Interview with AllianceBlock Co-Founder Qiao Wang: The Claude 4.5 Inflection Point and the Trillion-Dollar Opportunity Overlooked by AI

Original Article Title: Claude Opus 4.5's Breakout Moment & Investing in 2026 with Qiao Wang

Original Source: Empire Channel

Original Translation: DeepTech TechFlow

Guest: Wang Qiao

Host: Jason Yano

Podcast Source: Empire

Release Date: January 12, 2026

Key Highlights

This week, Qiao Wang joined the show to discuss how artificial intelligence is redefining what it means to be an investor in 2026. We delved into Claude Opus 4.5's breakthrough moment, Qiao's decision to invest in Google's deal, how to construct an investment portfolio in 2026, and how to allocate time thoughtfully in the age of AI.

Highlighted Insights

· The current market sentiment is overly optimistic.

· "About 40% of my portfolio is in cash."

· Cryptocurrencies are not very attractive, but there are always market opportunities. Focus on analyzing assets individually rather than looking at asset classes as a whole.

· The stock and Bitcoin allocation is roughly 50/50.

· The largest investment is in Google. Additionally, I hold Tencent stock, as Tencent is a very outstanding company.

· An area currently overlooked by the market is AI-driven biotech, with immense potential for AI in the biotech field.

· By 2026, we may see some unicorn startups with only one or two people.

· Some of the most successful AI startups are not companies like ChatGPT or OpenAI, but rather small businesses utilizing AI. The nature of the company moat has not changed, but the moat in the software field is indeed rapidly weakening.

· Code itself is no longer the bottleneck; the key is to design appropriate "prompts."

· Gemini's value has been underestimated by at least two orders of magnitude. "I would be willing to pay Gemini a monthly fee of $2000."

· I think Adobe might be this year's Google, as its current valuation is very cheap, and the market's view of it is somewhat similar to what it was of Google back then.

· Everyone should learn to code, and it is necessary. People can optimize certain aspects of their lives and work through automation tools.

· AI tools will make already efficient and capable people even more efficient and smarter.

· In terms of health, the three key factors are diet, sleep, and exercise.

Qiao's 2026 Investment Portfolio

Yano: About a month ago, you mentioned the current state of the market and expressed some level of concern. You also referenced the Internet bubble period, such as 1996, where some predicted the market was about to peak, yet it continued to rise for another three years.

Additionally, I saw a tweet of yours discussing the relationship between forward P/E ratios and the next decade's return rates. You presented some counterarguments, but you ended with, "I'm still scared." I'm curious, what is your current view of the market?

Qiao: I am scared because the market sentiment is overly optimistic. While I can't say this is true for everyone, you can see a lot of people showing off their investment gains on social media, especially given the market volatility of the past week, this sentiment is quite evident. Quantitatively, the current market valuation is close to historic highs. Of course, some may argue that current corporate profitability is higher, competitive advantages are stronger, the economic structure is more stable, coupled with loose monetary policy driving the market, the current valuation is justified. However, we cannot deny that the market valuation is indeed at historical highs. Therefore, I am not fully allocated to risk assets. In fact, about 40% of my portfolio is in cash. While I may miss out on opportunities if the market continues to rise, at least I can sleep more soundly.

Yano: I listened to Drucker and Muller's interview where he mentioned, "In my 50 years of investment career, this is the most challenging time I have encountered in predicting market trends." If even Drucker cannot predict, how can us ordinary investors possibly do so?

Qiao: That's why I'm more focused on stock picking. When you look at the overall market, the valuation is indeed high, but if you research individual companies one by one, sometimes you'll find very attractive investment opportunities. This way, even if the overall market drops, your sense of fear will also decrease. During the 2000 Internet bubble, although tech stocks did poorly in the following decade, small-cap and value stocks performed well, with an average annual growth rate of 10%. I believe this is the best time for stock pickers.

