In this episode, Dr. Prasanta Pal, founder of shiam.io, discusses his journey and the pivotal moments that led to the establishment of his company. He explains what exactly clean data is, its importance, and how his technology revolutionizes traditional AI approaches by enhancing the accuracy and reliability of AI models.
Dr. Pal's work in clean data has yielded some rather amazing results in fields including medical imaging, space imaging, and underwater pipelines. He even talk about possible further applications such as the early detection of diseases and the ability to uncover hidden information in various domains.
The episode explores the future of AI and how it will enable humans to focus on more creative problem-solving while delegating repetitive tasks to AI systems. Overall, it sheds light on the significance of clean data in unlocking the full potential of AI and inspiring individuals to embark on their own AI-driven journeys.
Check out SHIOM
Timestamps:
[00:00:00] "AI Wave podcast with Dr. Prasanta Pal"
[00:03:17] Medical data analysis vital for better life.
[00:08:20] Restoring Mona Lisa: Objective approach vs. tradition.
[00:11:32] Early ultrasound findings reveal unseen details.
[00:15:54] Exploring space reveals hidden layers of information.
[00:21:51] AI helps with repetitive tasks, boosts creativity.
[00:24:24] AI solves boring problems, promotes flourishing humanity.
[00:28:52] AI revolutionizes hazardous conditions, medicine, human capability.
[00:33:50] AI is serious business, have fun!
In this episode, Dr. Prasanta Pal, founder of shiam.io, discusses his journey and the pivotal moments that led to the establishment of his company. He explains what exactly clean data is, its importance, and how his technology revolutionizes traditional AI approaches by enhancing the accuracy and reliability of AI models.
Dr. Pal's work in clean data has yielded some rather amazing results in fields including medical imaging, space imaging, and underwater pipelines. He even talk about possible further applications such as the early detection of diseases and the ability to uncover hidden information in various domains.
The episode explores the future of AI and how it will enable humans to focus on more creative problem-solving while delegating repetitive tasks to AI systems. Overall, it sheds light on the significance of clean data in unlocking the full potential of AI and inspiring individuals to embark on their own AI-driven journeys.
Check out SHIOM
Timestamps:
[00:00:00] "AI Wave podcast with Dr. Prasanta Pal"
[00:03:17] Medical data analysis vital for better life.
[00:08:20] Restoring Mona Lisa: Objective approach vs. tradition.
[00:11:32] Early ultrasound findings reveal unseen details.
[00:15:54] Exploring space reveals hidden layers of information.
[00:21:51] AI helps with repetitive tasks, boosts creativity.
[00:24:24] AI solves boring problems, promotes flourishing humanity.
[00:28:52] AI revolutionizes hazardous conditions, medicine, human capability.
[00:33:50] AI is serious business, have fun!
Chris [00:00:00]:
Welcome, ladies and gentlemen. I'm your host, Chris Parisi, and welcome to the AI Wave, the only podcast that combines the AI revolution with businesses and economies, with a little touch of Rhode Island, bringing hope to this ocean state. In this episode, It's a great 1. We learned a lot. We're gonna talk to the brilliant Dr. Prasanta Pal. And in this interview we're going to discuss how AI has been utilized for data cleaning. What does that mean? What does that sound? Just listen in. This is a great 1 because we explore topics such as the Mona Lisa, space, ultrasounds and medical technologies. It is a great episode. So stay tuned, buckle up and get ready to ride the AI wave. We are here with the 1 and only Dr. Prashanta Paul talking to us about AI and everything that he's been doing for quite some time with AI. The AI revolution is now, but Dr. Prashanta Paul has been doing this for quite some time. Thank you so much for joining us. It's a pleasure, Chris, to be here. You're doing amazing work, spreading the word around AI and the awareness that's come around, the good and bad part, everybody should know. And it's very contemporary. And I think I'm deeply grateful for making this happen. Thank you so much. Well, I appreciate that. And like we've said, you've been doing this for quite some time. You know, when was your first array into AI?
