Original Title: From Hierarchy to Intelligence
Original Author: Jack Dorsey
Translator: Peggy, BlockBeats
Editor's Note: While most companies still view AI as a "productivity tool," Jack Dorsey pushes the question forward: Is AI rewriting the operational logic of organizations themselves? As the co-founder and former CEO of X and the founder of Block, he has long focused on the relationship between technology and organizational forms.
This article starts from history to revisit why companies have evolved into their current forms and why this structure is beginning to loosen. From Roman legions to modern companies, the organizational evolution over the past two thousand years has always revolved around the same constraint: completing information dissemination and collaboration within limited "management spans." Hierarchical structures, middle management, and matrix systems are essentially different solutions to this problem.
The emergence of AI has, for the first time, impacted this premise. When information can be modeled, understood, and distributed in real-time, does the organization still need a coordination mechanism centered around "people"?
Similar changes have already appeared in reality. Recently, People reported that the AI company "The Dark Side of the Moon" has a team of over 300 people with no departments, no ranks, and no OKRs or KPIs; collaboration relies on direct communication rather than layered reporting; five co-founders directly manage 40-50 employees each. Meanwhile, agents have been embedded in daily workflows, able to quickly organize information, design products, and even generate code. This structure is not merely about "removing management," but rather about placing complexity into recruitment, mobility, and tools systems.
Related Reading: “How The Dark Side of the Moon Works: Over 300 People with No Departments or Ranks, 5 Co-founders Directly Manage 40 to 50 People Each”
Using the practices of Block as a starting point, this article further proposes a more radical vision: transitioning from "hierarchical organizations" to "intelligent companies," replacing traditional information routing systems with a "company world model + customer world model + intelligence layer," even reconstructing middle management itself. This is not only an efficiency issue but may also represent a rewriting of organizational forms.
Here is the original text:
According to Sequoia Capital, "speed" is the best indicator of whether a startup can succeed. Most companies still view AI as a tool for productivity enhancement, with only a few starting to focus on how AI changes the way people collaborate. Block is demonstrating a completely new path: fundamentally reconstructing organizational design and using AI as a compound competitive advantage that continuously amplifies "speed."
The Starting Point of Hierarchical Organizations: From Roman Legions to Modern Companies
Two thousand years before the emergence of corporate organizational charts, the Roman army had already solved a problem that still troubles large organizations today: how to coordinate thousands of people in cases of limited communication and great distances.
Their solution was to establish a nested command system while maintaining a relatively stable "management span" at each level. The smallest unit is the "tent group" (contubernium), composed of 8 soldiers sharing a tent, equipment, and a mule, led by a decanus. Ten tent groups make up a century (century, actually about 80 people), commanded by a centurion; six centuries form a cohort; and ten cohorts constitute a legion of about 5,000 people.
At each level, there are clear commanders responsible for summarizing information upward and conveying orders downward. This structure from 8 → 80 → 480 → 5000 essentially serves as an efficient information transmission mechanism, based on a simple yet critical premise: an individual can effectively manage only 3 to 8 people directly. The Romans gradually figured out this rule through prolonged warfare. To this day, the structure of the U.S. military still broadly follows similar logic. We call this constraint the "span of control," which remains a fundamental limitation that all large organizations cannot avoid.
The next significant change came from Prussia.
After suffering a crushing defeat at the Battle of Jena in 1806 against Napoleon, Scharnhorst and Gneisenau led military reforms, proposing an uncomfortable reality: reliance on individual genius must be replaced by reliance on the system. They established the "General Staff," training a type of officer whose role was not combat but planning operations, handling information, and coordinating across units. Scharnhorst's original intention was to "compensate for the deficiencies of incompetent generals by providing them with the capabilities they lack." This effectively laid the groundwork for "middle management": a group of professionals responsible for information transmission, preemptively calculating decisions, and maintaining the coordination of complex organizations. Meanwhile, the military also clearly distinguished between "line" and "staff" functions: the former is responsible for advancing core tasks, while the latter provides professional support. This distinction is still widely adopted by companies today.
