly? You might be watching something and you're not quite sure if it's been AI generated. What sort of roles do companies need to consider and then what information do consumers also need?
- [09:30]
Unnamed Panelist
I really strongly believe that if you're produced, you know, AI generated material, be it text or images, you've got to let people know, like if we don't can't tell what's real and what's not, be it text or images, then our information systems is broken. You know, it's really, really important that we know what's real and not echoing.
- [09:58]
Dr. Matt Lease
Something Susan said earlier. First, we think information literacy, AI literacy has really become an essential part of education today. People have to know something about this because whether you're using an AI tool or just reading something online, it's getting harder and harder to tell what's real and what you should believe. And if we really want our democratic societies to stand up, we need to have people who can be informed voters when it goes to the ballot box. So the Republicans ran an ad showing the future dystopian world under Biden, and they did have a label on the video showing AI generated. If you looked for it, they voluntarily did that. But there's no requirement to have those kinds of disclosures in these ads. Now we have lots of questions again at the state level and the national, however, at least here in the US about what kind of disclosures we should have to have about our AI generated content. But you know, our foreign adversaries are people who are doing disinfluence campaigns. They're not going to certainly worry about any of this kind of stuff. So I think we have huge issues both at beyond sort of product level, issues of national security and etc.
- [11:09]
Dr. Susan Levy
From a cybersecurity perspective, this is a huge issue. And for me personally, it feels horrible to go onto the likes of Instagram or anywhere else like Reddit. You don't know if any of the content these days is real. It is actually getting harder and harder. I was, six months ago, I was like, there's no way I'm gonna fall for this type of stuff. I know what looks like AI generated content, but the way that it's been patched together now, sometimes when a person makes things, it can be sloppy. Now that they're learning from that sloppiness and they're putting the AI into the sloppiness. They've got, you know, people talking on one side of the corner, they'll move it around. It's like that looks like a real person, they've got a voice on top of it, and then they'll tell this story. Everything in that was completely AI generated, from the face to the voice to the story. So there is this theory of this, this dead Internet, and it does worry me greatly, is like the Internet has been massively influential and especially when Google came out with an amazing algorithm allowing us to access all of that real, real data, all that real information. And now so much of it is going to be generated and it's just not going to be as useful. And that's really what we talk about. What is intelligence and what is intelligent content versus what is Artificial. And I think this is really important that we say artificial content is artificial.
- [12:27]
Unnamed Panelist
And what is an intelligent use of a very powerful technology? Like, you know, if you think of children, you know, you can now, you know, ask language models to come up with a script for, like, a kid's cartoon, feed that into another model that might develop animations, put that on YouTube that would use recommender algorithms to send it to kids. So kids could consume cartoons that no human was involved in, you know, and no human knows that they've seen it. It's not transparent. And that's happening right now. That's not a clever use of. Of a powerful technology.
- [13:03]
William Fitzgerald
So I want to just go back to a few things. You know, particularly when you mentioned, like, younger kids, a lot of, you know, AI is now popping up on different social media. You can interact with it on Snapchat. Just want to touch on sort of transparency around minors, too. There might be sort of language around if you're watching an ad on tv. But for kids, when they're interacting with AI and might not even know what AI is or what. What that is like, I don't know how parents approach or talk to their kids about AI or what do companies have to consider when creating content for kids that might be different as an adult? When we understand what's happening behind the scenes.
- [13:45]
Unnamed Panelist
If you know how language models are built and what they do. You wouldn't let a kid anywhere near an AI companion because it's not safe. It couldn't be tested. You know, it's like a little girl playing with a doll, and the doll starts talking back, you know, but no adult can hear, you know, and it develops this personality and can start whispering. You wouldn't know. You wouldn't know. It's totally untested and unpredictable and. And untested. The effect on them, you know, and children and adults, too, but children and, you know, there's old. It's not only children. It's vulnerable people. You know, when you're a kid, if you have a doll or a LEGO figure or whatever, or a superhero, that's real, you know, so we have no idea the effect of these. And so no, children shouldn't be explained to. They should not be near them.