Dr.Pal [00:01:37]:
Well, AI has been there for a while because I've been an academic person and most of my career has been in academia. So AI has been in different disguise like machine learning, model building, which we don't often explicitly say, but main thing happened, I would say, 5, 10 years ago when the neural network came into the reality. And then we realized that it has a lot more potential than the traditional machine learning tools. And that's when people started calling AI. And it's always better to feel and see things to actually realize the potential. Like we see AI-driven tools happening, like self-driving car or semi-self-driving car, and Alexa, and Google Home. So these are like a reality. Then we see what has been conceived as machine doing some of the activities of we humans are very good at are now happening for real. I think the real question is, how far can we go and how much should we go? And that's kind of the interface I'm interested in and building technology around to empower the process, because the end of the day, whatever we do should be for human evolution in the right direction.
Chris [00:02:48]:
That's well said. And you talked about, you know, what are the next steps of AI evolution and how can we utilize that? And that's a topic I would love to discuss. But let's talk about how you have been utilizing AI and you talked about how you've been in academia, but you're more of an entrepreneur these days. Tell us a little bit about the pivotal moments that led you to establish Shayom. I want you to tell them a little bit about that as well.
Dr.Pal [00:03:17]:
Yes, I think Shaom came out of a painful process of realizing that medical data are not being fairly looked at. And your life, my life, our future, everything depends on how good we make analysis and evaluation about our mental health, physical health, everything around our health. The problem is the moment you collect some data and try to make an inference around it, there is so much ambient noise and artifact because of the reality of things, then we are not doing fair judgment to the information underlying that data. And I have been doing a schizophrenia study when I was making prediction about how much is the degree of this disease certain person has. And I realized only depending on a parameter, the person has schizophrenia or not. And that was my wake up call. This cannot be true because objectively, we know these people have schizophrenia, but my data analysis tool is making a decision based on whether or not they have it. So, and same thing is happening for cancer diagnosis and COVID testing and stuff like that. I think the goal is that again, end of the day, we are doing AI to make our lives better. Very simple. So we want to minimize the false positive or negatives. And then we also want to maximize the prediction capability so that when an unknown condition happens, we want to leverage that from our past knowledge.
Chris [00:04:50]:
Wow, well said. So it sounds like you're really into data cleaning. Yes. Would you say, and kind of reducing the noise around data. What are some ways that you're using AI to quote-unquote clean data?
Speaker C [00:05:08]:
So before answering the question, let me ask you a very simple question that do you drive better in mid-February with lots of snow or in a summer day like today, when everything is very clean and clear.
Chris [00:05:21]:
Oh, during the summer days like today. Yes, but you are the same driver.
Speaker C [00:05:26]:
I think I'm doing the pretty much same thing for AI, that AI is a model building based on the data. That is good when the data is clean like today. And when there's a lot of stuff around it, the AI is going to train on that as well, which is often unnecessary. So There are 2 problems there. 1 is we are spending a lot of time and money to train the models because there is a lot of garbage in the data. And the second part is the accuracy of the prediction is actually much more less because when is there is a lot of noise. And AI intrinsically does not solve this problem. So that's why we need a tool to clean the data, then train on AI models. That makes sense, because currently we're all being,
Chris [00:06:11]:
we're all training the, let's say, chat GPT. That's a model that we're all collectively training to let them know what's good, what's bad, right? Because the AI is not necessarily able to decipher what data is clean or not clean. It's pretty objective to it. So what you're saying is it's kind of our job and maybe your mission to have clean data so that when the AI models are processing machine learning, providing whether something may be cancerous or not, it's operating out of a lens of clean data. Is that correct? This is absolutely correct. And I'll add to that is that the way I look at data
Dr.Pal [00:06:53]:
during my academic years, that it's like a colorful onion. You have those cells of the onion and you peel it off, you get the outer layer and you get a new color and a new dimension into the same thing. And there is no stopping point, unless you want to stop, of revelation of the information. So basically when we're training on AI, there is some objective measure that we want to maximize, right? Whether or not it will rain tomorrow, we want to maximize that outcome. So the moment the data specific to that particular objective function is optimized, then your model will do great. So we are not building AI per se, but we are making the data clean and available for a given context so that we can maximize the utility of AI. Wow, well said. That was impressive.
