AI or Not
Welcome to "AI or Not," the podcast where digital transformation meets real-world wisdom, hosted by Pamela Isom. With over 25 years of guiding the top echelons of corporate, public and private sectors through the ever-evolving digital landscape, Pamela, CEO and Founder of IsAdvice & Consulting LLC, is your expert navigator in the exploration of artificial intelligence, innovation, cyber, data, and ethical decision-making. This show demystifies the complexities of AI, digital disruption, and emerging technologies, focusing on their impact on business strategies, governance, product innovations, and societal well-being. Whether you're a professional seeking to leverage AI for sustainable growth, a leader aiming to navigate the digital terrain ethically, or an innovator looking to make a meaningful impact, "AI or Not" offers a unique blend of insights, experiences, and discussions that illuminate the path forward in the digital age. Join us as we delve into the world where technology meets humanity, with Pamela Isom leading the conversation.
AI or Not
E069 - AI or Not - Chris Voicu and Pamela Isom
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Welcome to "AI or Not," the podcast where we explore the intersection of digital transformation and real-world wisdom, hosted by the accomplished Pamela Isom. With over 25 years of experience guiding leaders in corporate, public, and private sectors, Pamela, the CEO and Founder of IsAdvice & Consulting LLC, is a veteran in successfully navigating the complex realms of artificial intelligence, innovation, cyber issues, governance, data management, and ethical decision-making.
Your smart home can turn on lights, but can it understand your day, protect your privacy, and help before you even think to ask? That’s the gap we dig into with Chris Voicu, a mentor, engineer, professor, and part-time entrepreneur building the Prototype Living of Tomorrow (Plot) at Kennesaw State University.
We talk about what it really means to create an intelligent environment, where sensing and AI are integrated into infrastructure, not bolted on as a voice command. Chris breaks down how multimodal AI can pull context from voice, images, and real-world signals, and how agentic AI and embodied AI could power everything from assistive robotics to delivery systems and community-scale automation. Along the way, we keep the focus on human-centered AI that augments people, especially as the aging population grows and caregiver capacity gets stretched.
We also get honest about the hard parts: latency, compute limits on real devices, and why privacy, safety, and security have to be designed from the first sketch. Chris makes the case for edge AI and local AI models that keep sensitive data in the home, plus a practical hybrid approach with encrypted backups and selective sharing for hospitals, banks, and other services. We close by clearing up a major misconception about AGI and why today’s systems are still far from human-level sentience, even as capabilities accelerate.
[00:00] Pamela Isom: This podcast is for informational purposes only.
[00:26] Personal views and opinions expressed by our podcast guests are their own and not legal advice.
[00:34] Neither health tax, nor professional nor official statements by their organizations.
[00:42] Guest views may not be those of the host.
[00:51] Hello and welcome to AI or Not, the podcast where business leaders from around the globe share wisdom and insight that we need right now to address issues and guide success in our artificial intelligence and digital transformation journeys.
[01:07] I am Pamela Isom and I am the podcast host.
[01:11] Okay,
[01:12] so if you're listening,
[01:14] we have a special guest today.
[01:17] His name is Chris Voicu and Chris is a mentor and engineer,
[01:24] a professor and part time entrepreneur and he's the leader of something called plot, which I won't explain here. I'm going to let Chris do it. So Chris, welcome to AI Or Not.
[01:37] Chris Voicu: Thank you so much Pamela. And thank you so much for the opportunity to join you today on the conversation, especially about AI because it's starting to develop quite a bit. And I also want to say a big thank you to La Rhonda for making the introductions as well.
[01:49] Pamela Isom: Awesome. Thank you for joining me. I am excited to have you here. You have a lot to say and you've been involved and a lot of things that I think are really relevant here.
[02:00] So it's a, it's fabulous to have you as a guest on the show.
[02:03] Tell me more about yourself, your career journey.
[02:07] I would like you to share more about PLOT and your initiative with Kennesaw State University.
[02:12] Chris Voicu: Certainly. So to get started, actually I started a business when I was 18 years old. I worked full time in my own startup while I was in school doing my undergrad as well as doing my graduate studies.