Yano: How do you see investing in stocks and cryptocurrency tokens now?

Qiao: Currently, I believe the appeal of crypto tokens is not significant, but there will always be opportunities in the market, such as in 2022. Many worthwhile investment opportunities emerged at that time. Therefore, I am still focused on individually analyzing assets rather than simply looking at asset classes as a whole. If you look at the whole picture, there are hardly any particularly attractive assets now, except for the US dollar. But by analyzing assets one by one, you may find some opportunities. As for cryptocurrency, I still hold some Bitcoin and a small amount of tokens, but the overall investment proportion is very small.

Yano: If we try to break down your investment portfolio, 40% is cash, how is the remaining 60% split between stocks and cryptocurrency? What is the proportion of Bitcoin and other crypto assets?

Qiao: Between stocks and Bitcoin, it's approximately 50/50. I still hold a significant amount of Bitcoin, but this is not my ideal investment portfolio because selling Bitcoin would incur high taxes, and the proportion of cryptocurrency tokens is very small, less than 1%.

Yano: So what stocks do you currently hold?

Qiao: My largest investment is in Google, and I also hold Tencent stock. Tencent is a very outstanding company. Although its business is relatively low-key, its fundamentals are very robust.

Yano: Do you think Tencent is better than Alibaba?

Qiao: Currently, yes. Alibaba has a stronger advantage in the AI field, but faces fierce competition in its retail business. I also hold some Amazon stock. I think the market sentiment last year was very unfavorable to Amazon because the stock prices of all tech giants rose, except Amazon which showed weak performance. Additionally, I am very optimistic about Amazon's positioning in robotics technology. All of these investments are long-term, with a ten-year goal.

Yano: Amazon is indeed a leader in robotics technology, and they may become the first major company to have more robot employees than human employees.

Qiao: In the past five years, Amazon's human workforce has remained stable, but the number of robot employees has grown by 20% to 30%. They have significantly improved their profit margins in recent years despite not having high revenue growth.

Yano: In the healthcare sector, such as Eli Lilly or other companies, what are your thoughts?

Qiao: I hold some Eli Lilly stock, but this is a short-term trade rather than a long-term investment because my knowledge of the pharmaceutical industry is limited. I don't know what will happen in ten years, but I know that ten years from now, I may still be using Amazon, while the pharmaceutical industry has too many competitors. Although Eli Lilly has strong patent barriers, there are many cheap and effective gray market peptides that directly compete with Eli Lilly.

I think one area currently overlooked by the market is AI-driven biotechnology. While everyone is focusing on robots, drones, and AI chatbots, I believe AI has immense potential in the field of biotechnology.

Yano: I'd like to hear your thoughts on the automotive and financial technology sectors. Do you own Tesla or Rivian stock?

Qiao: No, Tesla's valuation is too high, and I don't know much about Rivian yet.

Yano: What about in the fintech space? Such as Robinhood and Coinbase?

Qiao: In terms of valuation, I think Coinbase is more attractive, but neither company is a particularly outstanding investment target. I hold stock in Coinbase, and Robinhood is overvalued.

Yano: Do you think there will be a super app like "Swiss Army Knife" developed in the West?

Qiao: Are you referring to a super app that integrates messaging, payments, social, and other functions like WeChat? I think in the financial field, there is already a similar trend, such as Robinhood. But if it's a super app that comprehensively covers all aspects of life like WeChat, I currently don't see that possibility.

The Breakthrough Moment of Claude Opus 4.5

Yano: I want to talk about Claude and Opus 4.5, and I think the best entry point is mentioning a "Wow, this is game-changing" moment you tweeted about. In the past few years, there have been several similar moments that gave me the same feeling: one was the release of ChatGPT, another was the first inference model, and there was Tesla's FSD V13 (Full Self-Driving). Why do you think Opus 4.5 is a "Wow" moment?