Chris [00:07:42]:
It makes a lot of sense of how the importance of data cleaning is in this AI revolution. So the work you've been doing is incredibly important. And so we can give our audience maybe a specific example. You and I were talking, we met at a few events for Founders and Friends, the startup community here in Rhode Island. And you were telling me about the Mona Lisa. I was like, whoa, what did you just say Dr. Paul? Yes, you're working on cleaning the Mona Lisa. Can you give our audience an understanding of what the application would be like in that specific example?
Dr.Pal [00:08:20]:
That's a great question. And the moment I say Mona Lisa, there are 2 parts to the problem. 1 is the art and beauty around it and the historical weight that it carries. And the second part is the data means there is data behind the art and beauty that we talk about. Now, because of the mother nature, it has been like a 500-year-old painting. So there are a lot of cracks and speckles around the original creation of the Leonardo da Vinci. The Mona Lisa we see today is not the 1 that was created originally. Now there has been many many efforts to restore it chemically as much as possible but still if you go to the museum or see it online, you will see it's definitely not what Leonardo da Vinci created. Now I'm asking the objective question that can we restore it back to its original creation as much as possible, like an engineering fix of a broken bridge? Like, can we make it more useful given the destruction that has already been done? And we did some preliminary work on that, like removing some of the patches, and then we got mesmerized, wow, we see like completely new features of Mona Lisa that nobody has ever seen. And I have access to some very established painters who says, wow, things has not been done this way. What has been done so far is something like more of a cleaning and making it look better. Our approach is exactly opposite. We don't try to make it look better or worse. We try to go to the objective reality because who am I to say whether Leonardo's color choice was good or bad. I'm no 1 to say or no 1 has authority to say that. So this is the whole philosophical difference between traditional restoration and what SIOM is dedicated to that get the original beauty out of Mother Nature from art and painting and scientific data, space, images, audio, video. So it doesn't really matter but the the machines and the degradation effect of the temperature and nature, particularly the third law of thermodynamics make thing happen a bit chaotic, even if you don't want to. And this is the undoing process of that. I am saying, mother nature is adding a lot of creativity. Can we undo that actually to get back to the original?
Chris [00:10:33]:
Wow, so what you're saying is great. You know, you're talking about a objective take on art by cleaning the data to make the, let's say the Mona Lisa as an example, what it was originally depicted, right? Not necessarily what Dr. Prasanta Pal thinks Leonardo da Vinci meant as much as what did he actually draw in that specific example. And you talked about other examples, you just said space, you said medical imaging, you know, if we could talk maybe about medical imaging that's an important 1 You know, I know recently I have a one-year-old and when we were Looking at the ultrasounds. I said do we have a baby or a little alien inside of you, right? You really can't tell much about that, but is that something that you and your company that you started up would be able to make more clean?
Dr.Pal [00:11:32]:
This is an awesome question coming out of the reality of our experiences. And I had a very similar experience when my daughter was born. We had the ultrasound image, and we're very curious, how does she look like? But again, as you said, she looked like an alien. And then because of the very turbulent condition in the mother's womb, it's very hard to image using ultrasound modality. And then some of my early findings using this technology made me look at things that probably nobody else have seen, something like the eye is popping up. The teeth, or rather the jawbone, the formation, early formation of the jawbone are popping up. Then we started seeing the nose, and like, but there is no nose in the original image. So we could not believe at the beginning that this can't be right. But on the other hand, we did not make any effort to put the eyes or the nose anywhere. So it's like naturally coming out. And it's just like being around a gold mine. And if you start kind of shading the soil and the mud around, you start seeing the speckle of the gold. That was the early process when you start seeing things that should be there, like a bright speckle should be around the eyes, and then we validated that, hey, this looks like eyes. And now if we can extrapolate that to MRI, CT scan, and all other modalities of ultrasound, you look at the scale at which you can impact the entire medical diagnosis and intervention space.