[02:24] And what you will notice throughout my entire career is I kind of went in different fields. I started in basically high powered electrical systems, then moved quickly to low powered electronics.
[02:34] And then for my graduate studies I started looking at things like telecommunication. And then I started being focused a lot on the biomedical aspect.
[02:42] And the reason for the changes is I noticed that the things you learn in one discipline are 100% applicable in another discipline.
[02:49] And so throughout this entire process I realized that you can learn a lot of things that you can apply, for example, in fields that are more important like biomedical, because biomedical will allow you to kind of help individuals, humans for a longer term solution and being able to provide different solutions that can take the things you're learning and apply in that exact area.
[03:10] And so AI is the next big thing. And AI is basically being integrated into that. And I was fortunate to have the opportunity to work with it a lot earlier and be able to actually develop a lot of the models.
[03:21] And with all of this experience, I've had the opportunity to be a faculty and in robotics and mechatronics engineering at Kennesaw State University.
[03:29] And I've also had the opportunity to lead an initiative called the Prototype Living of Tomorrow.
[03:34] And this initiative has a really nice area. It has 25 acres, kind of like a small campus and a home where we're basically trying to integrate intelligence in the living systems, including in your home, but also outside of the home.
[03:48] And we're also looking at how can we make an intelligent campus,
[03:51] not a university campus, but a living campus, like a community. And integrating that intelligence, 100% deals with AI and integrates AI in.
[03:59] Pamela Isom: And you said in and outside of the home.
[04:03] Can you elaborate more? And you said the campus. So can you just talk about those three? What do you mean there?
[04:10] Chris Voicu: A hundred percent. So the home of tomorrow, beyond what a smart home is today. Because right now, systems that are intelligent, like Amazon, Alexa or the actual Google systems, they basically are reactive.
[04:22] They wait for you to provide a command, turn on the music, turn on the light, turn off the light. They do have some sensing capabilities. Like, you know, did you leave the actual refrigerator door open?
[04:31] Or some small intelligence systems into it?
[04:35] But what we're talking about is integrating intelligence into the infrastructure. Just imagine like the wall being able to sense not just the temperature, but vibrations,
[04:43] the floor being able to sense when you step on it. Imagine the actual space that you're living in is so intelligent that the same way you have human perception, the similar sensing capabilities integrate it into the actual home so that anywhere you're at within that space,
[04:59] you can communicate with somebody that cares about you 24 hours a day in a continuous fashion.
[05:05] Well, that system can exist on the outside as well. For example, if you're walking in the park or if you're going through different areas and you need some logistical help, be it, you know, delivery systems, you can integrate all of that intelligence so that it is proactive, gets to learn about you and be able to provide you support without you having to ask for it.
[05:24] And the initiative at Kennesaw State University is exactly that. We're trying to kind of put that together. How we have the space, we have the home, we have the devices.
[05:33] And now we're basically doing the research to make sure that it is plausible and fully functional.
[05:40] Pamela Isom: How real is that? Is that in the near future? Like, where do you see this? How real is this?
[05:46] Chris Voicu: It's starting to become a reality pretty fast. We already see AI systems advancing drastically fast on a weekly basis. We see new models, and these models are slightly different from all AI models.
[05:58] But we see a lot of new models being put on the market every week now, pretty much every day really.
[06:05] And some of these models can understand the context of what you're saying, can understand what you want,
[06:11] simply by just having some textual information.
[06:14] Now, that textual information can be automatically gathered from images, can be automatically gathered from voice, for example, and sound. So when you're speaking,
[06:23] that converts it into an actual statement that can be read by a machine and understand exactly what you're requesting.
[06:29] Well, that can go beyond that, because if you take a frame like an image, for example, and you're sending that to a model that can understand what the image is,
[06:39] understand what the actual scene describes, then all of a sudden it understands the environment the same way we understand it when we look at things.
[06:47] And so it's not too far out. I think in the near term we'll be able to apply these. I will be able to actually apply them in the space and be able to actually automate the space with a lot of different types of systems, both integrated, but also robotic systems,
[06:59] things like humanoid systems, things like delivery system, autonomous vehicles, all of these could be integrated together and function as a single unit.