Qiao: I can't explain to you what happened behind it, but I can tell you how I, as a user, felt. The last time I wrote code might have been a year ago, but seriously writing code was something three years ago. From 2010 to 2017, before entering the cryptocurrency field, I was always involved in quantitative trading, writing code every day, from low-level C++ to higher-level Python data science code. That was my seven-year domain, where I also wrote some front-end and back-end code. But since leaving Messari, I haven't seriously written code, only worked on some personal projects in recent years, and only took time to do so at the end of each year.

What's different about Opus 4.5 is that before, you could quickly put together a demo, but the final 5% was always challenging to have AI complete entirely; it still required an excellent engineer to handle details like errors and edge cases. However, in Opus 4.5, I only needed to tell it in simple English what I wanted to do, provide a very clear specification, but as long as the specification was clear and comprehensive enough, it could complete it in one go.

Claude and Opus may be going through a hype cycle on Twitter, as I had an engineer friend tell me that OpenAI's latest GPT-5 Pro is almost as good as Claude.

Yano: I've used both of these, but actually Opus is what got me to switch from ChatGPT for the first time. The memory feature in ChatGPT was something I heavily relied on, to the point where I felt like I might just keep using it because it knew everything about me, but now Opus 4.5 has been a game-changer for me.

There's a new way of working now that I'm not sure what to categorize as anymore, whether it's a chatbot or something else, as the line between the two is getting blurry. For example, I just used Opus 4.5 to get something done. We have an eight-person sales team at Blockworks, and typically we need to decide who should be responsible for selling to each account and assign priorities to the accounts like Tier 1, Tier 2, etc. Each salesperson has their own judgment, like someone handling 10 Tier 1, 30 Tier 2, and 100 Tier 3 accounts. This allocation work is often complicated. So, I handed this issue over to Claude and had it consolidate the relevant information. I told it that if a token's FDV (Fully Diluted Valuation) is over $1 billion, that account should get additional priority; it's even better if the token's project is a public company.

Claude extracted relevant information from multiple data sources like CoinGecko and CoinMarketCap, ranked the accounts, analyzed each salesperson's past eight years of transaction records with Blockworks, understood the types of products we sell, and based on each salesperson's likelihood of transacting with these accounts, allocated the accounts accordingly to the right people.

Qiao: I also realized the situation you're talking about during the holidays, where the boundary between a general-purpose chatbot and traditional coding is gradually blurring. When conversing with a general-purpose chatbot, it usually generates a piece of code on the fly based on your needs, and now when you need to code, you don't even have to code yourself, just express your needs in simple natural language to achieve it. This is causing the functionalities of a code assistant and a general-purpose chatbot to merge over time.

Yano: Exactly. So why didn't you just use Replit or Lovelace directly to do your Vibe code work? Why choose Claude instead?

Qiao: I feel like they target different user groups. I haven't tried the latest version of Lovelace yet, but I feel like Lovelace is more suitable for users who want to quickly build a nice-looking presentation or simple application. They perform well in front-end development, although I haven't used them myself. So I initially used Cursor with Opus 4.5, then tried Claude and Opus 4.5. To be honest, in terms of productivity improvement, I didn't feel a significant difference. For me, they perform similarly, but that's just my personal experience.

How Does Artificial Intelligence Impact Startups?

Yano: You've witnessed the growth of thousands of crypto startups. How does artificial intelligence now impact these companies?

Qiao: The impact of artificial intelligence on startups is huge, especially over the past three years. Basically, I would ask in each group, especially the technical co-founders of each startup, since the release of ChatGPT in 2022, how much has your productivity increased? Almost every time, the answer is higher than the last time. This trend is very consistent. The most recent group's response was that their productivity has increased by about 3 to 4 times.

I think this impact is more significant in early-stage startups, while its effect on later-stage large companies is relatively smaller because one limitation of a code assistant is its "context window." For example, if Claude can handle a million tokens, it means it can efficiently help you kickstart a new project and work well. But if you ask Claude to modify the entire Google codebase, that's almost impossible, right? So these tools are more helpful to early-stage small startups and relatively limited for large companies.

