In early 2026, Block cut roughly 40% of its workforce, more than 4,000 employees, even though the business was growing. Gross profit reached about $10.36B in 2025, up roughly 17% year over year. Soon after the layoffs were announced, the stock jumped around 20–25% in a matter of days.
A few days ago, Jack Dorsey and Sequoia’s Roelof Botha published the essay “From Hierarchy to Intelligence.” The piece frames the layoffs as part of a “deeper organizational shift.” The argument centers on how companies coordinate information and how AI changes that constraint.
The essay’s historical framing starts with the Roman military. Command structures were layered because communication was slow and information had to travel through small spans of control. The Prussian General Staff formalized another coordination layer, creating staff officers whose job was planning and operational support. Corporate management inherited much of this structure. Middle management became the mechanism that routed information, aligned teams, and translated strategy into operational work.
Block’s argument is that software can now perform a large portion of that coordination task.
The proposed operating model
The essay outlines an organization built around an internal “intelligence layer.” This system would operate on two continuously updated models.
The first is a company world model, derived from internal artifacts such as code, tickets, design documents, incident reports, and operational metrics. Block’s remote-first workflow produces a large amount of machine-readable documentation, making it easier to assemble a comprehensive internal dataset.
The second is a customer world model, built from the transaction flows running through Cash App and Square. These systems observe both sides of financial activity: consumer spending and merchant revenue. Over time that data can form a detailed picture of financial behavior at both the household and business level.
According to the essay, the intelligence layer analyzes these models to identify opportunities and operational problems. Product capabilities such as payments, lending, payroll, or card issuing become modular building blocks. The system can combine them dynamically when the data suggests a need.
The human organization is simplified around three roles:
• individual contributors producing work
• Directly Responsible Individuals (DRIs) assigned to specific problems for defined time periods
• Player-coaches who still contribute as individuals while mentoring and coordinating others
Permanent management layers disappear in this framework.
Where the strategy has real grounding
Several elements of the proposal align with practical trends in modern software organizations.
Block’s remote-first workflow generates a large volume of structured artifacts. Design discussions, decisions, and operational records exist primarily as text and structured data. This environment lends itself to building searchable internal knowledge graphs and applying language models to analyze operational patterns. Many companies are experimenting with early versions of this idea through internal copilots and operational dashboards.
Block also has unusually rich financial signal. Through Cash App and Square, the company observes millions of consumer and merchant transactions every day. That dual perspective makes it possible to detect merchant cash-flow stress, shifts in consumer spending behavior, or emerging demand patterns across small businesses. AI systems that analyze these signals could support proactive financial services such as lending offers or liquidity tools.
The company’s financial position creates room to attempt this type of experiment. With billions in annual gross profit and continued growth, Block can absorb the risk associated with restructuring both its organization and its product strategy.
Where the proposal becomes vague
The essay provides very little operational detail about how the intelligence layer would function.
Key questions remain open. The architecture behind the “world model” is not described. Data governance, accuracy guarantees, and update latency are not addressed. In a financial services environment, automated decisions also require regulatory oversight. Lending offers, transaction monitoring, and risk scoring operate under strict compliance rules. The essay does not explain how those constraints would be encoded into automated decision systems.
The organizational design raises similar questions. DRIs receive authority to solve specific problems, but the mechanism for resolving conflicts between competing priorities is unclear. Performance reviews, promotions, and compensation decisions still require some form of structured oversight. The essay replaces traditional titles without fully explaining how those responsibilities are redistributed.
The product examples presented in the essay are also hypothetical. One scenario describes a merchant whose cash flow tightens. The intelligence layer identifies the problem and assembles a short-term loan tailored to the merchant’s revenue cycle. Another scenario involves a consumer whose transaction patterns indicate a major life change, triggering new financial tools and account features.
The essay does not indicate whether these systems are already deployed, piloted in specific markets, or still conceptual.
Why the narrative resonated with investors
The organizational story arrived alongside the workforce reduction. Investors responded positively. The combination of layoffs and strong financial guidance suggested rising operating leverage. The AI narrative framed the shift as a structural transformation rather than a cost-cutting exercise.
Public markets respond strongly to stories about productivity driven by automation. The idea that a smaller team could run a complex fintech platform with the help of AI coordination tools fits that narrative.
Whether the internal systems currently deliver that level of leverage remains difficult to evaluate from the outside.
The broader context around Block
Block has also faced regulatory scrutiny in recent years. Investigations and settlements have addressed fraud risks, dispute-handling practices, and anti-money-laundering compliance within Cash App. Financial regulators continue to monitor the company’s systems for transaction monitoring and sanctions enforcement.
Those issues shape how readers interpret the intelligence-layer vision. Automated decision systems operating on financial data must satisfy regulatory oversight as well as operational efficiency.
The essay focuses primarily on the opportunity side of the equation.
A grounded interpretation
The underlying idea reflects a real shift in how companies coordinate work. Modern organizations produce large volumes of digital artifacts. AI systems can analyze those artifacts, detect patterns, and surface operational insights that previously required layers of management.
What the essay offers, however, remains largely conceptual. The proposed “world model” and “intelligence layer” are described in broad strokes, with little explanation of architecture, governance, reliability, or regulatory constraints. Operational questions about accountability, conflict resolution, compliance, and system failure are absent. The document reads more like a vision statement than an implementation plan.
The timing also shapes how the narrative lands. Block had just eliminated roughly forty percent of its workforce while continuing to post strong profits and growth guidance. Presenting the layoffs as part of a transition toward an AI-coordinated organization reframes the decision as technological progress rather than a restructuring move aimed at improving margins. Investors responded enthusiastically, and the stock moved sharply upward after the announcement.
Viewed in that context, the essay functions as a narrative device as much as a strategic roadmap. It gives shareholders a forward-looking explanation for a dramatic workforce reduction and aligns the company with the broader excitement around AI. The language of intelligence layers and autonomous coordination creates the impression of a company inventing the future of work.
There is little evidence in the essay that these systems currently exist at the scale implied. Most examples remain hypothetical, and the operational mechanics behind them remain opaque.
For critics like me, that gap between narrative and implementation raises an uncomfortable possibility. The story helps stabilize investor confidence, softens the reputational impact of layoffs, and redirects attention away from regulatory issues that have affected the company in recent years. The result is a compelling piece of storytelling about an AI-native organization, delivered at a moment when the company benefits from exactly that story.
Whether the underlying systems eventually match the scale of the vision remains an open question. Right now the essay reads less like a concrete operating blueprint and more like a carefully timed narrative designed to protect the future of the company.
