NVIDIA has been receiving a lot of attention recently. What NVIDIA has been able to do is to create a good ecosystem. For example, they provided engineers with free distribution of things like CUDA. As a result, people continue to use it, and they have no choice but to stay in that ecosystem. In that sense, NVIDIA is number one. If you listen to what they say, using AI technology, some new services are emerging in the world. It is a picture of how the services provided by Microsoft, Google, or Apple are greatly improved by meeting this AI technology. That's how you can understand it, right? Yes, that's correct. And also, there is something else to consider. I haven't talked much about semiconductors, but semiconductors are important.

Because fundamentally, if you look at it, as semiconductors improve in performance, artificial intelligence can only improve in performance. But if you look at it now, big tech companies are all into semiconductors. Not only artificial intelligence semiconductors but also going into servers and various things. They are all into semiconductors now. And there is a reason why semiconductors are important. If you look at not too long ago, we experienced a lot of difficulties due to the lack of semiconductors. Therefore, there are uncertainties in the supply chain, and such factors. Also, similarly, you need semiconductors to maximize the performance of the product. Optimization is important. Semiconductors, and then working together with devices. So now, because this optimization is possible, we have no choice but to do semiconductors. And next, what we need to consider is that big tech companies also need timely supply of semiconductors to launch their products on time.

Yes, if you look back, Intel used to be very strong, but due to semiconductor issues, they couldn't supply Apple and others on time, which made Apple very upset because their products couldn't come out properly. So, that's what they tried to do. Another thing to consider is that there are also such issues, and you can't ignore the price aspect. The price difference is huge between receiving and making it. It's significantly different. So, in terms of price competition, it can be quite advantageous. In fact, Nvidia's gross profit margin is about 78%, but if they make it themselves, big tech companies can expect to save costs significantly. Since we talked about semiconductors, I would like to talk a little more about Nvidia. Ultimately, we will continue.

You keep emphasizing the part where advancements in AI technology are backed by improvements in hardware performance. I would like to talk a bit more about NVIDIA. NVIDIA has been receiving a lot of attention recently. Everyone knows about it. It's like, if someone doesn't know NVIDIA, they're considered a spy. Everyone knows about it. NVIDIA has now gone beyond just NVIDIA, and new terms like God NVIDIA are emerging. Recently, if you look at their stock prices and performance, they have been extremely good, right? So, they are receiving a lot of attention, and their market capitalization is expected to surpass big tech companies soon. They are in a position where they are chasing after them. Yes, they are closely chasing after them. So, there are high expectations for the future. However, there is a reason why they can do this. Firstly, they were able to establish an ecosystem in the MBA university.

I made it, so when you look at it now, in the case of NVIDIA, the ecosystem strategy is like this: at first, it doesn't make money, but now, in the beginning, it creates an ecosystem for a certain industry to grow, enabling it to grow. Then, it pours a huge amount of money to create an ecosystem that can grow. For example, in the case of CUDA, they distributed it all for free to engineers. So, for engineers, being able to use it for free is a huge benefit. Also, the performance is quite good, so they end up using it continuously. They have no choice but to stay in that ecosystem. They have no choice but to use NVIDIA chips. It's a situation where they have no choice but to use it. If you look now, Kakao is the same. It's the same transition, but at first, everything is given out for free, so I don't want to use it either.

It happened like that, and I had no choice but to use it. People are using it, so I can't avoid it. That's why it has become the current situation. The rain is the same. As I continue, now I can't get out of it. So, they used a strategy to confine. A locking strategy, for example. By doing this, they have made it impossible for customers to escape. So, in my opinion, this ecosystem is likely to go a certain way for a few years. Of course, many companies are developing semiconductors and MPO now. However, it's not just about developing semiconductors. So, software, then ecosystem, services, everything should be provided collectively. In that sense, NVIDIA is number one. So, simply comparing the performance of hardware, whether it's AMD or something else.

There may be some room for follow-up on things like this, and if a big tech company develops its own chip, that could also be another potential risk factor. However, the ecosystem that NVIDIA has built is much more influential than any hardware competitiveness. I think this is a good summary. I have a feeling that this will continue for several years to come. Of course, the stock price may fluctuate, but when it comes to AI semiconductors, NVIDIA is expected to maintain a dominant competitive position for the time being. Considering the ecosystem that NVIDIA has built, it was mentioned that any in-house chip may not have a significant impact in the short term. Nevertheless, big tech companies are making a lot of efforts in developing their own chips. How is the current situation of big tech companies in developing their own chips?

Could you tell us a little more about this? As you can see, many Korean companies are developing their own chips. I mentioned the reasons as well. Therefore, taking these semiconductor chips is very helpful for the competitiveness of big tech companies. They are developing everything on their own for a long time. If you look at Google, they have a PPU, a processing unit. They are developing these chips now, and new chips keep coming out. There is a chip called V5P, which is a chip that trains interestingly. Also, if you look at Xc1 CPU, it has been released. Next, Google is also making chips for their phones. Furthermore, Microsoft is doing the same. They have AI GPUs and in the cloud, Cobalt 100 was recently announced. Amazon is also doing the same.

