From Nvidia to Apple and Microsoft: Why are companies’ fortunes and fortunes so different when they all embrace AI?

Barrons

2026.11.19 21:48

Full text 6131 words

AI is changing the way companies create wealth. Standardization work such as retrieval, calculation, drafting, and verification are increasingly performed by tools. The spread of technological capabilities does not mean that gaps between companies will disappear. On the contrary, when computing power and information become easier to obtain, how leaders ask questions, allocate resources, handle conflicts, and determine long-term goals begins to more directly affect whether the company can transform technology into sustainable value.

Drucker once separated management from leadership: management is doing things right, leadership is doing the right things.

This sentence was more like a reminder in the industrial era, but today it has become a reality of division of labor.

Searching, calculating, drafting, checking, these tasks of “getting things right” are being taken over by AI in batches. What is left for leaders is: what is worth doing, where the boundaries are drawn, and who bears the costs.

There are no standard answers to such questions, only judgments.

Where does a leader’s judgment come from? In "Action Inquiry" and subsequent research, David Rooker and William Tolbert divided managers' behavioral logic into seven types: opportunists, diplomats, experts, doers, individualists, strategists, and alchemists.

This model, derived from observations of managers' behavior over time, describes how people understand power, time and conflict under pressure, rather than a wealth hierarchy. This article attempts to connect this model with the resource allocation methods of leaders of listed companies, and then discusses its possible wealth consequences.

Let’s look at two sets of numbers first. On the demand side, Gartner predicts that global terminal spending on AI models and platforms will reach US$64 billion in 2026, a year-on-year increase of 63.4%[5]. On the supply side, from the end of 2022 to November 2026, NVIDIA's market value rose from approximately US$360 billion to US$4.19 trillion, Apple's market value rose from approximately US$2.05 trillion to US$5.17 trillion, and Microsoft's market value rose from approximately US$1.78 trillion to US$3.87 trillion [6]. All three companies are rising, but the growth patterns are several times different: Nvidia is about 11.6 times at the end of 2022, Apple is about 2.5 times, and Microsoft is about 2.2 times (Figure 1).

The same technology wave, the same group of customers, and the same set of foundry production capacity fell on the reports of three companies, but they had three fates. The difference is not who works harder, but who puts AI into their system.

Therefore, the real distinction in the AI ​​era is not "whether to use AI", but "what to use AI for": technology is a public good, and cognition is private.

If the same model and the same computing power are put into different decision-making frameworks, the gap in output will widen like compound interest over time.

This article first discusses the four decision-making premises for AI changes, and then uses seven-order behavioral logic to observe the possible wealth consequences of different leadership styles. What is discussed here is the mechanism of action, which does not mean that a certain behavioral logic must correspond to a certain wealth result.

1. Bottom-level reconstruction: changes in four decision-making premises

Management science deals with choices under limited resources: modeling, setting constraints, solving problems, and then adjusting based on feedback. In the industrial era, the market is relatively stable, information transmission is slow, computing power is expensive, and the goal is single. These four premises suppress the difference between low-order and high-order behavioral logic in the long-term strategy and cannot be seen in the short term. It is these four premises that AI operates on, and the difference in returns between levels is brought to the table.

(1) Popularization of computing power: professional information is no longer scarce

Experts in the industrial age rely on long-term experience and information possession to establish authority, and their decision-making method is one-dimensional extreme value optimization.

After the efficiency of retrieval, comparison and data deduction has been improved, the work that used to take senior analysts a lot of time to complete can now be completed in a shorter time by young people who have just entered the industry with the help of AI. This does not mean that professional skills are no longer important. AI can help people obtain information, process data and compare plans faster, but it cannot replace an experienced professional to judge whether the data is abnormal, whether the plan is feasible, and what risks may be taken behind the decision. In the past, the advantages of professionals were often reflected in mastering more information and having richer experience. Now, the gap between information acquisition and processing capabilities is narrowing, and the difference between professionals is increasingly reflected in their understanding of the problem: whether they can find anomalies in a set of data, whether they can see risks that were not considered in the plan, and whether they can judge the possible consequences of a decision.

