Yesterday Today Tomorrow

Guest (Male): Yesterday, connecting past. Today, with an outer view. Tomorrow, to understand future. Yesterday, today, and tomorrow is the program that covers the current contemporary social issues in the light of our history. To understand our yesterday, to live fully today and tomorrow. Yesterday, today, tomorrow with In Sung Kim.

Inseong J Kim: Hello, this is In Sung Kim from Yesterday, Today, and Tomorrow. I have to tell, I have to share with our audiences how beautiful Phoenix, Arizona is. I mean Arizona. It's so beautiful. Every year we forget how beautiful it is. And I'm not encouraging everybody to move into Arizona because it's already saturated. But I have to be honest, it is so beautiful. Weather is amazing. So we are enjoying every minute of it. Every morning walk is incredible. So I'm going to finish my uh program with AI. And I want to start with a scripture. Said there is nothing new under the sun is in the Ecclesiastes in chapter 1 verse 9 and 10.

So if anyone is not a Christians who is listening to this program and go back and open the Bible and see there is the world world said there is nothing new under the sun and what that means. So we all believe we're looking for new things and AI is a new things and we believe the AI will save us financially, uh intellectually. That's what we're looking for. But is it going to save us? That's something to really uh think about. And but I'm be honest to be with you that AI is incredible. I'm using it.

And and so I'm going to talk about positive side as long as we use it as a tool, but knowing what it is is important. That's what I want to share. So artificial intelligence feels brand new. But the questions it raises are very old. Who can I trust? Is this person selling me something? Is this medicine safe? Who knows my secret? And people ask those questions in marketplaces in 3 thousand years ago, even many years ago. As it I think is as human existed after the fall. Before the fall, probably we didn't have to ask these questions, but after the fall uh we asked these questions.

Today uh we asked them of AI. We asked to AI. And I share a little bit about how ancient ancient um study that they've been accumulated over years in in China in Asia and using that book called that's kind of uh it's a basically it's a data science in in China. They've been using that. It's only trans formed to a machine today and using mathematics to find uh the solutions or answers on the data. So why this verse uh is really fits and this tool AI is brand new, but human nature is not. And a persuasion, trust, privacy are ancient problems as well.

So today uh we look at where the AI genuinely helps and where the old trap are simply wearing a new clothes clothes. So that's what we see even um even in the election that we experienced, we don't watch the main media for debating or uh constructive discussion what they are plans for to uh when they uh have actually job uh as job descriptions. It does we're we're changing everything right now. So we don't need so much campaign either. It's just quiet suddenly. People make a decision from mostly YouTube. And uh and then soon it's going to be AI uh more asking questions about who that person want to run the office and want to know their backgrounds.

So it's not about someone who is actually running. It's about information surround that person. Either includes truth and a lie altogether. And so we hope that this social media revolutionize of the main media that we thought that it's a propaganda all those years yet. I don't know personally it's worse than main media to me that AI bots dominating the narrative of the story of individual who's running for the office at the same time. Although human even history that today we're experiencing it's a very interesting time. The bottom line is people are believing the lie. That is the interesting part.

So like again going all the back go back to ancient history. It's in the Bible. People people's been believing from the day one the lie to all the way today to lie. And so that's the negative part of the AI that we have to be careful that using in our time. But there's a a lot of positive part that use AI as a just tool. So as long as human beings stay same and nothing will truly change. Every tool, every kind of knowledge can be used for good or for evil. So AI is not making decision.

Human who believe lie are making a decision. We have to make it clear about that because we always thinking AI is AI is going to destroy us. No, human are we doing we're doing to ourselves. So we thought the computers will solve the problems and so they solved a lot of problems. But the people who got to the internet first won the race as far as the market goes all gathered into the fewer hands because this mass information of the selling through the internet and it allow to acquire the wealth. And as along with that whoever supplies faster was winner.