In the case of Infineon, they produce inference chips, and then they also make chips like the Graviton or the Landonji. Although it's not mentioned here, if you look, the company with the strongest chips is Apple. Apple almost started developing chips from the very beginning, and now, from the M1 chip to the upcoming M4 chip, they are leading in chips. So, the M4 chip is a very top-notch chip. Still, they continue to develop further, don't they? When it comes to the A-series, Apple is solidifying its strength in the semiconductor field. By developing their own chips like this, if the utilization of these in-house chips increases, even if Nvidia has built a robust ecosystem, it could still be a significant threat. However, while one may think so, it's not necessarily the case. Because the strategy of big tech companies is fundamentally to sell to others.

We are not making it to create our services or products now to enhance their performance, so that's not competitive. Of course, it could reduce demand now. It's about making our own chips, not for selling. Of course, there are some sales talks now, but the basic strategy is to make our services and products better. Ultimately, in terms of overall sales, increasing the market share from other entities that do not have this BigTable is actually a much better strategy. Also, if you look at it, there are more companies designing these chips. If you look now, not only automotive companies, but also IT companies are making chips. For example, they announced that they will open AI. However, this kind of demand is huge.

But now, it's not about competing with Dia, but rather about building their own homes to live in. Of course, Chairman Son Jeong-hee is talking about replacing the Vidia chip, but it's not that easy. Earlier, he mentioned that it's about replacing their own demand, not about selling chips to others. It doesn't involve competition. That's what he seemed to say. Next, I looked at the number of AI-related employees by big tech companies. This data was released just a few days ago. The data is from a company called Glass AI, which is a research company specializing in AI. So, if you look, Amazon has the most employees, with over 10,000. That's why the number is so high. But in reality, in terms of ratio, other big tech companies may have a higher number of personnel.

Because there are many, that's right. Next, if you look at the second place, now there are quite a lot in terms of Microsoft, over 7,000 people. Next, including OpenAI, then Meta, then Google, and Apple. Unfortunately, as I mentioned earlier, there was a slight lag in artificial intelligence, so I think maybe it happened due to investment. I think about it now. Although it doesn't show the growth rate of the number of employees, just looking at the absolute numbers, isn't it a good indicator of how much investment these big tech companies are making in AI? Also, in fact, Silicon Valley is where the most talented people in the world gather. Even in a place where talented people gather, the hottest talent is in AI experts. In fact, it may not be the most comprehensive, but it shows how these big tech companies are looking at AI in a somewhat simplistic way.

I don't think there are many people doubting AI now, and if you are considering investing, you should probably keep an eye on the theme of AI. Do you also invest, professor? Yes, I do. Well then, from an investment perspective, how do you think about AI? What promising technologies are there? We hear a lot of stories, and if someone asks how to approach AI from an investment perspective, could you possibly provide an answer? AI is quite good. Yes, overall, it will continue to be good in the future. It seems like we can't stop this trend of artificial intelligence. It will continue to grow in the future, and there is a possibility that the growth rate will increase. The same goes for semiconductors. Yes, most semiconductors in the future, not all, but about 1%, will be AI semiconductors. In this way, the market will grow enormously, and also.

Similarly, big tech companies will also make huge investments now and in the future. Because now, I can't invest alone, I have to follow this trend. If others choose first, there will be a huge ripple effect. So, many companies are likely to continue investing. In my opinion, it is necessary to pay close attention to companies related to artificial intelligence. Not only semiconductors but also big tech companies. Riding this trend seems to be a good idea. I remember that when smartphones first appeared in the world, around 2012-2013, stocks related to smartphones received significant attention in the market. But since then, there has been a transition from PC-based to mobile ecosystem.

In the evolving mobile ecosystem, companies competing are attracting attention in turns, and in fact, it seems that this period of around 10 years has continued. So, what lies ahead for AI? Of course, it's difficult to determine the exact period now, but rather than 1 or 2 years, from an investment perspective, it is important to continuously seek out stocks that can benefit from the growth of the AI industry for a considerable period. Yes, that's correct. Because AI is just beginning now. And the same goes for semiconductors. As I mentioned earlier, we are only up to the 2nd generation, and we haven't even reached the 3rd generation yet. We are still progressing. You briefly mentioned MPU, which may still sound unfamiliar to many people. Beyond GPUs, moving on to MPUs and concepts related to AI that we are not familiar with will eventually become more familiar to us. Yes, that's right. So, what is an MPU? It is now mimicking some of the human nerves.

Similarly, now, semiconductors developed for artificial intelligence, such as GPUs, are not developed for artificial intelligence, but for graphic processing devices. Therefore, the power consumption is enormous, and there are many unnecessary functions. However, MPUs, which are developed for artificial intelligence, are suitable. Also, what is important is power efficiency. The power consumption is quite low. However, the current development has reached a certain level, but future semiconductors will not be the same. What will come out is not exactly the same as humans, but similar semiconductors will emerge. It is uncertain what exactly it will be, so what it means is that neuromorphic semiconductors, if you look at the human brain, neurons and synapses are connected.