(2) Information transparency: the space for short-term speculation narrows

The reason why opportunists are able to profit from information arbitrage is often because the information is in the hands of a few people. If something goes wrong, they can shift the responsibility to others, or even try to cover up the traces left by their actions. Sowell once expressed the view that people are more likely to continue to make mistakes when the cost of their mistakes is borne by others. The development of digital technology is changing this situation. Transaction records can be traced, information dissemination speeds up, and the behavior of individuals and companies is easier to verify. In the past, some fraudulent activities or conceptual hype may have taken years to be discovered by the market, but now they are often exposed quickly. Once market trust is lost, the price a company pays may far exceed the initial gains gained through arbitrage.

(3) The pace is accelerating: the buffer period for maintaining stability is shortened

"Boiled frog" is often used to describe the gradual loss of adaptability of tissues in slow changes. The environment changes little by little, but the organization is accustomed to maintaining the original state. By the time the problem is realized, the opportunity to adjust may have been missed. This situation has become more prominent as technology iterations accelerate. Diplomat-type leaders value team relationships and internal stability. When faced with digital transformation, job adjustments and process reorganization, they are often worried that changes will cause conflicts, so they repeatedly postpone them. In the short term, the organization has remained stable, but the original problems have not been resolved. When market changes force companies to adjust, they often have to pay greater costs. The Tao Te Ching says, "To deal with problems before they exist, and to cure them before they are in chaos" emphasizes dealing with problems before they appear.

(4) Moving the goal upward: from efficiency optimization to value trade-off

Traditional business management often revolves around efficiency, cost and performance. After AI takes on more and more standardization work, simply improving efficiency is no longer all about enterprise competition. The questions leaders need to consider have also changed: When companies pursue efficiency, will they harm the rights and interests of users? When you reduce costs, will it impact employee development? Will long-term interests be sacrificed when pursuing short-term performance?

These questions are difficult to calculate to arrive at a unified answer. There may be conflicts between different goals, and leaders need to make trade-offs based on the environment and actual conditions of the company. AI can help managers analyze the costs and benefits of different options, but it cannot decide which goals should be prioritized for managers.

2. Seven-level stratification: seven behavioral logics and their wealth consequences

Rucker and Torbert summarized managers' behavioral logics into seven types, and believed that these behavioral logics reflect the different ways managers understand problems, handle conflicts, and use power. It should be noted that behavioral logic cannot directly determine a person's wealth level. Managers' judgment will affect how the company allocates resources, but the final operating results the company can achieve are also affected by factors such as industry environment and market cycles, and accidental factors cannot be ignored. As the theory of bounded rationality points out, decision makers cannot have all the information and cannot foresee all outcomes.

(1) Opportunists: local zero-sum optimization, short-term bubbles will eventually return to zero

When opportunists make decisions, they usually only look at the immediate personal benefits. As for the cost that others will have to bear and the impact on the long-term development of the company, they are not their main considerations. This behavioral logic can be summarized as a local zero-sum solution: the increase in personal benefits is not achieved by creating new value, but at the expense of the interests of other participants. Decision makers only calculate what they can gain, leaving the resulting system costs, externalities, and long-term consequences to organizations and other stakeholders.

When this kind of behavioral logic enters business operations, it will show an obvious short-term tendency. For example, some companies are keen on pursuing AI concepts. Announcements frequently mention access to large models, but they do not have corresponding technical teams, actual projects, and R&D budgets. When problems arise in operations, managers’ first thought is often to look for external causes instead of examining their own decision-making and management responsibilities. The problem is not solved, and the organization continues to fall into firefighting and blame-shifting, and even needs to package data to maintain external expectations of the company.