[07:08] Pamela Isom: Interesting. Talk about humanoids some more. So we had talked earlier about differentiating between what we mean by human like and the human centered capabilities.
[07:21] Can you talk some more about that?
[07:24] Chris Voicu: 100%. So the human centered capabilities is augmenting the capabilities of individuals. And during different stages of life, we may need more or less help.
[07:34] For example, during the early stages, we might need additional support to kind of have the experiences that might exist in areas that we don't have access to.
[07:42] Different spaces, including different cultures, but also different spaces physically, like being able to touch items and be able to feel things that we may not have the opportunity to do.
[07:51] So.
[07:52] And so the human to AI interaction in that aspect is a little bit different than human like. What we're trying to is humans have a really, really unique system. We basically have billions of sensors integrated into us.
[08:06] And so we can feel a lot of things, things like vibrations, temperatures.
[08:10] We can feel the actual environment and be able to comprehend what's happening rapidly.
[08:16] Machines can't do that.
[08:17] And so what we're trying to do is work towards how can we develop systems that are more human like as well. So because if they can understand the environment the same way we do and have a better environmental Awareness,
[08:28] then they can provide us help because they can understand what kind of help we need and when we need it, even before we ask for it.
[08:37] And so that's what I was referring to as far as like, the human like is mostly in the idea that if we can get machines to see the environment the same way we do, then they can make better decisions as well.
[08:48] Pamela Isom: So how exactly is that safe?
[08:51] Chris Voicu: So the safety part is definitely an important aspect, and that's why it has to be thought about, especially for different areas,
[08:59] be it both physical safety, where you actually have hard physical locks, as well as the architecture designed to be initially integrating safety not as a policy,
[09:09] but in the initial conception.
[09:12] Because undoubtedly these same benefits you can get from like an exoskeleton helping you walk or stand up can also be a really harmful system if it goes beyond the safety limit.
[09:22] And so it's safe by integrating safety into it from the very conception,
[09:29] like privacy,
[09:30] that's other side of it. So the privacy also needs to be considered. So privacy is one of those things that is critical and is also something that a lot of individuals fear.
[09:40] So privacy needs to be kept as an utmost importance. And so how do we tackle that? For example, in our initiative,
[09:47] we look at privacy in two modalities.
[09:50] One,
[09:50] privacy is really important to be kept with yourself or with you personally. And so we're looking at models that will live in the home and not in the cloud. And so local models will allow for a lot of benefits, including more rapid responses,
[10:05] but it will also allow us to kind of keep all of the data we want private locally and not provide it to the world.
[10:12] And when we want to release it, we can certainly do that.
[10:15] And the reason that is important is because sometimes we do want to give information to our medical professional in different areas,
[10:22] but otherwise, the data that we want to keep private are private thoughts. For example, even the ideas that we might have if we have new discoveries,
[10:30] are we talking privately with a colleague or family member? We don't want that to be existing somewhere in the cloud. And so that's why we're looking at it in a different modality on creating systems that can also collect data so that it can understand you and understand everything about you,
[10:47] but keeping that local so that it is at utmost private.
[10:53] Pamela Isom: The work that you're doing is just fascinating. I mean, I know we're trying to get there. I know we've been saying everybody's been talking about AGI,
[11:00] and this sounds like more like a more realistic approach to where you want to take this, right, this sounds more like real world applications and just a more realistic way of communicating the automation that we want and the sensing capabilities that we want to be integrated into the AI.
[11:20] I like your definition of the human like,
[11:23] because we know that one of the goals is to get the AI to be more human like from the standpoint of humanoids, but not taking over what humans should and could be doing.
[11:38] Right. So you showed the intermix of human like and human centered capabilities.
[11:45] And I think that's really important.
[11:47] Before I ask you another question, I want to know in that arena, is there anything else that you wanted to share with me?