In large companies, the most effective way to leverage a code assistant is by establishing clear interdepartmental abstraction layers. This can break down complex tasks into small pieces, narrow the context scope, and then input them into an AI system to make it easier to understand.

Yano: However, I don't think using the term "productivity" to describe the impact of artificial intelligence is entirely accurate, especially in early-stage startups. If you are a company with 100, 200, or even 1000 people, a 3x to 4x increase in productivity might be a significant change. But for early-stage startups, they wouldn't even consider the question of "productivity." Their thought is more like: Why should we hire more people? I often see this situation in the companies I've invested in and my friends' startups.

They might think that since we have AI, we don't really need to hire new employees anymore. This phenomenon is really interesting. For example, I recently built two tools in the sales area: one is a sales commission calculator. The sales team often asks me how much they can earn. Previously, we needed to involve the finance team for calculations, and the finance team found it troublesome. Now with this tool, the problem is solved.

Another tool is a sales data dashboard. This tool can calculate the cost of the dashboard. Previously, we needed to involve the data team, and consider whether they used Dune or Gold Sky to index the data. If not, we had to pay additional fees. Now, I have developed a dashboard cost calculator based on all ports.

What is the significance of these small tools? They mean that we no longer need to hire someone specifically for sales support, so by 2026, we may see some unicorn startups with only one or two people. These startups may have already taken off, although they have not yet reached a billion-dollar valuation, they are already rapidly growing.

Qiao: Absolutely. I know many people who are now running a $10 million annual recurring revenue business alone, often former engineers from companies like Meta and Uber, who are tired of the bureaucracy of large companies and have chosen to start their own businesses.

Yano: However, I've noticed an interesting phenomenon: some of the most successful AI startups are not companies like ChatGPT or OpenAI, but rather small businesses utilizing AI. These companies are unwilling to disclose their core business to the public. Usually, if you have a startup, the business is doing well, and revenue is growing rapidly, you would want to tell the whole world, raise funds, or promote it on social media.

Do Moats Still Exist?

Yano: How do you view a company's moat? Do you think the definition of a moat will change?

Qiao: The essence of a moat has not changed, but moats in the software field are indeed rapidly weakening. For early-stage startups, there is hardly any moat to speak of. Companies like Facebook, Google, Microsoft, and Apple still have solid moats. AI code assistants cannot destroy these moats. For example, Apple's moat is its developer ecosystem, while Microsoft's moat is the high cost for users to switch from PC to other platforms. In the cloud services field, AWS, Azure, and Google GCP have moats in the form of high costs for customers to switch cloud platforms. Additionally, platforms like YouTube with vast proprietary data can develop very powerful video models using that data. Similarly, Microsoft's enterprise software is a crucial tool. Of course, you can replicate the features of Office, but will enterprises really switch from Office to other software? For them, these tools are too important, and the switching costs are too high.

However, I recently noticed an example that might be one of the biggest undervalued cases in the current financial market, and that is Adobe. Photoshop and Adobe's Creative Suite are very famous, and there is a viewpoint in the market now that the latest video and image generation models will replace Adobe's products. But I believe this view is completely wrong because Adobe's moat lies in its enterprise-level integration capabilities, many enterprise users using Adobe's Creative Suite store their images and videos in Adobe's cloud. For these creative industry professionals, the cost of switching from the Adobe cloud to other services is very high.

Furthermore, many creative workers have been using Photoshop for many years, and the operation has become muscle memory. For them, switching to other tools is very difficult. Therefore, Adobe's current P/E ratio is only 12 times, which is unbelievably undervalued for such a high-quality company.

Yano: Suppose you work at a company, such as a 500-person, ten-year-old startup. How should the company respond? Many young entrepreneurs may already be familiar with these technologies, but for a more mature company, it may not be easy to adapt. Employees may just throw things into ChatGPT and send automated emails.