From 2023 to 2024, some companies appeared in the domestic A-share market. They only issued announcements on the grounds of accessing general large models and optimizing office processes, but did not have corresponding technical teams and implementation projects. Relevant news once drove up stock prices, but as market sentiment receded and performance was gradually disclosed, stock prices fell back significantly, and individual companies were even issued delisting risk warnings. There is a similar situation in the primary market: some startups have used AI as a selling point to obtain financing, but have not yet produced mature products. After the funds are consumed, they can only liquidate and leave the market. Whether it is a listed company or a start-up, the problem is not just that the technology has not kept up, but that the company has not developed the value creation capabilities to match its market narrative from the beginning. Without actual business support, it is difficult to convert the attention and financing brought by a concept into long-term competitive advantage.

The ability of opportunists to make profits over a period of time is often related to insufficient market information. The phenomenon of bad money driving out good money described by Gresham's law is based on certain information asymmetry conditions. When it is difficult for external participants to discern the true operating status of a company, companies that lack actual value support may also benefit from poor packaging and information. However, as transaction traces, information disclosure and behavioral traceability increase, this type of practice becomes increasingly difficult to maintain in the long term. Torbert's research on behavioral logic also pointed out that behavioral patterns that treat others as tools and center on self-interest are most likely to lose their living space when the environment becomes increasingly transparent.

The problem is that the gains themselves are unsustainable. What opportunists often obtain is a one-time transfer of benefits rather than new value created through products, technology and organizational capabilities. Zero-sum income cannot form compound interest, and it is difficult to accumulate as a competitive barrier for enterprises. The more transparent the information, the narrower the arbitrage window, and the means that can be used repeatedly in the past are more likely to fail after a certain risk exposure. An enterprise may gain temporarily, but ultimately pay the price of damage to reputation, loss of market position, and even unsustainable operations.

(2) Diplomat type: Prioritize compromise and solutions to constraints, and the stock of stability maintenance continues to shrink.

Diplomat leaders place great emphasis on team relationships and stability within the organization. They tend to make conflict avoidance an important management principle, sometimes even prioritizing it over business goals. When encountering situations that require adjusting personnel, changing processes, or reallocating resources, they are often worried about causing internal conflicts and would rather slow down the progress of reform than disrupt the existing balance. In order to maintain team harmony, performance requirements can be appropriately lowered, and reforms that originally need to be promoted have been repeatedly postponed. Over time, the means to maintain stable relationships become the goals pursued by the organization. This is what management science calls goal displacement.

The experience of an old domestic offline retail company can illustrate the consequences of this management method. The management was worried that the transformation of the middle office and the reconstruction of the online business would affect the original team, so it postponed the digitalization of the supply chain and the upgrade of the product selection system for several years. But the external market will not stop because of the internal needs of enterprises to maintain stability. In 2015, online retail sales of physical goods accounted for 10.8% of total retail sales of consumer goods; by 2024, this proportion has risen to 26.8%. Consumption channels and traffic structure have undergone significant changes. The company's original business model has been affected, and the number of stores and market share have also declined. By the time management has to initiate transformation, the company has lost part of its market space and has far fewer resources for reform than in the past.

Similar problems are not limited to the retail industry. In traditional industries such as manufacturing, retail, and finance, some companies have also postponed digital transformation and AI transformation on the grounds of "maintaining stability." The management was worried that the reform would break the original interest pattern and cause personnel adjustments and internal conflicts, so they continued to postpone decisions. But the market won’t wait for internal consensus among companies. Technology is iterating, competitors are adjusting, and customer needs are changing. Reforms that could have been completed gradually when there was still room for operations have often become more difficult and adjustment costs have risen by the time operating pressures truly emerge.

