AI Brain Shrink Innovation! You Won’t Believe What’s Happening!

 




AI Brain Shrinkage – The End of ChatGPT?

GPT 4.5 and AI Innovation: Learning Methods, Economic Costs, and the Secret of Chain of Thought

1. AI Learning Evolution and the Integration of Human Education

Have you ever wondered why AI responses are much safer and friendlier than you expected?
The reason they’ve become five to ten times smarter than before is because many people have been ‘teaching’ AI.
As mentioned by the CEO, AI refines its responses by receiving human feedback not only through pre-training (preliminary mass learning) but also through post-training (follow-up guidance).
In this process, it also learns to avoid political, criminal, and sensitive topics.
Ultimately, AI can think like humans and provide emotional and safe responses.

Another point to note is that we’ve experienced people directly correcting and selecting the right answers to the question, “How can I study effectively?” for AI.
This process is called alignment and post-training scaling.
In other words, AI iterates through sequential thinking (chain of thought) with human collaboration, providing more accurate and rational answers.

2. GPT 4.5 and the Problem of Economic Costs

With the introduction of GPT 4.5, it’s worth noting that the intelligence of artificial intelligence has increased, as well as the learning and usage costs have increased significantly.
While the pro model accessed by general users costs around $20 per month,
the price per token for companies using the API method is 70 times more expensive than before.
In other words, significant investment and cost are required for advanced AI services.
The key point here is that rather than simply inputting a lot of data, the focus is on how to make it ‘smarter’ rather than ‘bigger’ to maximize cost-effectiveness.
To overcome economic limitations, AI developers are not infinitely increasing the size of models, but instead,
by introducing efficient learning and collaboration methods (e.g., chain of thought, distillation), they are trying to
reduce costs while maintaining or even improving performance.

3. The Core of Post-Training Scaling and Chain of Thought

Post-training scaling refers to the process of AI quickly supplementing itself by receiving human feedback after pre-learning.
For example, AI is made to produce answers through multiple thoughts (chain of thought),
gradually increasing the flow and accuracy of the answers.
This process is similar to discussing and reviewing a complex problem with a friend,
and reaching a final conclusion after several reviews.
In the chain of thought method,
① the question is broken down,
② step-by-step calculations and logical thinking are performed, and
③ the final answer is derived.
This sequential thinking method is the secret to how AI accurately solves complex problems that are sometimes difficult to solve with one mental calculation.
The ‘problem of 12 people calculating for 18 servings of food’ that the CEO exemplified is also
designed to allow AI to provide accurate answers through sequential thinking using this method.

4. The Future of AI Development and Global Economic Outlook

Currently, the development of AI technology goes beyond simply increasing the size of the model,
and is evolving into a method of collaborating with humans (agent collaboration model).
At the same time, generative AI is being applied to various industries such as digital marketing, customer service, healthcare, and self-driving,
and is heralding innovation throughout the global economy.
From an economic perspective, AI technology and innovative solutions increase productivity, reduce costs, and
It is expected to create an environment where companies can create new added value.
In particular, it is necessary to pay attention to how the latest AI technology and advanced learning methods such as chain of thought will be applied in the future economy.

< Summary >
• AI is generating increasingly safe and friendly answers through pre-training and post-training.
• GPT 4.5 is five to ten times smarter than previous models, but the economic cost is significantly higher.
• The chain of thought method improves AI’s ability to solve complex problems through step-by-step thinking.
• The AI collaboration model and agent development herald innovative changes throughout the global economy.
• These latest AI technologies are intertwined with economic, global, AI, technology, and innovation keywords, and the future outlook is bright.

[Related Articles…]
GPT Innovation Outlook
AI Economic Effect Analysis

*YouTube Source: [와이스트릿 – 지식과 자산의 복리효과]

– “챗GPT 이게 마지막입니다” AI가 뇌를 줄이기 시작했습니다, 이게 큰 겁니다 / 이경일 대표 (1부)




Yield-Chasing ETF

Covered Call ETF Investment Strategy Summary

If you’re interested in covered call ETFs, I’ve prepared some essential information that’s worth reading right away.
In this article, I’ve thoroughly organized the essential content you need to know about investment, dividends, stock prices, ETFs, and covered calls, categorized by period and item.

1. Basic Concepts and Key Products of Covered Call ETFs

A covered call ETF is an investment strategy that receives dividends from stock price gains and option premiums of the underlying assets.
Representative products include JP Morgan’s JP, JPQ, and Mirae Asset Global Investments’ QYLD, XYLD.
Listed by net asset size, all four are noteworthy products, and the annual dividend rates for each ETF vary, with JP at 7.36%, JPQ at 10.44%, QYLD at 12.99%, and XYLD at 11.98%.
Note that a high dividend rate is not always good. The balance between stock price gains and dividend yield varies depending on the product structure.

