Investment Guru Says | Shangya Investment Chairman Shi Bo: Don't race with quant, look for ten-bagger stocks in the physical world. Computing power is the "coal" of the AI era.

Each Daily (每经) reporter | Li Na Liu Jinxu
Each Daily (每经) editor | Peng Shuiping

In the spring of 2026, A-share market investors have been feeling a sense of helplessness.

High-frequency trading, quantitative funds, algorithmic “harvesting”… Faced with a network of invisibility woven by machines, many investors have begun to doubt: in an age dominated by machines, does investing based on fundamental research still matter?

Amid such confusion, however, Shuangya Investment’s chairman, Shi Bo, offered an entirely different answer.

“ I just came back from a recent investigation,” he said in recent communications with investors, “I went to take a look at companies that make optical chips and optical modules. I mainly looked at things like the machine equipment and inventory, and I also了解 how the company handles its day-to-day management.”

While quantitative models are eager to capture price spreads in 0.01 seconds, Shi Bo still insists on measuring the industry with his own two feet and observing details with his own eyes. This seemingly “clumsy” approach, however, has allowed him to keep crossing bull and bear markets. Data from a third party shows that over the past 5 years, Shuangya Huo Shui No. 1 Fund’s returns have also been far above the average level of quantitative funds.

In his view, real investing has never been about racing against machines, but about moving in step with the times.

Image source: Provided by the interviewee

Discussing the investment dimension:

Direction matters more than effort

Over more than three decades of Shi Bo’s professional career, he has already formed and refined his own investment philosophy. The first principle is value investing.

There are many paths to achieve value investing. But in his view, buying stocks is buying companies. He emphasizes the need to learn how to do the math—compare a company’s total market value with the value of its actual assets, and look for companies that are severely undervalued.

But before judging whether valuations are high or low, what is even more important is determining the direction of the era—and that itself is the first lesson in value investing.

“Chinese residents’ wealth mainly comes from real estate, from the dividend of that era,” Shi Bo said. “The first dividend wave was the dividend from urbanization, the dividend from real estate. The second dividend wave was the dividend from the internet. And now, we are entering the AI revolution.”

In his view, the AI revolution is not a continuation of the internet. It is a cognition and productivity revolution that happens once every 250 years. 250 years ago, when Watt invented the steam engine, human beings entered the industrial age—physical labor was amplified by tens of thousands of times. And this time is the amplification of intelligence.

“ The internet only addressed the problem of the production relations,” Shi Bo said. “But the AI revolution can directly generate tokens and create productivity.”

“ The AI revolution is bigger and faster than the internet revolution. It is the fastest revolution in human history in terms of penetration rate. It is neither the internet nor the computer era. This is a cognition and productivity revolution that happens once every 250 years,” Shi Bo stressed.

In his view, the essence of investing is monetizing cognition. “Direction matters more than effort. Every dividend wave is the dividend of the era, not an individual’s capability. What you need to do is identify the direction and hold on to it; the rest is up to time.”

Discussing research:

From the workshop to the cafeteria—‘people’ is the first focus in investigations

In Shi Bo’s investment philosophy, quality investing is a core principle next only to value investing. Companies whose value is undervalued may not necessarily rise; what matters is the quality of management. He insists on investing in companies whose management serves shareholders’ interests and possesses great corporate qualities. And on-the-ground research is precisely to examine “people,” the core asset.

During his time at Harvest Fund (华夏基金), Shi Bo was known for “diligence.” He once investigated 10 listed companies in a single month. The years of research work was condensed into 10 pages of paper and 111 indicators. It is this research habit that runs throughout his career that drives him to keep searching for the next inflection-point industry.

Before the gold market took off, he basically visited all domestic gold companies— from Shaanxi to Shandong, going deep into mine shafts to inspect the mines. Later, he also traveled all the way to the Solomon Islands to investigate “Wangguo Gold” within the tribes. It was precisely this kind of deep on-site visit that made him see in advance the investment value of gold. In 2023, he judged that the U.S. dollar’s position would be challenged and that the U.S. would enter a rate-cut cycle, so gold would have enormous room. The subsequent market proved his judgment: the related gold stocks rose by more than ten times.

Even now, when managing private funds, he still maintains an extremely high investigation frequency. “ Now I need to research at least four listed companies every month,” Shi Bo frankly said, “and every stock I buy, I’ve gone on-site to see and to inspect.” From his words, you can feel his enthusiasm for research—that is not a task, but a genuine love from the heart.

But the way he conducts research follows his own logic.

“ When I go to investigate, I absolutely have to look at the production lines,” he said. “For companies in the computing power industry chain, you’ll see whether the number of machine units is enough, whether the equipment is running at full load, whether inventory levels are high or low—these details are more real than any financial data.”

Besides the workshop, he also pays attention to places that are easy to overlook. A company’s quality of management can’t be seen from the financial statements. Financial statements can be embellished, but the cafeteria can’t fool people. In his view, these details explain more than the income statement. How a company treats its employees often determines how it will treat shareholders.

He applies this research approach to identifying two types of companies.