[11:55] Chris Voicu: So 100%,
[11:57] when we talk about human like, we don't want systems like robotic systems to replace humanity because that would be a disastrous trend. And so what we're looking at is how can we have a lot of these humanoid systems provide support and just thinking about the aging population, because by 2030,
[12:14] one out of five Americans will be over the age of 65.
[12:18] And so we will need additional support and have help us, for example, pick up the actual groceries from outside if somebody delivers it to the door, or being able to actually make the food, or being able to actually prepare the home so that whenever you're ready to go to bed,
[12:33] you have everything prepared.
[12:35] Those type of systems will augment the capabilities and will augment the solutions of what currently exists. Because we don't have enough support for like caretakers and we don't have sufficient caretakers to actually take on the role to take care of the aging population.
[12:49] And so on that aspect, we want to be able to augment, but we don't want to replace the individual. Because in the end, all of these discoveries that we're making is to help people,
[13:00] not to replace them.
[13:02] Pamela Isom: So we need to keep that front and center 100%.
[13:05] Chris Voicu: Without it, all the work we're doing will be for nothing. Because if they end up eventually taking over,
[13:12] which is a fear that a lot of people have,
[13:15] and so it is a valid fear, but at the same time the benefits significantly outweigh the disadvantages. And so we shouldn't stop developing it just because there is a possibility of something like that,
[13:29] because we can see the support that it can provide us, like treating cancer or being able to keep people alive, not just a hundred years, but even 300 or above years, for example, if we can discover some biological cues.
[13:42] Pamela Isom: So describe an intelligent environment.
[13:46] Chris Voicu: So an intelligent environment is a little bit more unique than, for example, like a smart home. An intelligent environment is that that has intelligence into the actual,
[13:55] for example,
[13:57] infrastructure like your walls, your floor, your actual systems in the house can collaborate together.
[14:04] And so even if you have, for example, appliances like stoves and et cetera, you might in the future have an actual robotic hand that can prepare the food and then you work together.
[14:14] And that requires you to be able to understand the environment really well.
[14:18] And so you'll have to have really good sensing capabilities and then be able to collaborate between them. Also requires a different modality, because now all the systems have to work together.
[14:28] They can't just work independently because the manufacturer is different.
[14:32] They have to be able to talk to each other. And so that is one thing where intelligence will be a little bit different than what smart homeopathy.
[14:40] Pamela Isom: And how does that tie to agentic AI that's on the scene today?
[14:46] Chris Voicu: So agentic AI is basically a full integration of everything I've been describing, because agentic systems are basically deciding, sensing, acting. And then now you can also include interacting because you're interacting with individuals or other machines, for example.
[15:01] And so in order to do this, your systems have to be able to make decisions and understand the context. They have to have really good sensing capabilities,
[15:09] and then they have to be able to act in the real world.
[15:12] And so what we hear about, for example,
[15:14] embodied AI or physical AI, those are really critical because it allows us to take software systems and software solutions and put them in the real environment where they can actually, for example, prepare food, open the door, deliver things to your door,
[15:30] do things in the park, for example, turning on the lights as you're walking through it, provide you support if you're too tired,
[15:36] or even deliver you a water, for example, regardless of where you're at in the park if you need it.
[15:44] Pamela Isom: It's not funny, it's serious. It's just fascinating that it's getting closer to, to reality. So I'm, I'm impressed.
[15:53] So the, the next question that I have is, and it kind of builds upon what you've been talking about, but can you describe the challenges as well as the opportunities?
[16:04] I think you've kind of gone into some, many of the opportunities, but where are the challenges?
[16:10] Chris Voicu: So the challenges undoubtedly are threefold. One of the challenges is in the development of AI systems. How do we make them more intelligent? How do we make them faster? Because right now latency is a big issue.
[16:22] So right now, if you ask, for example, even ChatGPT,
[16:26] give me an email, for example, it takes some time to generate that email.
[16:30] And so instead of taking time away from you by waiting for the prompt, we need to make them faster.
[16:36] And so improving the models is one of the challenges. The second challenge is taking that intelligence and putting it on a real system, because real systems don't have the same resources that a super computer lab, for example, might have.