- [14:38]
Dr. Susan Levy
I think I have two kids, two years old and four years old. I'm definitely on the fence a little bit. You know, I used a computer from a very young age. My father was great getting myself into technology, and I'm like, why can't I give my. You know, this is a new piece of Technology, that's exactly what I was using back in the early 90s. I was given a computer. Why shouldn't my child have it? So I definitely have this, you know, dilemma. But then I'm like, where is this going? Especially, you know, when you've seen what's happened with social media over the last 10 years of these echo chambers of the bias that we do have smashing into each other. But even with social media, if I use that as an example of technology having this huge power of the technological algorithms, which in many ways are artificial intelligence by themselves, I think that we now look for human connection more than maybe we have in a very, very long time, maybe ever. Because we've binged on our own bias for such a long time, where it's like, you know, I feel something's wrong, something feels off, and, you know, I keep hearing the things that I want to hear. The algorithm keeps giving me the things that I want to hear, and they feel that their life just isn't fulfilled in that way. So after binging on social media for 10 years, I'm not suggesting that this is a good thing to do, but what happened is that they're looking out now and saying, actually, I want a different point of view. And they end up going and saying, hey, maybe this isn't the best thing to do. I've heard my own opinion so many times and blown up as much as it can be, and now I'm looking for a different opinion. It's hard to see that obviously in this world because we've got such division, but I have seen that in other people where they're looking now for that opposite viewpoint, which could mean something positive could happen 10 years down the road after AI kids. The kids.
- [16:25]
William Fitzgerald
You know, when we're talking about transparency and content creation, there are instances, obviously, perhaps, when companies haven't been transparent and the users start to pick up on, you know, what has been user generated, and then they later come out and say, yes, this has been AI generated. When will we get to a point where we accept that companies are using this and be okay with it? And it's just. That's just a part of life. Do you think that we're getting closer to it? Is it farther away? What needs to happen? If transparency is there, will we be able to accept it?
- [16:58]
Dr. Susan Levy
You know, it's not something that we've. There's a vote out there saying, do you want to accept this or not? Like, companies are going to create advertisements where with AI because it's cheaper, and that's Just what they're going to do. And, you know, Europe definitely is going to have to put in a lot of different regulation to curb this, but that's not going to touch the States, let's be honest. Like, it's not going to. It's not going to happen in America. So in reality, I think we're going to have to be forced to accept that. And I'd say it's probably in the next six months to a year where you're going to have to assume that something there has AI in it. And that's pretty much now, from a.
- [17:36]
Dr. Matt Lease
Design perspective, we sort of talk about this notion of an uncanny valley when the technology is sort of human but not quite right. It's actually much more upsetting to us. And so it's better to sort of design it a little bit less so where we feel a little safer and more comfortable using it. And so like a great example in Pixar movies, they've done a great job of when early on, not trying to make two human, like characters, keeping them looking a little different. And then we're much more comfortable with them. So I think there's this design perspective about trying to think how you present the technology again instead of trying to get it perfect design for where it actually is. But I think in terms of what we have to deal with today, I think it's already there, it's out among us. And I think the challenge isn't necessarily even. Is the content I'm looking at fake? But a political figure says whenever they do anything wrong, oh, I didn't do that. That was Genai. Right. We're so used to that now, being among us and accepting it now you can explain away anything you don't want to say you did by just saying it was fake.
- [18:33]
William Fitzgerald
Maybe let's talk a little bit about legislation and regulation. What we're seeing in the eu, what we're seeing here. There was a big AI summit in Paris recently. Some of the leaders were aligned in certain ways to big topics around deregulation. Maybe we can break that down a little bit of where sort of everyone stands, stands and what that means that we aren't necessarily aligned.