Chris [00:13:02]:
Wow. So imagine that, that we would be able to have access to data that is more clean so that we can have a better understanding of what our future child looks like. And maybe at a microscopic level when we're doing biopsies, we can say, you know, well, let's kind of have more of a objective take on this data and maybe that will help decipher things like whether something's cancerous or not.
Dr.Pal [00:13:30]:
Potentially. That's 1 of the pain point that I'm trying to, and that was my own pain point looking at data that people ignore stuff that is not very much statistically significant, but maybe there is that 1 cancer cell that will evolve into like a million cancer cell. So, and existing methodology does not give us access to that subtlety of the cancer cell. So what we are trying to do is that even if there is 1 cancer cell, Can you identify that and recover that whatever is the underlying tissue conditions and whether or not it is cancerous? Right now, things are happening at a very gross level. So when things are really, really bad, we know. But we know it anyways from the pain of the patient. Our approach is that no, like the science can help us even knowing it way before it is actually manifested as a disease. That is when you can actually totally undo the process, totally like reverse the process of having cancer or Alzheimer. We did a little bit of work on Alzheimer where even I'm not an expert in Alzheimer, but we are seeing from historical MRI images, the development of the disease in MRI scans are much more clearly tractable than those done in the traditional method. And again, you extrapolate the same process for every other field. And that's the goal of Xeom is do things at scale. That has been 1 of our like recent objective. When we can do that, we can literally revolutionize the medical field while everybody
Chris [00:15:00]:
will be doing their job just better. All right, so that is great to hear, Doctor. The fact that we can get a little bit more excited about the medical field, but I do have to ask, a lot of trailblazers know I'm very much interested in the beyond, in space, right? And you did mention that there's a lot of data up there in space. We've had some great advancements in technology when it comes to telescopes and how we're able to go even farther than ever before, right? The most recent discoveries that we've been all talking about. Can you tell us about how your technology can make the space a little bit more clearer? Okay, that's a great question. And 1 of my, as a physicist, 1 of my favorite sweet spots. Okay.
Dr.Pal [00:15:54]:
I love to talk about space and I'd love to go to space someday if it is more accessible. 1 problem with the space is that it's not like this studio where there is beautiful light and somebody standing there to capture What is happening and this is the biggest problem? It's a very wild condition under Extremely constrained situation. So the lighting condition is not perfect. The sound waves are not perfect. Almost nothing is perfect. But you happen to get there with the camera. And then you capture whatever bits and pieces you can. And to us, scientists, it's a piece of gold. Now the question is, is that all we can make out of it means the visible arena of whatever is given to us or actually there is, as I said, like the colorful onion, there are many layers of information that we found recently with our fun experiment with James Webb Space telescope selfie to be specific. And first we thought it is all garbage because we started seeing things that makes no sense, literally makes no sense. It should not be there. We started seeing things like alien's eyes, and we started seeing things that was like in complete darkness, like a structure is popping up and how can there be a structure in the middle of nowhere and then debating with other fellow scientists and happens to be that we got in touch with the project manager for James Webb a couple of weeks ago who validated that hey, those structures are the heat shield in the background, but the specific way we are doing stuff, it is, as I said, we are revealing layer by layer, and that kind of gives us the privilege to see things that's not visible as like a bright spectrum of the light. That's often, we as common people often see and believe that, oh, that is the only truth out there. But the way we are approaching that, hey, truth lies in many, many layers. If we unravel 1 after the other, then we get a 360 degree view of things. So essentially what we are doing with space imaging, and I would extend that to underwater low light condition imaging as well, because I'm in conversation with the US Navy people, so that under extremely wild condition of the undersea, with very little light, you can start seeing things that could be an obstruction and often they make accidents there, okay, and very expensive accidents that we can avoid by actually knowing if there is obstruction using the same kind of methodology. And as I said that with this technology, we have access to seeing things out of things that nobody has ever seen. And it's a highly non trivial and very counterintuitive process. That gives me excited, I keep losing hair, but start becoming very happy. Because we see things from extremely old historical images that nobody has ever been able to see the way we are seeing. And this is where I want to differentiate between the seeing and the perception. I mean, you see things, but we don't perceive things because they are often hidden. But I am giving an access layer to the information so that we see and when we want, we can perceive all the depth of the information out there.