In the 1840s and 1850s, American railroad companies introduced the military hierarchical system into the commercial world.
The U.S. Army supplied a large number of West Point-trained engineers to railroad companies, who brought military organizational thinking with them. Line and staff structures, divisional divisions, and bureaucratic reporting and control systems all initially originated in the military. In the mid-1850s, Daniel McCallum of the New York Erie Railroad Company created the world's first organizational chart to manage a railroad system spanning 500 miles and thousands of employees. Prior informal management methods suitable for small railroads had become ineffective, resulting in frequent train collision incidents. McCallum institutionalized Roman hierarchical logic: clarifying power and responsibility levels, defining reporting relationships, and structuring information flow. This became the prototype of modern companies.
Subsequently, Frederick Taylor (known as the "Father of Scientific Management") optimized the internal system. He broke work down into specialized tasks, assigned them to trained experts, and managed them through quantitative metrics rather than intuition, thus forming a "functional pyramid" structure—an organizational form that maximizes efficiency within the existing information routing system.
The first major stress test of this functional structure occurred during World War II with the "Manhattan Project." The project required physicists, chemists, engineers, metallurgists, and military personnel to collaborate interdisciplinary under extreme secrecy and time pressure to achieve a singular goal. Robert Oppenheimer utilized functional divisions at Los Alamos Laboratory but insisted on open cross-department collaboration, resisting the military's tendency for "isolation." In 1944, when "implosion" became a key bottleneck, he reorganized teams to establish cross-functional groups, which had no precedent in the corporate world at the time. This model worked, but it was a wartime exception driven by a few exceptional individuals. After the war, the corporate world faced the question: Can this cross-functional collaboration be normalized?
Post-war corporate growth and global expansion made the limitations of the functional structure increasingly obvious.
In 1959, Gilbert Clee and Alfred di Scipio of McKinsey published "Creating a World Enterprise" in the Harvard Business Review, proposing the "matrix organization" framework, combining functional specialization with divisional structures. Under the guidance of Marvin Bower, McKinsey helped companies like Shell and General Electric implement this model, achieving a balance between "central standards" and "local flexibility." This system became the paradigm of "modern enterprises" in the post-war global economy.
Subsequently, to address the complexities and bureaucratization of the matrix structure, new management frameworks continued to emerge.
In the 1970s, McKinsey introduced the "7-S model," distinguishing between "hard elements" (strategy, structure, systems) and "soft elements" (shared values, skills, staff, style), emphasizing that relying solely on structure cannot guarantee organizational effectiveness; cultural and human collaboration are also essential.
In recent decades, technology companies have conducted more radical experiments with organizational structures.
Spotify introduced cross-functional squads and rapid iteration cycles; Zappos experimented with Holacracy, eliminating management titles; Valve adopted a flat structure with no formal levels. These attempts revealed the limitations of traditional hierarchies, but none fully resolved the issue: Spotify returned to traditional management after scaling up; Zappos experienced significant employee turnover; Valve's model struggled to scale to hundreds of people. When organizations grow to thousands, they must revert to hierarchical coordination due to the lack of more effective information routing mechanisms.
This constraint is identical to the problems faced by Romans and the U.S. Marine Corps in World War II: narrowing the management span means increasing levels, which slows down information flow. For two thousand years, organizational innovation has tried to circumvent this trade-off but has never truly broken it.
So, what is different now?
At Block, we are beginning to question a foundational assumption: organizations must rely on humans as coordination mechanisms and adopt hierarchical structures. Our goal is to replace the function of hierarchy with systems. Currently, most companies are merely equipping employees with AI co-pilots to make existing structures operate a little better, but the essence remains unchanged. What we aim to build is another form: a company that itself is an "intelligent agent" (or even a small AGI).