- [18:56]
Dr. Susan Levy
The big thing here, and I've said this is like you've got one side, which is Asia, China, predominantly, who's got a very, very big, big box at AI. You know, they've got massive funding and you can't really see what's going on. It's a whole different, different world. And they've, you know, they can do whatever they want with censorship. And then you've got. The United States is very capitalistic about it and you've got different players. And again, they're these huge AI is going to come from. If you want to talk about AGI, it's going to come from either the States or it's going to come from China. But when you talk about Europe, in reality, what Europe has done so well is the culture is there. Like, we've got some of the best laws and regulation towards how people are treated at work. For me, Europe, especially Ireland, can really be at the forefront of saying humans are going to use this as a tool. We need to actually focus more on the human development of it and say, hey, Irish people, European people are going to be the powerhouse of showing us what actually intelligence is in the first place. I think that's what's going to become very clear here. We're talking about is this AI generated content or not? I think our views of intelligence are going to change. And I think that some of the best examples of actual real human intelligence is going to come from Europe and that's. That's what's going to happen. So you're going to have the United States building artificial, you're going to have China and Asia building artificial. You're going to have Europe saying, hey, intelligence, this is the eye, this is the intelligence that we have behind us. And we're starting that with a lot of the things that we're doing about regulations. But in reality, it's going to be the people showing us that, you know, maybe they can work three days a week and they've still created something really creative and they've done something amazing. I think that's going to happen in Europe and it's going to happen in Ireland.
- [20:43]
Unnamed Panelist
So I was at the AI Summit in Paris there. There certainly was no sense of a backing down from the AI act for sure. And I'm involved in some of the progression of developing the guidelines to implement this AI Act. And I don't think anyone could say it's perfect. And it certainly started being written at a time around 2018, say, when the technology was very different. I think if you were to write an AI act now, you do it differently. Genai wasn't around, so the technology has moved faster. But once you have something in place, then you can change it. But a big focus probably is the Digital Services act is probably going to be more impactful in terms of the AI that we use every day, recommender algorithms and mandating transparency. And I think that's the one that would be quite contentious.
- [21:37]
William Fitzgerald
I think also part of the discussion that happened at the summit too is whether you can have regulation in place without necessarily curbing innovation. Where do you feel like that intersection will be? Like, do you think we have to figure out how the two can dance together and make sure that like, different countries are aligned? If we have more regulation in the U.S. perhaps, perhaps elsewhere, other countries will surpass other nations. And we're all trying to be superpowers in AI. So how does that fit into the equation?
- [22:12]
Unnamed Panelist
It is a big discussion in the eu. How do you get that balance right? The answer to that, I think is we have got it right before. Like who wants to go to a chemist and buy a medicine that hasn't been rigorously tested on? Nobody would buy that. And the same goes for technology and technical products. So we've done it in a different. And if you have a company and you're rolling out a product that has AI in it and it's making crucial decisions about people's lives, you want to be given a checklist to make sure, okay, I've done this test, so I've done my best to make sure this is safe so it distributes. So if something goes wrong, at least it's not all your fault. You can say, well, I complied with this. So I think creating that safe space, distributing risk can only be good for.
- [23:06]
Dr. Matt Lease
Innovation in many ways. We often do what's called just in time compliance. So let the autonomous car out there and only after it runs into the bicyclist do we figure something out. Maybe that's not a good example. Maybe we should be thinking a little better about how we can anticipate some of these things in advance. So I think one of the most important things we can be really doing is convening the stakeholders to have these conversations and sort of thinking about in the democratic societies of tomorrow, in exchange of human rights and human dignity, what are the sorts of rights and protections we want to aspire to? Another thing that we must do is sort of engage these multi party stakeholder conversations because you need the policymaker with the technologist, with the researcher who can really sit down and think about these sorts of things. And we need these multi stakeholder groups and sort of understanding how technology is going to be impacting all these different constituency groups. But I think the point you make about sort of national security or sort of the international climate is true. We in democratic nations will be doing more to protect our citizenry than in autocratic countries. And so we have to be aware that, that that is going to be an imbalance driving some of the technology development. How can we partner among the nations to stand up for the kinds of rights and technology development against some of the nations that are not.
- [24:24]
William Fitzgerald
It's interesting too how different platforms and different companies operate in different countries. How will companies need to work with other countries when there are different ethics at play or different sort of visions and regulations on the table?
- [24:44]
Dr. Susan Levy
Carefully. That's the biggest thing. I mean, what's happened with a lot of large multinationals is that we've, especially recently, some of the big, big things have happened to the meta Google in the European Union. Like it's, it's a dangerous thing. So they have to go in there very carefully and say, how do we want to be operating? But again, I think it's important that they're in those areas and they're going to stay invested in those areas because of the people, like they're not doing that just because of the market itself. You know, they have massive partnerships with these countries. They have huge, again, a lot of our people are working for these companies. So they want access to that because they want the people there not just as a, you know, to sell the products to them, but they want to get it, so they will do it. And I think that it's, it's a hugely important thing that Europe maintains and says, hey, here's this hole that you need to run it through because this is how we want to live our lives. This is the choice that we're making from a big picture and I think that they should do that accordingly.