Chris [00:18:59]:
That's incredible. Dr. Prasanta Pal, We're talking about how we can clean space data, how we can talk about medical imaging, how we can get that ultrasound, detect potential cancerous cells ahead of time by just being able to clean the data. I mean, this is incredible stuff that you're working on. How does AI help you achieve these applications that you're working on?
Dr.Pal [00:19:24]:
So the crux of AI, 1 has to understand is that we have some input data from various sources and then we create a model, a very robust model that is the effective representative of the data that has been trained on. So what we are doing here is that how more robust we can make so that the speckles from the data are not necessarily removed, but given its own space. 1 of the mistakes that people have been doing, including myself, is we often discard the data thinking, hey, this is noise. It has no meaning. We literally reinvented that real estate of noisy space, and I'm saying that who am I to say noise is noise? There is a lot of different kind of information. If we dig deeper, we just get a very orthogonal type of information space for which I don't need to do a separate measurement. So you basically give me 1 image and I'm giving you 5 different images with very orthogonal and very different kind of information space. So that is Basically, we have 2 eyes. Would it be good if we have 5 eyes and see all different dimension and depth of stuff? That is what is happening here. So the AI is coming here to help us make the robust model building process. And then it's like we are helping AI to build better model and AI is helping us to make the predictions much better. So it's more of a very interactive process and we are helping each other. And I think this is the way our next generation of AI will revolutionize because we'll be based everything on data that is meaningful.
Chris [00:21:06]:
That's incredibly well said. And when it comes to data that is meaningful, you just gave some great examples and applications of how we can make data more meaningful and how we can have what you say a collaborative relationship with AI. And at the top of the show you talked about your relationship with AI where you know there's definitely some positives but there's also some maybe limitations. I don't know if I would call it a love-hate relationship that you have with AI, but maybe it's a complicated relationship that you have with AI. What would you say are some limitations that you feel, whether it's in your specific field or in general when it comes to this AI revolution?
Dr.Pal [00:21:51]:
I think let's trace back a little bit. Why are we even doing this? We are doing this to make humanity better, make our life a little better. And we as human species, 1 thing that is very common to all of us, we don't want to do the repetitive and boring tasks, okay? So let the, I think this is a good thing when you can leverage that AI doesn't complain about doing the same stuff again and again. I will complain. You will complain probably, okay? So I think this is the way I look at it. It's an opportunity for us to be more creative. Some people talk about, hey, are the radiologists are going to lose their job? I would say, no, they will be solving a different level of problem. See, until every single human being, every single species on Earth are happy and flourishing, we have problem to solve. Until that day, we have a lot of problems. I think what AI will change is we'll not be solving the malaria problem anymore because that has been solved. We'll not be solving the early detection problem anymore because that has been solved. Maybe we can dig deeper into much higher order problems like how can we be happy all of us? How we as human species can come together not to destroy each other, but actually go beyond the spaces to maybe colonize Mars, but that literally means that we are going beyond our limitation. I think those are the kind of problem we'll be solving by leveraging what AI can do at scale. I don't see that as a threat, not necessarily, but it is same as like fire and nuclear power. It is up to us how we are going to use it. It's pretty much of the same category. So we are not using blockers anymore. We are using like Uber or other kinds of vehicle services. Does it like make the need for going from point A to point B obsolete? No, the need is still there. Just we are using a different mechanism. We are solving a different kind of problem. So as long as human need is concerned, Those are still there, but just the flavor and the depth of it is changing. And AI will help that as long as we as humanity also evolve along with AI. And we become more conscious individuals along with AI becoming conscious.