[16:51] So whenever we talk about supercomputing, we're talking about warehouses of these machines. Well, a robot system like a humanoid or even an intelligent car can't have one of those on location.
[17:02] And so how do we develop these systems to basically be just as good locally as they are in the cloud? And so that's another challenge that has to be overcome.
[17:12] And there are several different ways that this has been tackled.
[17:16] And then finally, the last challenge,
[17:18] how do we ensure that privacy and safety and everything is integrated, including security? Because we've already heard in the news a lot about security being a big deal,
[17:28] it's going to potentially get even worse. So in this particular aspect, we need to be able to actually have solutions as early as possible, so that if there's an AI model that attacks computers worldwide,
[17:40] there are some mitigation techniques or some basically system capabilities that will allow you to protect yourself and your data and even some of the private information that you might not want to provide.
[17:52] Pamela Isom: And you said, we're starting to address this today. Can you elaborate a little bit more?
[17:58] Chris Voicu: 100%. So, for example, the challenges would be models. We personally are working on developing new models that can work efficiently by doing a lot of tool sets or tooling that integrates safety into it.
[18:11] So you have some hard limits that cannot be overcome. Even if a machine model, for example, requests you to do something, there is a hard limit on what it can do.
[18:21] And so we're developing those models to basically be efficient, but also to integrate safety into it.
[18:26] At the same time, on the computing side, if we use these smaller tool sets and smaller systems that we can literally tie together,
[18:36] then they can work just as efficiently as a cloud model, but in a local system. So we're improving that a little bit as well.
[18:43] On the privacy side, though, that's the one that is the most difficult because we certainly want to keep it as private as possible.
[18:49] And so we, the approach that we're taking is, as I had mentioned earlier,
[18:53] we're trying to develop systems that can accumulate large amounts of data on location, because then all of your data stays close to you and then you can choose what you want to release.
[19:05] You don't necessarily have to worry about it being in a third party cloud that may or may not be 100% secure.
[19:13] Pamela Isom: I appreciate the idea.
[19:15] The concern that I still have is too much localization because, I mean, there's always good and then there's always that opportunity. Right. So how are we then going to balance that?
[19:29] Because, you know, like a bank, I don't want all my stuff at home so that it's accessible if I make a mistake and forget to lock my door or I don't want everything right there.
[19:42] I need it in a bank somewhere. Right. So I think about that here. I don't want all my information local.
[19:48] So how are we going to deal with that? Because now everything is. We move to localization, which I'm in favor of. I'm in favor of edge AI, I'm in favor of as much as possible that, that we can do,
[20:00] put it at the edge.
[20:01] But I'm cautious because it's vulnerable. The more you localize, isn't that introducing another vulnerability in some aspects again?
[20:12] Chris Voicu: So whenever you move everything local, one of the biggest challenges, for example, is data loss. Because if something happens,
[20:18] the home burns, for example, the systems go up in chaos and you might lose access to all of that data.
[20:24] And so the idea is to allow you to make backups that would be encrypted, fully encrypted, if you wanted to on the cloud.
[20:31] And then there is the ability to actually share data that you would want to share with the cloud system so that every system can interact with. Yeah, they say, for example, you mentioned your banking system will have access to your banking financial information.
[20:42] Then your hospital would have. The hospital would have access to your medical records. And all of those would basically have subsets of your data.
[20:51] But the core data especially, because now machines will be able to learn things about you on a continuous basis.
[20:57] We're talking about from the minute you wake up to the minute you go to bed, they, even while you're sleeping, they'll be able to collect data and have that information continuously.
[21:07] And so what we don't want to happen is have that on the cloud because then that data can be maliciously used. They know when you're not home, they know when you're going to bed, they know your scheduling.
[21:17] So that information would be better at home. And in some aspects,
[21:21] if it gets lost, it's not the end of the world. Because as long as it already has all the information it needs and you already have all the predictors, and you already have your wellness board, for example, or how are you doing,
[21:33] then if the data that was collected from the continuous fashion was missing, it won't affect you personally. Because the core data you care about has already been sent to the cloud, for example.