And that created today's problems, but at the time we didn't know. So it's a mechanical decisions that how computer and then AI developed and along with that economic decision has followed. So where it's going to head is I think majority of us already know what is going to happen. So we're living a very very interesting time. So we uh we all believe the social media is amazing things that revolutionize of human problems that bias or gap of rich and poor or telling the truth, authentic everything. But look at what TikTok do these days TikTok does today.

It just make a mess over the information. So again, who's using the tool of this machine is important. So can we AI solve the human problem? Only when it used it properly in ethic minded and by good people with a good intention. So I'm genuinely hopeful about the AI in some way as a tool. But there is other area that we have to be very cautious about when AI was used by evil intent. What it does it look like? So we left off our program last week about the market research that how good uh it is that we can use this AI.

So the big company AI put the big company research within the reach uh of the small businesses. So you can gather public information on competitors and prices sort of the hundreds of customer comments into the themes 70 per 7% of the marketing professional already used ChatGPT or similar tool. And also a US small business use more than doubled from 23% in 2023 and 58% in 2025 among small business using AI. 85% reported higher sales and 85 4% higher profits. So people who knows AI are winner in our time as far as sale goes. So Unilever Unilever says AI made its customer inside analysis about 60% faster scanning over thousand outside the data source each month.

And the time from the product idea to the research brief brief fell from the months to days and product test uh rounds dropped it from five or six to one or two. So these are the companies own figures. So their productivity uh the time actually was saved. And a Harvard Business School and Microsoft researchers had AI play thousands of the toothpaste shoppers. A real consumer study value the fluoride in toothpaste at the $3.27. The AI estimated three $3.40. And traditionally study cost $20,000 and take three to six months. AI took under 15 minutes. And that's the reality of today.

So the researchers see it at first the screen before talking to real customers and then not a replacement. So uh these are practical tool that people can use it in a positive side. But at the same time, AI can't replace real customers, cannot replace the real customers. Vendors claim systematic responses match real survey the panel 80 to 90% of the time at the 10th of the cost. So that's the reality. And so uh false ad can come back as a truth. Nothing new under the sun that we shared. And 1917 the famous snake oil was tested and found to be mostly mineral oil and beef fat, chili pepper and turpentine.

That was 1917. The seller was find the $20 at the time. Today AI read the ads along with everything else. So a claim repeated on enough website can come out sounding like a neutral research. But it can be biased because ads are now inside AI tool. ChatGPT began showing the ads in February 2026. OpenAI says quote ads do not influence the answers of ChatGPT gives you unquote. But the promise is worth stating than holding them too. So whoever spends money put the more articles um the more information the inputs and comes out as output to AI answer those questions.

So keep in mind it's not a magic, it's not a mystery. The more data that um AI has it will persuade by of those datas. So whoever putting more datas wins basically. And so in health issue that a lot of people trying to use AI for uh the health and some say we're not going to have a doctor anymore. It is going to be true. So part let's talk about positive side. AI help us about understanding uh the more than ever before. It explains us. It can turn medicine leaflet into the plain words. You can keep asking until you get it.

The best results comes with a an expert in charge and Sweden AI helping radiology uh radiologist across the 80,000 women found out 20 more breast cancer 20 % more breast cancers with no rise in false alarm and 44 % less reading work. In the UK, the way to hear the cancer diagnosis fell from 14 days to three. So there is a lot of positive side about what's happening with AI with health. Saving lives from the sepsis. Johns Hopkins used AI early warning system across the five hospitals more than 760,000 patients visit is spotted a sepsis nearly 2 hours sooner on average and death fell about 18 %.

And we can just go on and on. In a study across the two uh 205 UK GP practices doctors using a AI uh uh steps scope uh steps to scope were 2.3 times more likely to detect the heart failure. And 3 % three 3.5 times more likely to detect irregular heartbeat um and 1.9 times more likely detect heart valves disease in patients if legs still needs uh confirming test. So these are really good tools. And a medicine diagnosed uh with AI. And uh rental sertib a drug for the lung disease uh idiopathic a pulmonary uh fibrosis was discovered and designed with AI.