Chris [00:23:58]:
You know, 1 quote that we had and that I've always been saying from the beginning on Episode 1 as well is that AI will not replace you, it will enhance you. That's something that resonated with you and that we talked about and you can easily tell it as well that what you just said, right? We're gonna be able to do more with AI. It will enhance us and allow us to focus on other opportunities
Dr.Pal [00:24:24]:
that present itself. Yeah, I would say this is more of an opportunity in my opinion, because I think human species should be more creative about solving problems that are of the next generation kind. So that the pedagogical problems, the boring problems are given to AI. And as I said, that unless every single human being on Earth, every single species on Earth, are actually flourishing, blossoming, we have problem to solve. And so let's make sure that we as humanity take that leap forward to be each other's friend. And this is a great time to do so. And I would say what Shium does differently than many other data science tools is, you remember the time when this guy John Dalton came up with this atomic theory a long time ago? So he said, hey, gold is a big lump of gold. But if you break it down to atoms, then there is only 1 atom of gold, and we only have more gold, and same goes for every other element. Literally that day onward revolutionized the field of chemistry and physics and our understanding of the microscopic world. And now we know, oh, things has its fundamental origin. And once we know this, we can reconstruct or deconstruct as per need. We are pretty much doing the same thing with the data that I got really motivated by John Dalton's work of looking at this massive structure, but he thought, okay, there are atomic elements behind it. We are looking at data the same way. We break it down into atomic elements like pixels and bits and bytes. Then we deconstruct everything, and then we see what is the thing we want to optimize? Is it for baby image? Then there are certain kind of reconstruction process that make it targeted towards the baby image enhancement, as opposed to space image. So this is a fun process. And we really enjoy it, because every day we see new things, and sometimes we think it is nonsense. Sometimes it is, but some other time we see fantastic things. And we'll slowly reveal those to the public as we are more confident. And it is a scientific work. We need to make sure it is validated. But I'm very fortunate that I have a group of scientific collaborators, like very rigorously. There are PhD work going on on this work. So that way, we are a little bit slow but very steady in our endeavor to present very accurate but mind-boggling data from whatever has been done historically and in the future. I mean, you are right. This is a very exciting time. You talked about not only on the data cleaning aspect, but also on the opportunity that this AI revolution
Chris [00:27:00]:
gives us all, is that we can now focus on more creative processes, focus on other opportunities and problems that we couldn't have the time to solve. Now we do with this AI revolution. So, it's a very exciting time to be alive, that's for sure.
Dr.Pal [00:27:18]:
Absolutely. I mean, AI doesn't sing a song. It can compose a song based on whatever has been made, but that human creativity of making a fantastic piece of music is literally the crux of human genius, right? Or same goes for the Mona Lisa. I think this is where we should be focusing while AI is taking care of some of the regular mundane stuff like turning the lights on and off where we should not be involved. I think, yeah, this is a great moment to help humanity go to the next level, be very, very creative, and do stuff at scale so that individual efforts will not be put into doing the monotonous work where AI will be helping us. Then we can solve much, much larger and bigger problem to have a much stronger and more friendly humanity.