- [25:41]
William Fitzgerald
Do you think there are regulatory actions already in place or being talked about in certain areas that could potentially be rolled out elsewhere as like an example of, you know, something that could potentially work across industries or in general?
- [25:56]
Dr. Susan Levy
That's a really big question because I'm not sure if you'd bring European regulation here. I don't think that it would work in China. It wouldn't work in, for the most part. I don't think that is going to work. Like, we've really taken a stance again where Europe is about the people capital, whereas America is very much about the systems itself and just, you know, brute force. We will find a way into these technologies and, and China, again, it's the amount that's being thrown into it. So for the most part, I, I don't think that they can cross and play, but they, they can learn by moving into that country and then saying, how do we apply some of the principles? But Just a big actual regulation. It's kind of, it doesn't fit.
- [26:34]
Unnamed Panelist
A lot of times the principles do apply, you know, so, so the Prince, so the EU act, the Digital Services act, it all links back to the fundamental principles of human rights. So if you buy into that, then you hold on to those principles and you build from there and then that leads you to protecting human autonomy, freedom, freedom of thought, all that.
- [26:57]
Dr. Matt Lease
One of the fantastic and frustrating things about this beautiful US of A. Is we have these 50 different states. And so a lot of what happens is these states are sort of startups in terms of trying out different kinds of policy and regulation and seeing what works. So I think we're all looking around internationally as well as, you know, domestically, and trying to see how other people are trying to craft appropriate policies. What we can learn from different people, pick and choose the parts that make sense from us and kind of all learn together. And the most important thing is that we're diving in, trying to do it, because I think we can only learn by doing.
- [27:30]
William Fitzgerald
It's also just hard to write, I would think, legislation, or look at this more closely as technology is evolving so quickly too. So it's like, of course you want to do it as it's growing from the ground up, but it's. We don't even know really what six months look like. I mean, from a legal standpoint too, like what are legal experts saying and what are you hearing from that side?
- [27:51]
Dr. Matt Lease
The easiest thing to do if you're a company is the just in time compliance. So do whatever makes the most sense for your business revenue right now, but be prepared to adapt quickly if some new regulation comes, comes down the road. But I think, you know, the risk, of course, is if we don't think ahead, then we can have some of these problems that are much more expensive or harder to recover from later.
- [28:11]
Unnamed Panelist
What I heard from a company, which I thought was a really interesting point, she said she's the CEO of a company, she says, well, we just make sure we're ethical. Because if you're ethical in how you use AI and deploy AI, then you're going to be right side of the law because like it really is in early stages and you know, at the moment, AI regulation. So if you're ethical, then you'll be fine.
- [28:34]
Dr. Susan Levy
What is ethics in the, in the first place is something that I think a lot of people struggle with. You know, that's why, you know, it's hard to say, is it ethical? Because a lot of times we can't answer that. Question. So we kind of will put that back. Like, what does that mean?
- [28:49]
Unnamed Panelist
Yeah. So if you're making a decision about someone's life, have you looked at the data your model is trained on? Have you tried to. Have you tried as much as you can? And that's what the legislation says is, have you done your best to make sure your training data is bias free? Have you tested its decisions against different kinds of population to make sure it's not going to make stupid mistakes? You know, have you done your best and use state of the art methods to make sure that your product is safe?
- [29:17]
Dr. Matt Lease
I think what's really important is we do engage the researchers, the futurists who are really thinking and trying to anticipate what technology roadmaps are looking like. There's a lot of reports coming out of think tanks and different organizations trying to anticipate a lot. Of course there will be things we can't, but that's one thing we do. And then on the ethical front, I think one of the most important things you can do is engage all these diverse stakeholders in your conversations and thinking about the technologies we're building and who's being impacted out there. Because I think where we've had a lot of the pie in your face moments where people again, either face litigation or they face PR snafus, it's because they weren't thinking about some other group or they didn't have that other group potentially involved in the technology's development and that led to a dramatic oversight. So putting people first is one of the best things you can do sort of in technology development.