2. Analysis of Annual Dividend Rate and Total Return

The attractiveness of ETFs is revealed not only in the simple dividend rate but also in the total return from stock price appreciation.
Compared to SCHD ETFs, which are popular among dividend investors, covered call ETFs also have high dividend rates, but stock price appreciation may be limited.
In terms of total return, JP ETFs show overwhelmingly good performance, and JPQ can enjoy more stock price gains, although the dividend is somewhat lower.
In other words, when investing, you should not only look at simple dividends but also comprehensively consider the total return that is reinvested along with stock price movements.

3. Comparison of NASDAQ 100 and S&P 500 Based ETFs

QQQ ETF is an ETF that tracks the NASDAQ 100 index and shows the highest return when stock prices rise.
In comparison, QYLD and JPQ, which use a covered call strategy, show defensive power in a declining market, but in an upward market, a portion of the price increase is replaced by option premiums.
The S&P 500-based ETFs IVV and XYLD also show similar characteristics, with a defensive effect in a falling market, but the gap may widen in a rising market.
The right ETF selection depends on market conditions, and it is necessary to recognize the structural limitations of each ETF.

4. Selecting the Right Covered Call ETF for Each Market Situation

QYLD and XYLD are attractive in a sideways market because dividend income plays an important role.
In a bull market, JPQ and JP are advantageous for enjoying stock price gains, but it is difficult to enjoy the full rise due to the nature of the option premium strategy during a sharp rise.
Covered call ETFs, which play a defensive role in a bear market, are somewhat effective, but you cannot avoid the impact of a sharp decline.
Therefore, you should choose an ETF that fits your investment preferences and cash flow needs.

5. Portfolio Composition and Investment Simulation Strategies

It is better to focus on short-term cash flow management rather than a long-term investment perspective.
As an example of portfolio composition, a strategy of mixing S&P 500, NASDAQ 100 ETFs, and covered call ETFs is recommended.
For example, in the existing strategy of reflecting 15% of NASDAQ 100 and 70% of SCHD dividend and Dow Jones-related ETFs, you can add covered call ETFs such as JPQ to balance cash flow and stock price appreciation.
Simulation results show that as the proportion of covered call ETFs increases, the monthly dividend increases, but the total return tends to decrease slightly.
Therefore, it is important to clearly consider your cash flow needs and investment period when investing, and to allocate the proportion of covered call ETFs in your portfolio appropriately.

6. Precautions for Investment and Final Conclusion

A cautionary note when investing in covered call ETFs is that you should not make investment decisions based solely on the dividend rate.
Even if there are products with dividend rates of 200% or more, it may be difficult to enjoy the desired stock price increase effect due to high volatility and limitations of the option strategy.
In addition, a long-term investment strategy of putting all your funds into covered call ETFs is not suitable.
It is best to adjust the proportion in your portfolio according to the time when cash flow is needed and the amount needed.
In summary,
– Choose QYLD, XYLD if you prioritize cash flow
– JP, JPQ recommended if you value stock price gains
– You need to build a balanced portfolio if you want to enjoy both moderately
– You should also consider the difference in investment fees (e.g., 0.6% for QYLD, XYLD vs. 0.35% for JP, JPQ).

< Summary >
Summary of key points related to investment, dividends, covered calls, ETFs, and stock prices.
Covered call ETFs are an investment strategy that aims to generate both dividend and premium income,
Dividend rates and total return vary by product, including JP, JPQ, QYLD, and XYLD.
The appropriate ETF selection depends on the market situation—sideways, rising, falling—
It is important to manage cash flow and stock price gains in a balanced manner by adjusting the proportion in the portfolio.

[Related Articles: Covered Call ETF Investment Strategy, Portfolio Composition Method]

*YouTube Source: [이효석아카데미]

– 커버드콜 ETF, 하나만 고른다면 저는 이것입니다! #한나 #etf #커버드콜

  ● AI Brain Shrinkage – The End of ChatGPT? GPT 4.5 and AI Innovation: Learning Methods, Economic Costs, and the Secret of Chain of Thought 1. AI Learning Evolution and the Integration of Human Education Have you ever wondered why AI responses are much safer and friendlier than you expected? The reason they’ve become…

Leave a Reply

Your email address will not be published. Required fields are marked *

Feature is an online magazine made by culture lovers. We offer weekly reflections, reviews, and news on art, literature, and music.

Please subscribe to our newsletter to let us know whenever we publish new content. We send no spam, and you can unsubscribe at any time.