Chris [00:28:05]:
Well said, and with this AI revolution is that the most recent form of it, it has been exponential, right? We use that term a lot, but it truly has been exponential growth. And that's why we are, you know, riding this AI wave hard to let you all know, like it's coming, it's here, and it's going to change the way our world is forever. And that's going to happen now and soon. And looking ahead, talking about that, it's tough to predict exponential growth because the options are limitless. But where do you see the advancements in AI taking us over the next 2 years, 2 to 5 years?
Dr.Pal [00:28:52]:
I think some of the revolution area would be like exploration of space, for example, extremely hostile conditions and stuff like bomb squad and even cleaning, which we should not be spending our time on. So, anywhere under hazardous conditions, like underwater sea diving, for example, I am involved with another conversational project on underwater pipeline monitoring. That's a very hard problem. You have to put submarines and real people to monitor what is happening underneath. This is where I think AI can really marvel, and we need AI for that. Same goes for doing automated sample analysis in Mars so that we can see the signature of life, I think, rather than sending humans who are always at the risk of death. Medicine is actually being revolutionized and it will be. I think 1 of the projects I'm working on with the company, Phil, and my dear friend, Devon, is very deeply involved in the vision that he calls the human operating system, where we as humanity are not utilizing all the faculties that we are blessed with. So in school, we only enhance only limited faculties, and depending on our limited exposure to stuff. But we are in a mission that all of us can enhance our human capabilities, not just the AI capability. AI can help us enhance our human capability to be multidimensional that we are intrinsically, but we just forgot about it. Devon's vision is a human operating system where we, as humans, can operate at a much higher level scale when we can collect data from our brain, from our body, much more reliably using the technology I'm developing so that we can enhance our life at scale that we have never had access to.
Chris [00:30:50]:
1 interesting thing that I've always believed that will happen, and it sounds like you agree with this as well, is that currently we have our biological system, right? These little cells inside of us that are interacting in a system. But soon we'll have an abiological system integrating within us through nanotechnology with our biosystem, right? That's letting us know, oh, your heart pressure is up, your heart rate is up, oh, you know, there's these cells that are forming that could potentially become cancerous, and we'll have that whole system that can also report to our doctors and physicians and surgeons. That vision, is that something that you feel is very plausible and potentially something that could happen within, you know, 05:10 years? Oh yeah, absolutely. In fact, a couple of weeks ago I was at MIT's
Dr.Pal [00:31:42]:
Koch Cancer Institute where the whole topic was about nanobots like observing like potential lung cancer markers, essentially. And some of the problems are there. It's a very wild condition. And then even a little difference may mean cancer, or it may come of other origin. I think Microsoft is also involved with this kind of research and I'm in touch with some of those researchers where the intelligence intrinsically that comes from, I think, human creativity cannot be replaced by AI. AI can train certain stuff. I think 1 of the important areas would be that our human intelligence, how we can put into the AI intelligence so that they can interact in a very collaborative fashion. Right now, it's in its very early phase, it's happening, but in a very low dimensional fashion. But I think because the computational cost and everything is so high that we cannot afford to do that. But in the future, definitely medical diagnosis and intervention will be 1 of the key area so that people can live a full life, happily blossoming. And I think that will be the very impactful area that I can't wait to look forward to.
Chris [00:32:56]:
Dr. Prasantapal, it's been a pleasure chatting with you. I could chat with you for hours. I'm just so intrigued with everything that you have going on, what is going on in your brain. We love having conversations, as you can tell, Dr. Prasanta Pal is involved in many things that's gonna help our lives be better. And what I most love about you is your vision for humanity, and how with the utilization of AI, with technology, with data cleaning, that we can make the world a better place. So I really appreciate you hopping on. But I do want to give you the opportunity to our wonderful audience. Is there anything else that you would want to share that we didn't get a chance to discuss today? Yeah, sure. 1 last thing would be, first off, thank you very much for creating this AI wave, literally wave, particularly around Rhode Island. I mean, Wave is very much needed across everything we do.