One type is inflection-point companies—companies that are at the critical juncture where the supply-demand relationship reverses, and where an industry moves from 0 to 1 or from 1 to N. Shi Bo believes identifying inflection-point companies does not rely on financial statements, but on visiting the industry chain. You have to see whether the upstream raw materials are sufficient; you have to see whether downstream demand has erupted; you have to see whether competitors’ production capacity can keep up.

The other type is pivot companies—companies that have pricing power within inflection-point industries. “When the supply-demand relationship reverses and the industry enters an inflection point, you need to find pivot companies within it,” Shi Bo said. “Pivot companies are those with pricing power—their market share or profit margins are significantly higher than those of their peers.”

In his view, differences in management quality ultimately show up in these details. Good management can seize opportunities when the industry is in an upcycle, and hold the line when the industry is in a downturn. Poor management, no matter how good the track is, cannot run out of it.

Discussing the present:

Computing power is the coal of the AI era

In Shi Bo’s investment methodology, scientific investing is a consistent underlying logic.

His background in investment banking gives Shi Bo a sharp ability to sense the industry chain, business models, and competitive landscape. He divides industry investing into several stages:

The 0-to-1 stage is broad research. During this period, it is more like venture capital—what it pursues is the odds. The industry space is huge, valuations expand rapidly, but earnings certainty is insufficient. The 1-to-10 stage is deep research, where certainty improves significantly. At this time, you invest in pivot companies and leading enterprises, pursuing higher win rates. Industry barriers and first-mover advantages are extremely critical, so it is necessary to closely monitor marginal changes—especially to be wary of slower penetration rates.

Based on his judgment of the AI era, Shi Bo has built a clear investment main theme: the computing power industry chain. And his analysis of this main theme also reflects the way he uses supply-demand relationships, technical routes, and cost curves to infer the direction of industrial evolution.

In his view, the United States, leveraging advantages in technological innovation, has completed the AI breakthrough from 0 to 1. And China, leveraging strong manufacturing capability and supply-chain advantages, is expected to capture amplified opportunities in the industrialization stage from 1 to 10. This is exactly the core logic behind his long-term optimism about the computing power industry chain.

“Computing power is the coal of the AI era, the fuel of intelligence,” he analyzes. “The growth in computing power demand is exponential. Every breakthrough in each application will bring an explosion in computing power demand. The same is true for language models and video models. And once robots break through, it will be even more so. But on the supply side, it is constrained by the physical world— the downstream is the virtual world, iterating every day; the upstream is the physical world, and can only be refined little by little. This mismatch between supply and demand is the source of investment opportunities.”

Shi Bo describes the value distribution across the industry chain using elastic transmission: “ The downstream chips have the highest certainty but relatively small elasticity; the midstream optical modules have elasticity more than five times that of chips; the upstream optical chips have elasticity ten times that of optical modules; and further upstream resources like indium phosphide, tungsten, and rare earths have the greatest elasticity. In the industrial age, what makes the most money is coal; in the AI era, what makes the most money is upstream resources.”

In his view, AI computing power investment is replacing real estate and becoming the engine of the next round of economic growth. It is the most important investment main theme for the next 5 to 10 years.

“Global supply chains can’t do without China,” he gives an example. “Inside a computing power rack, there are core components like optical modules, PCB boards, and copper foil—down to upstream rare metals like tungsten and indium. The entire industry chain can’t do without China’s manufacturing capability.”

Discussing strategy:

Only if you can hold on—does that create the source of excess returns

In Shi Bo’s investment philosophy, the weight of the two words “hold on” is very heavy. He understands that only by holding long term can you benefit from the huge returns brought by compounding.

At the event, he repeatedly emphasized the importance of holding on. Most investors miss ten-bagger stocks because they lack sufficient depth of research and can’t maintain confidence amid volatility. If you truly research it thoroughly and identify the direction correctly, you won’t panic because of short-term drawdowns.

But Shi Bo’s understanding of “long-term holding” is not holding completely still. This is a dynamic process of tracking and continuously validating. He sets a benchmark company for the industries he follows. If the companies in the relevant industry chain have quarter-over-quarter growth rates for two consecutive quarters that are all below the benchmark company’s performance, he would also choose to sell.

Behind this dynamic tracking is also the company’s risk-control system. In his view, real risk control is not stopping loss after a drop, but thinking clearly before buying: how large could this company’s maximum drawdown be? When it drops by 20%, is it reducing the position or adding to it? If the answer is adding to it, then the investment is worth making.

Worth mentioning is that Shi Bo not only embraces the technology wave himself, but also requires the entire team to keep up. In daily work, he asks researchers to handle emails with “lobsters,” and to categorize and process the massive information in investment research reports to support research work. In his view, this is both an efficiency tool and a frontline position for understanding the industry. For a team investing in AI, you first have to use AI yourself.

“Volatility isn’t the risk—getting it wrong is the risk.” Shi Bo said, “Risk control isn’t stopping losses. Stopping losses is not risk control. Risk control is prediction.”

In this era of quantitative noise and market confusion, Shi Bo believes investors’ value is not in battling machines, but in moving in step with the times.

And perhaps this is the underlying password that keeps his thinking clear even as times keep evolving.

Cover image source: Provided by the interviewee