[21:44] Pamela Isom: I like it.
[21:46] So considering everything that you've talked about, we've hit on edge AI, we've hit on agentic,
[21:52] we've talked robotics,
[21:54] and the fact that AI needs to be sure that these capabilities augment, not replace,
[22:01] human capability and human agency.
[22:04] And so those are all interesting things. I even appreciate the discussion on multimodal, but I do have a follow up and that is multimodal and assisted technologies.
[22:19] Do you see that as a sense of urgency and is there an opportunity to expand capabilities?
[22:27] Chris Voicu: So it's definitely a sense of urgency because the more we can provide assistive technologies, the more help we can provide to individuals.
[22:35] And so the part time living up Tomorrow is actually collaborating with some of the other research centers at Kennesaw, including the MOVE Center.
[22:41] And what that entails is basically developing technologies that can assist individuals who might need, for example, rehabilitation,
[22:49] or they might need support with walking, or they might have some kind of ailments, including, for example,
[22:55] immobilizations from the leg. So they have to be utilizing a wheelchair.
[22:59] And in that particular case, these assistive technologies can help them, instead of using a wheelchair, walk in the near term or can help them do things that are limiting the individual.
[23:10] And so the faster we develop systems that are fully functional and fully basically applicable to individuals, the faster they can have some more flexibility. Because then if you have a system helping you walk,
[23:23] you can do things that you were not able to, for example, if you were in a wheelchair.
[23:28] And so we definitely have an urgency for that. We need to develop them as quickly as possible and we need to be able to deploy them.
[23:34] But at the same time, we need to make sure that they're safe. And so there is a layer of basically testing that needs to happen so that everything that you're developing is actually doing what you're intending it.
[23:47] Pamela Isom: Yeah,
[23:48] it's exciting,
[23:49] it's good, it's exciting, it's scary,
[23:51] it's real,
[23:53] it's good. We need it.
[23:55] We just need to be very careful and very strategic about what we do. And I agree there needs to be a sense of urgency. So that's some like serious balancing that we have to do.
[24:05] Right? Because there's a need and especially,
[24:08] you know, assistive. Assistive technologies is a passion of mine. These tools are expensive.
[24:16] The things that we need to support us are expensive. Walkers are expensive. Right? Like these. These tools are expensive and they don't always do the job. I know some people that have taken walkers and Just kicked them aside because they felt like they weren't helpful.
[24:33] Right. Or it was taking too long.
[24:35] Now it's incentive in the, in that case for the person that I know because it forced them to do everything they could to rehabilitate till they could really get rid of it because it wasn't as helpful as they would have liked.
[24:48] It wasn't working the right muscles.
[24:51] So I think that these types of assistive technologies needs to emerge and evolve. I definitely think that we need that. And I do think there needs to be a sense of urgency and I'm in favor of it.
[25:04] I like the work. Now, is this work that you're mentioning here, is that a part of the PLOT or is that getting into what we're doing, what you have going on with ieee?
[25:16] And while you're at it, explain more about IEEE and the CIS program.
[25:22] Chris Voicu: A hundred percent. So the work that I had mentioned, especially on the assistive technologies, is definitely integratable to IEEE as well. But that work actually is also part of plot, because PLOT is a living lab at Kennesaw State University.
[25:35] So we have a home, we have the space outdoor. And so we're going to take a lot of prototypes of things that are being developed in research and apply them in a real world environment.
[25:44] So that's one of the benefits of having that space at Kennesaw State, because these assistive technologies will be able to be tested from prototype to user, basic integration to commercialization on the same location.
[25:57] Having said that, one thing that I do want to add really quick,
[26:01] especially when we talk about assistive technologies,
[26:04] we are looking at systems that can be customized. And so that customization is the next big frontier.
[26:12] Instead of having to basically purchase something that is developed for pretty much an average user,
[26:19] you should have the ability to produce whatever assistive technology you're wanting to on location. And so that's one thing that we're also looking towards, is providing customized assistive units that can be developed in your own home.
[26:32] And AI systems will 100% help.