In a similar way to the human body structure, semiconductor devices that function similarly are likely to continue to emerge. Many companies are currently developing these semiconductors. There has been some progress, and if you look back, companies like Samsung are now incorporating processing and memory functions into semiconductor chips. In the case of semiconductor chips, one chip must handle computation, control, and memory functions all at once. For example, GPUs now come with HBM, causing various issues because they need to communicate with each other, leading to power consumption or bottleneck problems. Therefore, the future of semiconductors lies in chips that can perform computation and memory functions together in one chip. This is important because, for example, in autonomous vehicles.

If you look at it, the reason why autonomous driving cars are challenging is that when a car is driving and suddenly something or someone jumps in, it needs to react immediately and stop the car. But if the semiconductor performance is slow, what happens? It could lead to accidents. However, with the neuromorphic semiconductor coming in, it will recognize instantly and react right away. At the same time, almost simultaneously, accidents can be prevented. It is much more advanced for AI than now. Therefore, improving the performance of semiconductors is crucial every day. Yes, that's right. So, just like I mentioned about robots earlier, the same goes for robots. The response speed needs to be fast. It has to handle computation, memory, control, and everything at the same time. That's why the neuromorphic semiconductor is the answer for those robots.

Only then can the real age of AI open up. When you talk about it, semiconductors are actually now classified as the AI sector in the stock market. But when you listen to what you said, shouldn't we now classify them more as essential consumer goods? I just wanted to throw that thought out there. Today, especially regarding AI, I am looking into the competition in big tech related to AI and also what the future looks like. Our ACE ETF is actually a sincere operator in big tech and semiconductors. Our ACE ETF has newly listed a total of four items in the ACE Big Tech Value Chain Active Series. If what you said is the beginning of AI, then the stock market going forward can expect to benefit from the growth of the AI industry.

You may feel comfortable looking for stocks in the future, and for those investors interested in AI, it's essential to pay attention to ETFs. Let me briefly introduce a few ETFs. NVIDIA, Microsoft, Google, Apple, these are companies that are leading in AI competition. As these companies grow, it is expected that related value chain stocks will also grow. We have newly listed four ETFs focusing on concentrated investments in these companies. In summary, when we look at the growth process of the AI industry, we can see that certain leading companies with dominant competitiveness are driving the industry. They possess existing competitive advantages, established ecosystems, and substantial financial resources.

With such financial strength pouring into continuous technological development in the big tech AI ecosystem, I believe the influence it exerts will only become stronger in the future. Therefore, from an investment perspective, if I were to share my opinion a bit on investing in AI businesses, rather than investing broadly, focusing a bit more on leading companies and value chain companies that can support the growth of these top companies might be a slightly better approach. In fact, based on this judgment, we recently listed the Ace Big Tech Value Chain Active Series. Many people ask me how to invest in AI, and the most common methods are typically investing in the Nasdaq 100 or semiconductor ETFs, big tech ETFs, which are somewhat related to tech and growth stocks.

Investing in high-weighted ETFs can partially expect benefits from the growth of the AI industry. By incorporating existing technology ETFs into our Ace Big Tech Valuation Active Series, you can create a portfolio that anticipates benefits from industrial growth. I mentioned ETFs a bit earlier. Lastly, the professor has given us a lot of valuable advice. If you have any additional words you would like to share with our viewers, I always think that it is good to invest with a long-term perspective when investing in stocks. This is because stocks are greatly influenced by external factors. Even Samsung has dropped significantly, so you never know when these variables will arise.

Whenever events like the pandemic or other disasters occur, stocks are affected. However, in my view, such variables do not last for more than a year. So, in my opinion, companies that fundamentally grow, increase their sales, and profit margins will inevitably reflect in their stocks. If you look at a company that can grow for about three years, especially companies related to AI or those expected to grow in the future, investing in such companies for about three years for long-term growth is advisable. The stock market, in fact, fluctuates daily, but instead of reacting to these changes in the stock market on a daily basis, if you invest in industries with a clear growth potential for the next three to five years, you are likely to achieve excellent investment results.

It seems like you've done it. Today, we talked with Professor Kwon Yong-hwa about the competitiveness and prospects of Microsoft, Google, Apple, and NVIDIA in the AI industry. Thank you for your insightful words today, Professor Kwon Yong-hwa. I will greet you again next time. Thank you. Thank you.

다른거 살 필요 없이 이 '4개 주식' 사놓고 기다리면 큰 부가 될 겁니다. (권영화 교수 2부)

경제원탑은 우리 삶에 꼭 필요한 경제를 알려드리기 위해 탄생했습니다.
주식, 부동산, 경제전망 등 늘 어렵게만 느껴지던 경제상식들을 쉽게 전달드리겠습니다.

👉경제원탑의 구독자가 되어주세요
👉https://url.kr/ugmj59

※모든 투자에 대한 책임은 투자자 본인에 있으며, 
본 방송 정보는 참고 사항일 뿐입니다.

※현재 어떠한 커뮤니티 또는 리딩방을 운영하고 있지 않습니다. 
사칭에 주의하시길 바랍니다.