- [30:07]
William Fitzgerald
I guess from if you are a consumer, even if you work for a business or a startup or whatever company, we're all consumers at the end of the day, what does the everyday person have to get excited about in the AI world in the next weeks, months, years even?
- [30:24]
Dr. Matt Lease
So we're all excited about the progress that brings new innovations into our daily lives. But all of that has often been manual scientists following a certain process. We're thinking about using AI to innovate. So much of the way we work and the way we interact with one another, what parts of the technology, the scientific process, whether it's R and D in a company or science that happens in the university, what parts of that can be accelerated, automated, assisted, to really speed up that workflow and the pace of development, such that we don't just have the new things that are exciting to us every six months, three months, two months, but even faster. I think that's a really exciting thing to think about AI being partners with us in the scientific discovery process.
- [31:07]
Dr. Susan Levy
Yeah. And generally I'm very optimistic. I do think that once we figure things out, I think the tools we're going to be using now, just like it has been in the past with social media, when it comes to Google, when it comes to calculators, I think we are going to have more fulfilling jobs. I think that we're going to be doing things that we enjoy more. I think that, you know, I'm very excited about what's going to be happening. You can practice and you can do more with content creation. Are you going to be able to experiment more with it? Yes. Is it going to potentially like have an issue on critical thinking with education? At some point? Yes, but then there will be an urge for wanting something more that's deeper and there's always a human connection to connect. And I think that that is the part that people are looking for more and more. I think that's what Austin is about in a lot of ways. And I think because of technology, AI particularly, it is going to help us become more ourselves. It's going to be helping us become more human.
- [32:02]
Unnamed Panelist
Healthcare, the future of healthcare, I think, and science and the pace of innovations is going to be mind blowing. I think the amount of experiments you can do and simulated environments to conduct experiments and healthcare, it will be transformed, I think. And I'd love to see, you know, the intellectual energy that we spend, you know, getting teenagers to look at little black boxes, you know, for hours a day and solve, you know, big problems in the world instead, you know, redirect some of that mental energy on solving, you know, using AI for, to solve intelligent problems.
- [32:38]
Dr. Susan Levy
Whenever you're thinking about AI, for me it's about thinking about the eye. I do think that's something that I'm very passionate about. I think that, you know, it's just a tool. It's not going to replace what you really love about your job. It's going to replace first with the things you probably most hate about your job. And therefore you might have an extra 20% of your job that you didn't realize you had before. That's going to give you time and a space. Like, what am I actually passionate about? What do I really want to do in my job? And so AI is just about the eye. And that's, that's that intelligence and that's the human that never going to be replaced.
- [33:11]
Dr. Matt Lease
You know, whenever we have any new technology development, whether it's cloning or something like this, we're always asking these questions about how can we realize the optimal benefits for society at the same time anticipate head off, mitigate potential risks. So I think AI is just part of a much broader conversation we're having about that with all kinds of developments. A second one is, I think, a lot about information. If you're talking about, again, where something's real or not, whether it's AI generated, you have these same questions about information on the web. People can be writing stuff. They might not do it at the same pace, they might not do it at the same scale, but you also don't know if stuff there is necessarily believable or not. So a lot of the problems we have are really information problems rather than AI problems.
- [33:53]
William Fitzgerald
Amazing. Thank you guys so much. It's such a fascinating time, really, in tech and it'll be interesting how just next year at south by, how perhaps these conversations will be different or what's new. It's just evolving so quickly. So thank you so much for taking the time. Thank you so much. And thank you for everyone tuning in.
- [34:13]
Samantha Murphy Kelly
I want to thank once again the panelists, William Fitzgerald, Susan Levy and Matt Leese and the team at Ireland House. The Next Innovation is produced by Situation Room Studios and powered by Enterprise Ireland. Investing in the next wave of innovation. Innovation. Our executive producer is Christine Barata and our senior producer is Sharon Barreiro. Emily Beeman is our associate producer. Additional production assistance by Global Situation Room. I'm your host, Samantha Murphy. Kelly, until next time.