In a 12 weeks trial of 71 patients lung capacity improved by uh 98 mm um ml uh on the height highest to dose, but fell by 20 mm on placebo. Cuz there is trial is small, but larger trial is still needed. And it's still experimenting actually AI. Gather all the information and in inventing and creating the medicine as well. Question is uh we're prolong human life. In Korea, there I got some statistic about how many people are over 100 years old. And number is staggering. Yet baby is not being born. And so this solution for using AI to prolong the human life at the same time.

How do you balance of this is social cultural demography of of economic sense and also uh reality of who's taking care of who. I mean these are the question that we also factor into using this AI. And so we cannot just go to one direction without thinking about the other directions. What are we doing with this information? And are we choosing to really literally choosing to make what age of group like baby boomer or the 50 or 40 are they going to leave more than hundred and then forever and then we're not going to have a children anymore.

What are we doing here? And these are just something that uh scientists have to really factor into about issue not just for rich people want to live longer because they they had a good life then they want to live longer. And using all these tool to prolong their lives and not thinking about rest of us or are we thinking about human in general of improving the lives. Okay, then when we do that, why are we why Canada has number of growing number of euthanasia uh kind of mandatory euthanasia. So what is that mean? Which data are we following through?

And again uh human intellect has to be intervene in data uh science um time right now because AI can create such a mass data with information that can be very attractive. But without thinking overall picture of who we are, what are we doing to our society? What is the consequence of what we do today and it affect tomorrow? And what we do here is affecting tomorrow's life. So we saw what is happening with elections and what is happening with not just this country, but other countries. And I think we're the most I think a lot of most Americans are probably um least um opened for world media what is happening all over the world.

And we hear very biased messages. So I think understanding what is going on in the whole world and how do we apply to our time and fear driven. I mean literally YouTube is a fear driven media. Uh so I think what we need is uh sanity in our time to really uh walk away from it in time and just think about. And just alone with God. What is happening? And what do I supposed to do? And what is the basically truth like I shared in the beginning there is nothing new under the sun than is Bible relevant to our lives today.

That's a something we have to um process. So because the AI has the caution here, we're really living in the AI time. And it's very vibrant exciting time at the same time very interesting time. It's much different than all human history. So AI can be confidently wrong. These are these are the caution. 2023 test on 39 real medicine questions only about 10 answers were satisfactory a 10 answers. And one dose was off by the factor of thousand and some references were invented. And the lead a pharmacist warned quote healthcare professionals and patients should be cautious about using ChatGPT as authoritative source of the medication uh relate to information unquote.

By saying that I think AI company should be really be careful about legal ramifications because people are asking all kinds of questions especially this um legal or medical or psychological counseling without being a professional that asking questions and then later on um are they ready? Some I think hearing talk about do you have a insurance for this catastrophic event not just company wise or nation wise or individually. And with that saying that is the ethical mind of building AI carefully is important. So accuracy there's a real data and clean data and a real data risk privacy and that's reality.

And there may be a way to have both. That's something we're trying to figure it out. And uh federate learning sends AI to the data and not data to the AI. And there's a mixed information about how we use AI is need some intelligence and also study and we need a basic foundation to reflect. So thank you for listening Yesterday, Today, and Tomorrow. We'll be back next week. Thank you.

Guest (Male): You've been listening to Yesterday, Today, and Tomorrow with In Sung Kim. You can also find more from In Sung Kim at insungkim.org. That's i n s e o n g k i m dot org. Thank you for listening to the show.

Guest (Male): GLP ones are changing weight loss.

Guest (Female): My sister had amazing.

Guest (Male): The ideal treatment is

Guest (Male): Seems like everyone's a GLP one expert these days, but when it comes to your health, it pays to listen to someone who actually is. Orderly meds makes it easy to get started with personalized support along the way. And with your first month's order, we're giving you the second month free. Visit orderlymeds.com. Bringing orders to the chaos of GLP ones.

Guest (Male): Individual results may vary. Eligibility and prescription required, not medical advice. Visit orderlymeds.com for details. Compounded medications are not FDA approved.