We are not the first organization to attempt to transcend hierarchies. Haier's "one person, one unit," platform-type organizations, and "data-driven management" are all similar explorations. But they lack a critical element: technology that can truly fulfill the coordination function. AI is that technology. For the first time, a system can continuously maintain a model that upholds the overall operation of a business, and coordinate based on it, without requiring humans to relay information through hierarchies.
To achieve this, a company needs two things: a "world model" of its own operations, and sufficiently rich customer signals.
Block primarily relies on remote work, where all tasks leave behind recordable "traces": decisions, discussions, code, designs, plans, issues, and progress. These constitute the raw materials for the company's world model.
In traditional companies, managers are responsible for understanding team states and relaying information up and down; however, in a "machine-readable" organization, AI can continuously build this overall view: what is being done, where things are stalled, how resources are allocated, what is effective, and what is ineffective. This information, which used to be carried by hierarchies, is now supported by models.
However, system capabilities depend on the quality of input signals, and "money" is the most authentic signal. People may lie on surveys, ignore ads, or abandon shopping carts, but when they consume, save, transfer, borrow, or repay money, those actions are real. Block sees both sides of the transaction every day: seeing buyers through Cash App and sellers through Square, while grasping merchant operational data. This allows it to construct a rare customer world model—understanding financial behaviors of individual customers and merchants based on real transaction signals, and this signal continuously accumulates and strengthens.
The company world model and customer world model together form the foundation of a new type of company. In this model, companies no longer operate based on established roadmaps devised by product teams, but rather around four core constructs:
First, capabilities: basic financial capabilities such as payments, lending, card issuance, banking, buy now pay later, salary payments, etc. These are not products but rather underlying modules; they lack interfaces but meet reliability, compliance, and performance requirements.
Second, world model: including company model (understanding its own operations) and customer model (customer and market representation built on transaction data), gradually evolving into a system with causal and predictive capabilities.
Third, intelligence layer: providing capabilities for specific groups of customers at specific times, proactively offering solutions. For example, when the system predicts a restaurant's cash flow is about to become tight, it automatically assembles loan and repayment solutions and pushes them in advance; or when changes in user behavior suggest a move, it automatically configures new financial service combinations. These do not require prior design by product managers.
Fourth, interfaces: such as Square, Cash App, Afterpay, TIDAL, etc., which are merely delivery interfaces, with real value arising from the models and intelligence layer.
When the system attempts to assemble solutions but finds a lack of certain capabilities, this "failure signal" becomes the product roadmap for the future. The traditional method of imagining demand by product managers is directly replaced by real customer behavior.
In this structure, the organization will also change accordingly. In traditional companies, intelligence resides in individuals and is routed through hierarchies; here, intelligence exists within the system, with people at the "edge." The edge is where intelligence meets reality. People can perceive intuitions, culture, trust, and complex situations that models cannot capture, and play roles in ethical and high-risk decision-making. However, they do not need to coordinate through hierarchies because the world model provides the necessary context.
In practice, organizations will simplify into three types of roles:
· IC (Individual Contributor): experts in building capabilities, models, and interfaces;
· DRI (Directly Responsible Individual): mobilizing resources around specific issues or customer outcomes;
· Player-coach: participating in frontline work while cultivating talent, replacing traditional managers.
There is no longer a need for a fixed middle management layer, as the remaining coordination work is completed by the system.
Block is currently still in the early stages of this transformation, which will be a difficult process; some attempts may fail. However, we openly share this direction because we believe every company will eventually face the same question: Are you continually deepening your understanding of a complex issue?
If the answer is no, AI is just a tool to reduce costs; if the answer is yes, AI will reveal the true essence of the company.
Block’s answer is the "economic graph": connecting millions of merchants and consumers, understanding behaviors on both sides of transactions in real-time, and continuously accumulating insights. We believe that this model of "organizing the company around intelligence rather than hierarchy" will reshape the operational methods of various enterprises in the next few years.
The speed of a company fundamentally depends on the speed of information flow. Hierarchies and middle management slow down this flow. For two thousand years, from the Roman army to modern enterprises, we have not had better alternatives. But now, this premise is changing. Block is building the next form.
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