[26:36] Now,
[26:37] getting back to the IEEE,
[26:38] I am a big supporter for IEEE. IEEE is the largest professional organization in the world. And it's a really great network of individuals from all over the world in different areas and different facets of technology,
[26:52] but even outside of technology.
[26:54] And so this organization is a large community that if you need to learn or find a mentor or be able to progress through your career, it's a huge support supporting pillar.
[27:04] And so ieee, the Computational Intelligence Society, is basically the Artificial Intelligence Society within that professional organization And I've had the opportunity to to chair the Atlanta Section chapter as well as be the Inception chair for it, because I see a huge need for AI system knowledge and AI systems,
[27:25] period,
[27:26] to be integrated everywhere. And because of that reason, I also want students to be ready for that.
[27:31] And so IEEE provides a lot of benefits, including seminars,
[27:35] distinguished speakers, and different types of events.
[27:38] And providing that to students, the community,
[27:41] and tying them to industry professionals and everybody within the IEEE community is a huge benefit.
[27:48] Pamela Isom: That's exciting. So what can the listeners do to engage?
[27:52] Chris Voicu: There's a lot. If you ever want to come visit our prototype Living of Tomorrow, we would be more than glad to host a lot of the actual visitors and viewings. So you can reach out to me or you can send out an email to plotenasaw.
[28:02] Edu.
[28:03] At the same time, with ieee, the engagement is tenfold. There is a lot of volunteer opportunities.
[28:09] And so if you're a student, for example, or a community member that would like to volunteer, IEEE has a lot of basically societies. They have a communication society, they have a computer, they have a computational intelligence, which is the AI society.
[28:24] There's also basically the young professionals, which is basically a group of individuals that are just starting to go into their career phase. And so you can have a lot of events, opportunities, volunteer opportunities, networking opportunities,
[28:40] and those can be found@ieee.org but even with Kennesaw State University, if you're looking for collaborations, if you're looking to test the system,
[28:48] if you would like to see the space, we're always welcoming everybody from the community to students to everybody, including from hub businesses and potential partners.
[28:58] Pamela Isom: That's really good. Of course, you and I have started some collaborations and that's why you're here, because this is important and it's really valuable work going on. And my perspective always is do what you can to be a part of shaping the outcomes.
[29:14] That way you're more satisfied with the outcomes rather than waiting for it to happen and then throwing rocks at it. Right. So my perspective on all of this is,
[29:25] and the reason why I engage like I do is because I want to be a part of the solution and at minimum have some inputs into the equation.
[29:36] So I really, really am excited about the possibilities of things that we are going to do together.
[29:43] But also I hope that the listeners will get in touch with you and look at pursuing some of this.
[29:50] So as I move into the last portion, I'd like to know, is there a myth or a misconception that you would want to clarify or expand upon today?
[30:01] Chris Voicu: One of the myths might be kind of a little bit debatable, depending on who you might ask. For example, so the idea that AI systems are already humans, or the idea that AI systems become humans in the near term,
[30:13] that's something that may possibly happen in the future. But we're not anywhere close to the AI systems that we have today.
[30:20] They're very much controllable. They can very much understand what you're asking in a systematic manner, and they can portray emotion, but that's not really true emotion.
[30:30] And so we will see a lot of evolutions whenever we start developing new models, new capabilities, including the artificial general intelligence integration.
[30:39] But we're so far away from having an AI system be a human.
[30:43] And so what you were talking about earlier being part of the solution, that's important because whenever we're developing these models,
[30:50] the next patients. And so the applications that are going to be developed is going to be dependent on who's asking for whatever need might be.
[31:00] And so in this aspect, if we don't ask for solutions, we don't ask for basically systems that can do things that we want them to.
[31:07] Nobody else is going to. And so it'll give us the ability to have systems that can provide the exact need that we may have.
[31:17] And so that's one of the myths versus the other one. I don't foresee AI systems becoming human anytime soon. It'll be several, you know, 10, 20, even up to 100 years before we have sentient systems that can be fully human.
[31:30] Pamela Isom: Like,
[31:32] that's good.
