Trump has reached a consensus with NATO Secretary General Jens Stoltenberg on the framework for a security agreement in the Arctic region:
The aim is to strengthen the security defense capabilities of the Arctic region, primarily to counter any potential threats from Russia or China.
Related plans include deploying U.S. missiles in the area, implementing control over mining rights (explicitly aimed at excluding Chinese interests), and increasing NATO troops (to enhance NATO's military presence in the region).
Additionally, the agreement does not involve Greenland's sovereignty and does not include any territorial concessions.
Alright, regarding the previous threat to impose tariffs on the eight European countries due to the Greenland issue, it has now been decided that no tariffs will be imposed.
This is because, after meeting with the NATO Secretary General, both parties reached a framework agreement on Greenland.
(However, what exactly are the specific details of this agreement? I look forward to further detailed explanations.)
【Chatting: AI and Text Pollution】Since the emergence of ChatGPT, I have been exploring the topic of "text pollution".
The human-generated corpus currently used to train large language models (LLMs) is expected to face depletion by 2027—if not sooner. This means that future sources of corpus will mainly be generated by AI or products of human-AI co-creation.
Today on the X platform, there is an increasing amount of content written by AI that lacks "soul." For humans eager to enjoy the pleasure of reading, this is not only a torment but also a regression. (It should be noted here: I am not criticizing the act of using AI itself, but rather the soulless way of using it that relinquishes human rights to think actively, as if being pulled and enslaved by tools.)
Overall, for those who truly hope to gain information, insights, or simply want to have fun, they may adopt a "defensive" reading strategy, significantly reducing the click and open rates for "folded content" (i.e., long texts exceeding 280 characters).
For those trying to gain traffic through content creation, the situation has become even more severe; especially for creators without a core focus area or lacking professional depth, content production will turn into a highly challenging task.
In short, the large-scale application of LLMs has triggered text pollution. On one hand, this leads readers to become increasingly picky and defensive; on the other hand, it prompts content creators to begin frantically "involution," which in turn uses AI to create more junk information.
Because of this, traditional rigorous news production models, creative methods that can build genuine communication links between people, deeply investigative journalism, and on-the-ground reporting modes... these content creation models will have the opportunity to recapture people's scarce "attention."
If this is seen as a forward-looking study of the industry's profit model, or an exploration to uncover investment targets and create Alpha returns, I will refocus on those media companies that adhere to different creative models—everyone, you must hold on!
Recent events happening around the world have given me a deeper understanding of the term 'self-castration.' It turns out that some people really do make such a choice.
Today's capital market landscape can be described as 'rivers of blood'.
The S&P 500 index has given back all the gains since 2026.
The market is experiencing violent fluctuations: the VIX fear index has surged sharply, the dollar index has plummeted, and risk assets led by cryptocurrencies have faced a comprehensive crash.
Meanwhile, US Treasury yields have skyrocketed. Just as the Davos Forum is taking place, discussions worldwide about 'selling dollar assets' are resurfacing.
Well, let's take a look at the specific distribution of US Treasury holders 👇 Europe holds as much as 3.6 trillion, accounting for 40% of the total amount held by foreign holders of US Treasuries.
Wang Xin, Deputy Director of the Industrial Statistics Division of the National Bureau of Statistics, pointed out:
In terms of industry dimensions, several high-tech manufacturing sectors have shown outstanding performance. The value added in sectors such as integrated circuit manufacturing, aircraft manufacturing, electronic specialized materials manufacturing, and biopharmaceutical manufacturing have all achieved significant growth, with increases of 26.7%, 24.8%, 23.9%, and 12.1% respectively.
In terms of product dimensions, the rapid development of the "Artificial Intelligence +" model has effectively driven the output of related hardware, with production of storage chips and servers increasing by 22.8% and 12.6% respectively. At the same time, emerging economic growth points such as embodied intelligence and human-machine collaboration are injecting strong momentum into the robotics field, with production of robotic reducers, industrial robots, and service robots achieving rapid growth rates of 63.9%, 28.0%, and 16.1% respectively.
Today, the National Bureau of Statistics of China released the latest statistical data: the average weekly working hours for employees of enterprises nationwide is 48.6 hours. I wonder how everyone feels about this?
In addition, I also conveniently checked the weekly working hours of workers in other countries for your reference👇
According to data released by the National Bureau of Statistics: In 2025, the completed real estate development investment in China was 82788 billion yuan, a decrease of 17.2% compared to the previous year. At the same time, the sales area of newly built commercial housing fell by 8.7% year-on-year, and the sales amount of newly built commercial housing decreased by 12.6%; while the funds in place for real estate development enterprises decreased by 13.4% year-on-year.