Qiao: I believe you can't force a company to adopt new technology; you must let them find the immediate application scenarios of AI technology and see the significant impact it can bring. By the way, this is actually our experience at Alliance over the past three years. Three years ago, we knew AI would become an important trend, and we proposed to be an "AI-first" organization at that time, but we did not try to force AI into every corner of the organization.

Instead, we used AI to automate some specific processes. For example, we receive thousands of applications every year. Three years ago, I needed to read all the application documents myself, about 5,000 a year. This work was very exhausting and really drained me. But now, we have automated about 50% of the work using AI.

Yano: How did you do that? Did you use Opus 4.5? How was this system built? What was the input, and what was the output effect? Why can only 50% be automated? Why not achieve 99%?

Qiao: Yes, this software was developed by our engineers. I believe the code itself is no longer the bottleneck; the key is to design appropriate "Prompts". I wouldn't call this a secret weapon because everyone actually knows some characteristics of excellent founders.

I simply converted these rules of thumb about excellent founders or excellent startups into prompts to use for screening applications. Currently, the main function of this system is to exclude those obviously unqualified applications rather than directly tell me who the best candidates are, as I believe this part still requires the judgment of human interviewers. AI has not reached this level yet, but I believe by the end of this year, you will see AI and AI-driven venture capitalists outperforming humans in this regard.

Yano: I feel like the price of these tools is just too low right now, similar to when Uber was first launched, where a trip from the New York Financial District to the Upper East Side was only $5. At that time, you would think that price was obviously a mistake, but they used venture capital subsidies to attract users.

How much do you think these tools are worth? How much do you think these tools are currently undervalued? For example, how much would you be willing to pay for Opus 4.5?

Qiao: As for Opus, I'm not quite sure right now; it depends on what I can build with it. But taking Gemini as an example, I think its value is underestimated by at least two orders of magnitude. Right now, I only need to pay $20 to use its professional version, and I haven't even upgraded to Pro Plus yet.

Yano: If Gemini charged $2000 per month, would you be willing to pay?

Qiao: Yes, because its features are very powerful. In fact, I've said on Twitter that it functions as a research assistant, a junior researcher, a junior code assistant, a pretty decent medical consultant (can check a doctor's advice), and a pretty decent legal assistant. Overall, when you add up all these features, the price of $2000 is simply a bargain.

Google Investment Decision

Yano: How much time do you spend using apps like Gemini and Claude every day? Where did you squeeze out this time from your daily activities? For example, did you reduce Zoom meeting time, or did you spend less time on social media?

Qiao: In fact, I analyzed my time allocation half a year ago, which also prompted me to make the decision to invest in Google. I looked at my iPhone usage records and found that my top three most used apps are Chrome, YouTube, and Gemini, all of which are Google products.

Of course, I was initially concerned that ChatGPT might threaten Google's search business, so I talked to my wife about it. She gave me some inspiration, saying that her most common use case for Google Search is shopping, a function that ChatGPT cannot fully replace in the short term. I then did some research and found that over half of Google's search revenue actually comes from shopping ads, which made me feel that Google's core business is still very strong.

No one can shake its position in the short term. Furthermore, Google also has technological advantages like GCP (Google Cloud Platform) and TPU (Tensor Processing Unit), which are key points that I only realized later. I believe these factors all indicate that Google's moat is very solid. Therefore, last year, I decided to invest in Google, which was almost my only significant investment last year.

Yano: On New Year's Eve last year, I had dinner with a Google employee, and we talked about these things. Everyone was discussing their commonly used tools. That Google employee mentioned that many people overlook one fact, that Google has access to vast shopping data, and it seems that the market has not fully realized this yet. So, are there any other similar opportunities?