[31:33] So is there anything else that you wanted to share with us with the listeners and with me here as you move into your parting words of wisdom and or a call to action,
[31:45] what else is on your mind that you want to leave with us?
[31:49] Chris Voicu: So the most important thing, especially from the podcast title AI or not a hundred percent AI, because the benefits, as I had mentioned earlier,
[31:57] will be tremendous. You'll have a lot of support, you'll have a lot of help, you'll have a lot of things that it can do. It can't do it just yet, but it's getting there.
[32:06] So when you wake up, you basically just, you know, say, hey, prepare the coffee, or you say, hey, go ahead and put the tickets on the airplane. Those type of systems will start being more and more available.
[32:16] And so what I recommend to everybody, don't be afraid of it, start using it, start interacting with it, get to know how to use a lot of these AI systems because the younger generation is born into this technology.
[32:28] And so the minute they basically see the world. They already have things like AI systems that they can interact with. They already have tablets, they already have even more advanced technology.
[32:39] But if you wait to start using a lot of these systems, then you'll fall a little bit behind.
[32:44] And then that hesitation may lead to not getting the support you need simply because you haven't yet tried it. And so the biggest call to action,
[32:53] give it a try. It's really wonderful what it can do. And think about the benefits. And then if you have a need,
[32:59] don't hesitate to reach out to even universities and say, hey, this doesn't exist yet. Do you think you could put it together?
[33:05] You know, ideas come from anywhere, and then you might just be creating the next best thing that can help people.
[33:12] Pamela Isom: Okay, that's awesome. I do like your perspective on at least give it a try. Although sometimes it's hard to know where to get started. So I know some of the listeners struggle with that.
[33:23] Well, how do we get started? I already use,
[33:26] for instance,
[33:27] the GPTs. Right. So. But you are talking about more advanced activities and more advanced integration, so that's good.
[33:43] So one of the things that I am curious about is if we consider everything that you've talked about so far and the intelligent and connected capabilities that you're bringing to the table with the, through the PLOT initiative and some of the work with ieee,
[33:59] what are some of the differences between what you're describing and AGI? Can you help us with that?
[34:08] Chris Voicu: Yeah, definitely. So the systems we're talking about in the future, we'll use something like artificial general intelligence. But as of right now, the systems that we currently have are not really AGI.
[34:18] And the reason for that is for AGI to actually function,
[34:21] you'll basically need systems that, like humans, be able to take a lot of information in,
[34:27] produce different results simultaneously, and have those models or those systems interact with each other and adjust the model halfway through getting a result.
[34:37] The current systems that we have are not able to do that just yet. A lot of the models are basically systems that will give you one solution. And what they're referring to AGI.
[34:47] Right now, a lot of the other systems is the ability to have one model call another model to provide that result. So now you basically have like a voice with the text calling together.
[34:58] And so you're getting closer to AGI. But true artificial general intelligence will require that full single model that can do it all at the same time, the same way that the human brain basically works in parallel, be able to provide that solution and affect the answer from different inputs from different,
[35:16] basically, data streams or data types and be able to give you a unified solution that is not yet there.
[35:24] Pamela Isom: I see. So AGI is not in the near term, but what you're describing is closer.
[35:32] Chris Voicu: So it'll be in some aspects. The intelligent systems that we're hoping to develop will use AGI when it's available on database, because it's going to basically integrate intelligence and the sensing capabilities.
[35:42] And that multiple data streams, like sight from the vision from cameras,
[35:47] the audio from voice, and the actual, including things like vibration, ambient information, temperature, all of that will be unified into a single data unit.
[35:56] But using that with an AGI model still needs to be developed. And so those models are definitely under research. And so when there is a solution for that, taking all of that data, making a single model that works across all of them,
[36:10] then we'll have even more improvements,
[36:12] more capabilities.
[36:16] Pamela Isom: I really want to thank you for being here today and this has been a wonderful conversation.
[36:25] Chris Voicu: I just want to say thank you again for the opportunity and thank you for the invite for this amazing talk, and I look forward to speaking with you in the future as well.
[36:34] Thank you again