Dr.Pal [00:33:50]:
And you're doing an amazing job of bringing in people, creating their awareness and gluing them together for making this more accessible. I think I'm really grateful for that, particularly bringing me here and the opportunity to speak to your wonderful audience. The last thing I would say is, you know, AI is serious business. So people should be very serious about seriously having fun while working on it. Otherwise their brain will explode and they will have, you know, nightmares. So I think while working on this very complicated problem, at least I make sure I have enough fun, I'm really enjoying it, and being surrounded by the right people both personally and professionally, ultimately nothing can replace the human genius. I mean, that's the 1 thing I have learned and I have experienced. So always be with the right people who are always encouraging you, motivating you, and particularly wild ideas like this 1. When I started, if I did not have the wonderful people, like I should mention like Annette Tonti particularly, the director of Rihub, I mean, without her, I could not take things forward like this. And folks from Brown University who really loved the idea and filed a patent and to bring it forward. My academic collaborators from many elite institutions has been supporting me. Again, when things are established, it's very easy to move that forward. But at the early phase, I am saying, hey, I can change all the data. You have to have believers and who are actually supporting you to pursue that. Maybe I was completely wrong or maybe I was right. I think this is where you have to keep your faith and be surrounded by the right people, not necessarily who are always telling you the good thing about you, but always correcting you for the betterment of it. So I am personally grateful for those folks around me, both in my personal and professional circle, for keeping me encouraged and fired up all the time, entertaining me, and that's why that's my... And I would say I have a lot of developers and team members across the globe. I'm really grateful for them because often they are like doing stuff without getting paid a lot because as a startup, we don't have a lot of money, but I think that's the kind of privilege I have had and then I want to make sure that that culture goes on so that you have a bigger vision, you're working for something bigger than yourself.
Chris [00:36:04]:
No, I mean, we are AI powered, but we're human elevated. At the end of the day, it's all about the people that make up our organizations, our technology, and that are with us. You're right, AI can be overwhelming. It's always changing. Every day there's something new. So you have to always ground yourself. I have my one-year-old, Caden. He helps ground me. I come home, he does not care about AI. He just wants to go play with his little toys and trucks, right? And it's good to have that ground and it's good to have great people around you for sure. So that was well said. So I'm sure a lot of people are going to be more intrigued about everything you're doing. What would be the best way to keep in touch with you, our audience, if they want to learn more?
Dr.Pal [00:36:47]:
Are you active on any social channels that you'd want to share with others? Yeah, I wish I had more bandwidth for that. Maybe send me like a LinkedIn invitation. I've been lately really busy. And then on the website also, like I purposely kept it a little bit closed, but maybe we'll open a channel so that people can contact us. I think that would be helpful. Well, send a LinkedIn
Chris [00:37:08]:
invite to Dr. Prasanta Pal. He is a very busy man, so don't expect an immediate reply, but certainly someone you would want to follow, keep in touch with as he does great works in the field of AI and beyond with data cleaning. Thank you, Prasanta Pal. Thank you, Chris, and your team. You have an amazing team, and really grateful for getting this forward. We appreciate it. Thank you. Thank you. And that was our interview with Dr. Prasanta Pal. That was amazing. I learned a lot. I hope you learned a lot too. This is really exciting. It gave me even more hope about the possibilities of AI. Once we let the technology do the mundane tasks, for instance, it opens us up to be more creative and solve these never-ending world problems and just make our world the better place. So that got me fired up. I hope you enjoyed it, too. And if you ever want to learn more and talk to Dr. Prasanta Pali, he said, hey, follow him on LinkedIn. So look him up, send him a connection, tell him that he did a great job on this episode, and he's an open book. He is very excited about this, so definitely someone that you want to follow because he is going to be a leader in this AI revolution. And don't forget we have some great guests coming up so make sure you subscribe to our podcast and follow us on social media at trailblaze.marketing. We'll keep You