The concept of "goal displacement" proposed by Merton explains this phenomenon: in the process of pursuing a certain management method, organizations tend to gradually forget their original goals. Team harmony and internal coordination are originally intended to better achieve business goals. If maintaining harmony itself is regarded as the ultimate goal, the organization will easily lose its ability to correct errors. Drucker also made a distinction between coordinators and leaders: coordinators are more concerned with maintaining the existing order, while leaders need to determine the direction for the organization [9]. Of course, enterprise management requires coordination, but coordination cannot replace decision-making, nor can it be a reason to postpone reforms.

The "Tao Te Ching" says "do it before it exists, and fix it before it becomes chaotic", which emphasizes taking timely action before problems have arisen and conflicts have not yet expanded. The same is true for corporate management. Reform should often be initiated when there is still room for operation, rather than reacting passively after market share declines and operating pressure increases. The earlier problems are discovered, the greater the room for adjustment and the cost will be relatively low.

The ancients said, "If a small amount of chaos is not neglected, big chaos will result." This is also true in organizational management. Internal disagreements and contradictions are not terrible. What is really cause for alarm is the long-term avoidance of problems in order to maintain superficial calm. If an enterprise is reluctant to adjust and its original business cannot form a new growth curve, it will have to rely on consumption of stock to maintain operations. As competitors continue to improve their products, channels and management methods, the company's market share and competitive advantages will gradually weaken. This kind of elimination often does not happen suddenly, but after missed adjustment opportunities time and time again, the profitability and organizational vitality of the company are gradually eroded. It's a gentle knockout - no bang, just slow dilution.

(3) Expert type: single-dimensional extreme value optimization, linear growth meets the ceiling

Expert leaders are accustomed to solving problems from the professional fields they are familiar with, tend to seek optimal solutions around a single field and a single variable, and use personal professional judgment as the main basis for decision-making. In their view, professional ability is often more important than collaborative ability. This way of thinking can easily lead managers to fall into the dilemma of local optimality and global suboptimality: individuals may maximize their areas of responsibility without considering whether this approach is beneficial to the entire organization. He does everything himself, is unwilling to fully delegate authority, lacks interest in teamwork and human-machine collaboration, and does not pay attention to cross-department cooperation and organizational system building. Over time, the team gradually becomes accustomed to waiting for the leader to make judgments, making it difficult for other people's opinions and creativity to be used, and the development of the company increasingly relies on the leader's personal abilities.

DeepMind’s early development experience provides a case worth observing. AlphaGo and AlphaFold have achieved far-reaching technological achievements, but technological breakthroughs have not been immediately translated into a clear commercialization path, and related businesses have also faced greater investment pressure for a long time. As DeepMind is integrated into Google's industrial system and receives support from computing power, funding and business resources, its technological achievements have a broader application space and its commercial value is gradually released. There is a similar situation in China: an AI company that started with computer vision technology has achieved outstanding results in academic research and competitions. However, after listing, due to a relatively weak commercialization system, it was unable to achieve profitability for a long time, and its market value dropped significantly from the initial stage of listing. The two cases reflect a problem: technological leadership does not mean that an enterprise will be able to achieve commercial success. From technology research and development to market application, it also requires the cooperation of products, channels, capital, organizational management and other capabilities.

Judging from public AI startups, companies led by founders with a single technical background and lacking commercialization and ecological construction capabilities often find it difficult to fully unleash their technological advantages. In contrast, companies with complementary management teams that can combine technology research and development with market demand are more likely to form a complete commercialization system. Of course, corporate valuation is also affected by factors such as industry cycles, market environment, and financing conditions. It cannot be simply attributed to the professional background of the founder. However, team structure and organizational capabilities will indeed affect the speed and space for converting technological achievements into commercial value.