To facilitate everyone's reference, I have organized a trend chart 👇
From the chart, you can clearly see that the real estate development investment in 2025 has returned to the level of 2013.
By 2025, China's birth rate has dropped to 5.6 per thousand people. Data shows that the number of newborns decreased by 1.6 million, bringing the total down to 7.9 million; at the same time, the total population also decreased by 3.4 million, falling back to 1.405 billion. This is also the largest recorded decline since the Great Famine in 1960.
Currently, the trend of population decline poses a threat to economic growth, as the shrinking labor force and aging population will put tremendous pressure on pensions and social service systems. To address this, the Beijing authorities have implemented several measures, including cash subsidies, extended parental leave, and simplified marriage registration. Specific measures include providing approximately $500 annually for each three-year-old child starting in 2025.
The reasons for the decline in population are mainly due to the decrease in the number of marriages and the reduction in the number of women of childbearing age. Since 2020, the number of women of childbearing age has decreased by 16 million — this is seen as a legacy issue left by the previous one-child policy.
【Chat】To be honest, reviewing a vast array of diverse data daily has become my routine.
Since the first day I entered the industry, I have been dealing with data. In the early years, tasks such as data cleaning, analysis, modeling, and chart creation required hands-on involvement...
Later, thanks to the service upgrades from data service providers, the system can now automatically generate various high-frequency data, eliminating the need for manual operations.
With the rise of AI technology, service providers have further introduced Agent (intelligent agent) functionality. Nowadays, to query data, you only need to ask the Agent to obtain it. Although the current Agents occasionally appear somewhat "clumsy," sometimes misunderstanding or failing to comprehend instructions, their rate of progress is indeed visible to the naked eye.
From this observation, the investment from domestic financial and investment data system service providers in Agent development is unprecedented, while their overseas counterparts seem slightly less competitive in this regard.
If we talk about the spirit of "competition," it must be the Chinese... especially in the process of scaling from 1 to 100, or even from 1 to 1000, this execution capability is enough to showcase an overwhelming advantage in competition.
【Major Asset Class: Commercial Real Estate - Retail】
During the study of industry reports, a consensus has emerged in the market: retail commercial real estate is very likely to become a standout asset class in the coming period. Everyone understands that retail commercial real estate (essentially Shopping Malls) has suffered greatly from the impact of e-commerce over the past decade, leading to poor performance of such properties, and as a result, there has been almost no new supply of retail properties in the market.
Currently, the Yield of European retail properties is at a historical high.
(Here, it is important to educate everyone and please do not misunderstand the Yield data used in the investment field. A high Yield does not simply mean that the actual return on this type of asset is good, but rather that the market has high risk-return requirements for this type of asset. In other words, this means that it is currently not an asset that is highly favored by the market.)
However, precisely because the Yield of European retail properties is at a historical high, from the perspective of "mean reversion," institutions are likely to judge that the valuation and pricing of this type of asset may be at the stage of bottom reversal.
Here we introduce JP Morgan's "Long-Term Capital Market Assumptions." This report is regarded as an important cornerstone for institutional investors in strategic asset allocation, detailing the institution's benchmark predictions for global major asset returns over the next 10-15 years in 2026.
In this forecast, the top 5 asset classes ranked by expected returns are: Private Equity (PE), US Value-Added Real Estate, Japanese Equities, US REITs, and Venture Capital (VC).
As for the US stock market, the long-term return expectations for large-cap, mid-cap, and small-cap stocks are approximately 6.7%, 7%, and 6.8%, respectively.
【Perspectives on Housing Affordability in the Asia-Pacific Market: Why is Homeownership So Difficult】
Let’s delve into the data regarding housing affordability in the Asia-Pacific region. In the charts you see, the horizontal axis represents the **mortgage-to-income ratio**, while the vertical axis represents the **rent-to-income ratio**.
Specifically, let's look at the performance of several major cities:
**Hong Kong** is in the most severe situation, with a mortgage-to-income ratio that has **exceeded 100%**, and even for renting, the rent-to-income ratio has **exceeded 70%**. This means that for ordinary residents, achieving the dream of “homeownership” is currently almost impossible.
Shifting our focus to **Sydney**, the area is known for its high property prices. Data shows that its mortgage-to-income ratio is **70%**, while the rent-to-income ratio stands at **30%**.
Finally, looking at the market data for **Singapore**, its mortgage-to-income ratio is **65%**, and the rent-to-income ratio is at a level of **43%**.
In preparation for the upcoming speech, I have been intensively researching the latest performance of various asset classes recently.
Among all categories, the most intriguing are alternative assets: specifically including private credit, private equity, real estate, and hedge funds, etc.…
Once I complete the organization of all the materials, if time allows, I will then bring some detailed data and in-depth insights to share with everyone.