Qiao: I think Adobe might be this year's Google, because its current valuation is very cheap, and the market's view of it is somewhat similar to the view of Google at that time. I've never used it myself, but I know it has many use cases.

Yano: No new user will go to use Adobe. Just like Google, my commonly used apps are YouTube, I use YouTube very frequently, and also Gemini. But Adobe? I don't use it at all, so I don't think new users would choose Adobe either.

Qiao: But all new users will use Google. I think a common misconception in the market is that Adobe is not a consumer-oriented product, but an enterprise product.

Yano: Enterprise product is correct, but my company would never consider using Adobe; Figma is more popular.

Qiao: In fact, Figma and Adobe serve two different markets. In the market competition for new project presentations and website development, Adobe has already lost, and Figma has taken the lead in this field. What about Canva?

Yano: Canva targets the low-end market, more suitable for amateur users, while Adobe serves high-end enterprise users. So I agree with your point that Adobe's new user growth has indeed been stagnant for a long time, but its pricing power is very strong, able to continuously increase subscription fees.

How to Allocate Time Reasonably in 2026

Yano: As a Google user, I would like to discuss the topic of investment, but I am more interested in exploring one question: In 2026, how should we reasonably allocate our time? How do you think AI will change the way we allocate time?

Qiao: Well, I believe everyone should learn to code, and it is a must. If not, one may be left far behind . However, here, coding does not mean traditional programming but refers to optimizing some aspects of life and work through automation tools. I believe there will be many B2B SaaS software in the future, such as Gmail, Zoom-like tools, which everyone will still pay to use. But at the same time, everyone's workflow will have specific needs, and these needs are very personalized. Third-party software companies may not develop these targeted solutions.

Yano: For example, at Blockworks, we can develop a commission calculator ourselves. This is a great example; no one will develop a dedicated SaaS software to do this kind of thing.

Qiao: And when I say "everyone should learn to code," it doesn't have to be intimidating. Because today's technology has advanced to the point where you don't actually need to write complex code. You just need to converse in natural language with the system to achieve automation.

So, I would strongly recommend that everyone doesn't necessarily have to use a sophisticated tool like Claude Opus. I can recommend some simpler and more user-friendly tools. For example, Replit. The first time I realized the potential of artificial intelligence was through Replit. That's when I understood that AI is not just a chatbot; it can help you build anything. This technology is going to fundamentally change our world. I highly encourage everyone to try out Replit; it's really awesome.

Yano: What impact do you think this gap will have on the labor market?

Qiao: I think this will be like the proliferation of the internet, these tools will make those who are already efficient and capable even more efficient and smarter, while those who are inefficient may fall further behind. Artificial intelligence is an incredibly powerful tool that can significantly increase efficiency, but it ultimately depends on how much you are willing to use it.

Using AI Models for Investment

Yano: From a certain perspective, standing on the opposing side, do you really think the Warren Buffett Stock Tracker you spent time building is worth it?

Qiao: Yes, it did take up quite a bit of my time. Although the results have been good, more importantly, in building this tool, I gained a deeper understanding of the relevant technology.

This tool can be seen as a digital clone of investment masters like Warren Buffett, Charlie Munger, Howard Marks, Peter Drucker, and Bill Miller. It regularly scans thousands of stock codes and conducts in-depth research on each stock, attempting to simulate Buffett and Munger's investment logic.

The code part is actually straightforward, but the prompt design took me several months of continuous adjustments. I designed a very detailed set of prompts to simulate Buffett and Munger's thinking process in evaluating potential investment opportunities.

This process is divided into six steps. First, it collects information on six key aspects of interest through in-depth research models, then calls another API to perform analysis using an inference model. In-depth research and the inference model are two separate stages.

The in-depth research model is very good at gathering facts and data, while the inference model, although sometimes generating some incorrect information, excels in logical deduction. Once accurate data is fed into the inference model, its analytical capabilities far surpass those of the in-depth research model.