Lorber's theorem emphasizes that measuring a manager's ability should not only be based on how many problems he can solve when he is present, but also whether the organization can operate normally when he is not present. Expert leaders are prone to encounter limitations at this point: many things in the company rely on his professional judgment. Once he is left, it will be difficult for the team to make independent decisions. Torbert's analysis of expert behavioral logic also pointed out that over-reliance on existing professional knowledge and judgment framework will make it difficult for managers to jump out of their own cognitive boundaries and see where there is room for improvement in the entire organization. Relying on personal professional capabilities, companies can maintain technical or business advantages for a certain period of time, but this ability is difficult to replicate indefinitely, and growth is easily limited by personal energy and professional fields. The more accustomed a person is to solving all problems personally, the harder it will be for the organization to build capabilities that are independent of him. Personal abilities may eventually no longer be the driving force for corporate development, but instead become the upper limit for further corporate growth.

The first three behavioral logics reflect the different limitations that leaders may face in business decisions. Opportunists put personal gains above the interests of the organization and other participants, diplomats excessively maintain internal relationships and the existing order, and experts are easily limited by their personal areas of expertise. The three behavioral patterns are different, but all may limit a company's organizational capabilities and further development.

For an enterprise to achieve sustained growth, it is not enough to rely solely on individual professional capabilities, internal coordination or short-term profit motivation. Managers also need to establish an execution system that can continue to operate, break the existing framework when necessary, and consider resource allocation and long-term development from the perspective of the entire organization and industry.

How do leaders move from solving immediate problems to building organizational capabilities? How do individual professional strengths translate into team and system capabilities? How can enterprises move from partial innovation to the construction of industrial ecology?

In the next article, we will continue to discuss the other four types in the seven-level behavioral logic: doers, individualists, strategists and alchemists, and further analyze how leaders move from personal wealth creation to systematic wealth creation.

For more in-depth analysis and exclusive insights into global markets, multinational companies and the Chinese economy, welcome to visit the official website of Barron's Chinese website

According to the real-name requirement of the Cybersecurity Law, please bind your mobile phone number before posting a comment.

Express

2026-11-19 23:11

Director Zhu Lidan sues Wahaha Commercial Co., Ltd.

2026-11-19 23:02

Domestic commodity futures markets mostly closed higher at night, with black products leading the gains.

2026-11-19 23:01

International Energy Agency: Middle East conflict may reshape global energy landscape

2026-11-19 22:59

SpaceX plans to issue at least $32 billion in AI chip debt on Thursday

2026-11-19 22:55

Anthropic is expected to open at $99.60 on its first day of trading

2026-11-19 22:52

Livermore’s leading index of Chinese concept stocks expanded to 2%, while Pinduoduo rose nearly 13%.

2026-11-19 22:51

About 15% of shares in Anthropic IPO sold to retail investors

2026-11-19 22:50

APEC Connectivity Blueprint

2026-11-19 22:46

Anthropic's IPO doesn't place shares to majority institutional buyers

2026-11-19 22:45

Wang Yi: The meeting reached the Shenzhen Outlook, which is the most important outcome of the APEC Year in China

JPMorgan Chase says some customer information leaked in cyberattack on supplier

2026-11-19 22:39

The Nasdaq's gains quickly expanded to 2%, the Philadelphia Semiconductor Index rose 4%, and NVIDIA rose more than 4%.

2026-11-19 22:34

U.S. stocks opened: The Nasdaq opened 1.6% higher, storage, optical communications and other sectors rose, and Nvidia rose nearly 2%

2026-11-19 22:29

Yuanji Food passed the Hong Kong Stock Exchange listing hearing

Three land parcels in Beijing received 9.247 billion yuan in land auctions: China Construction Intelligent Land and Swire Properties jointly won the Chaoyang Jiangtai residential land for more than 4.9 billion yuan

2026-11-19 22:23

DeepSeek’s second round of financing lands

2026-11-19 22:17

Wu Yongming: The target of building global data centers exceeding 20GW may be lowered

2026-11-19 22:16

20 quick facts about News Network on November 19

2026-11-19 22:06

BASF announces termination of acquisition of Evonik

2026-11-19 21:58

Chongqing Luqiao: Shareholder Hangzhou Linxin plans to increase its holdings of company shares by 15 million yuan to 30 million yuan

Scan to download App