Next, I will call the inference model in the second stage, allowing it to simulate the "Digital Buffett and Munger" investment committee to analyze whether it is worth investing in a particular stock. Ultimately, it will provide a specific recommendation, which concludes the entire process.

Have you heard of a theory? Some believe that Renaissance Technologies discovered Large Language Models (LLMs) long before others, but they kept this technology secret, which may be why their investment returns are so high. However, I designed the Buffett and Munger model with the intention of avoiding competition with Renaissance Technologies. They excel in short-term trading, such as intraday trading or operations within a week. In this short time scale, it is indeed challenging to compete with the latest AI models.

Yano: So, are you also competing with companies like Susquehanna?

Qiao: Yes, but my model leans more towards long-term investment. In today's market, hardly anyone has the patience to hold a stock for more than 5 minutes, and this is where language models can play a role.

Yano: So, how do you combine the advice of different investment masters? For example, Howard Marks' advice and Peter Drucker's advice may be completely different, and their methods are diametrically opposed to Buffett.

Qiao: I will weight their advice and consider it holistically.

Yano: Do they all recommend buying Adobe?

Qiao: Yes, and an interesting phenomenon is that if you run the same prompt multiple times, the model may give different answers each time. So, if you run the same question multiple times and average the results, and each time the result suggests buying a particular stock, then the credibility of that recommendation will greatly increase.

It recommended about ten stocks, four of which are actually already in Berkshire Hathaway's portfolio. For example, Chubb, an insurance company, and Google.

How is Artificial Intelligence Changing Branding and Distribution?

Yano: How will artificial intelligence change the way brands communicate and distribute content? For example, as we mentioned before, Delphi, I don't know if you noticed OpenAI's recent launch of the Sora tool. Through Sora, you can clone someone's likeness and embed it into your video. This reminds me of the changes in marketing and branding. In the near future, brand marketing will obviously become more personalized.

Qiao: This trend has actually already begun. A few days ago, I saw an ad on LinkedIn that read, "Hey, Jason, as a Co-Founder of Blockworks, I really think you would like Rippling."

Yano: Artificial intelligence will make their ad algorithms more intelligent, and ads will become very precise. In fact, a few days ago, I searched for a term online — High Rocks. It's a marathon-style fitness event similar to CrossFit, and I've been training for High Rocks recently. So, when I searched for the High Rocks fitness app on the App Store, I found some apps specifically designed for High Rocks training. But High Rocks is a very niche activity, so there shouldn't be so many specialized apps for it in theory.

I found this strange, so I wanted to know if these apps were really designed for High Rocks or if they were just customized ads for users searching for "High Rocks." Sure enough, these apps are actually some general fitness and nutrition apps, but their ad targeting is very precise, so I think this kind of personalized advertising will become more common in the future.

Health and Longevity

Yano: What do you think is the most effective change in terms of health?

Qiao: Ultimately, the most important things are still the three basic factors: diet, sleep, and exercise. Back in 2021, I was very focused on optimizing details, such as trying various supplements, sauna, and other methods. But after four to five years of personal practice, reading a lot of research, and listening to countless podcasts, I found that nothing is more effective than getting a good 8 hours of sleep every day, maintaining a healthy diet, and sticking to regular exercise.

Yano: So, how do you manage your diet and exercise?

Qiao: To be honest, I no longer obsess over optimizing my diet to the extreme. It was causing me a lot of stress. I just try to maintain a healthy diet without putting too much pressure on myself.

Yano: This reminds me of a neurology meme, where the person in the middle does elaborate food prep every day and takes 17 supplements in the morning. And your current attitude is like, "Just eat healthy, don't overthink it."

Qiao: Yes, I find that if I try to optimize everything to the extreme, like Brian Johnson does (Note: Brian Johnson is an entrepreneur known for extreme health management), I would feel a lot of pressure. And stress leads to increased cortisol levels, which is not beneficial for longevity.

Original Article Link