April 02, 2026 ChainGPT

Block Cuts 4,000 Jobs as Dorsey Bets AI Will Replace Middle Managers, Shift to Composable Fintech

Block Cuts 4,000 Jobs as Dorsey Bets AI Will Replace Middle Managers, Shift to Composable Fintech
Headline: Jack Dorsey says AI will replace middle managers as Block trims 4,000 jobs — here’s how the company plans to work Jack Dorsey and Sequoia partner Roelof Botha argue in a new essay, “From Hierarchy to Intelligence,” that Block’s recent cuts — about 4,000 roles out of a workforce of just over 10,000 — are not merely cost-cutting but a structural pivot: replacing layers of middle management with AI-driven systems that coordinate work and compose products. Why managers exist — and why AI could make them redundant The essay frames corporate hierarchy as a historical solution to a single problem: routing information in organizations too big for one person to understand. Managers collect context from teams, relay strategy from leadership and keep groups aligned. Dorsey and Botha say modern AI can perform those functions continuously and at scale, eliminating the need for many “messengers” and middle layers. Two “world models” and an intelligence layer Block plans to swap management layers for two complementary AI models that feed a centralized “intelligence layer”: - An internal model that continuously ingests code, decisions, workflows and performance metrics to recreate the context managers traditionally held. - An external model that maps merchant and customer behavior using transaction flows from Cash App and Square. Together, those models would let Block dynamically assemble financial products to match market needs — not by following a fixed roadmap, but by composing modular capabilities on demand. Product development becomes composable Instead of long, static roadmaps, Block would break the business into modular capabilities — payments, lending, card issuance, payroll, etc. When the system recognizes a user need (for example, a merchant facing a seasonal cash-flow gap), it assembles a tailored solution from existing modules. If a needed capability doesn’t exist, that gap becomes the highest-priority item in a system-generated backlog rather than a human-crafted product plan. A smaller, flatter org — three roles With coordination absorbed by AI, Block says it will slim organizational structure and operate with three role types: - Individual contributors who build and maintain the system - Directly responsible individuals who own specific outcomes on 90-day cycles - Player-coaches who remain hands-on while mentoring others What Dorsey points to — and the pushback Dorsey told Wired the move was prompted by a capability leap he noticed in December in tools like Anthropic’s Opus 4.6 and OpenAI’s Codex 5.3, which he said now handle large codebases effectively. But current and former Block employees told the Guardian that the reality is messier: roughly 95% of AI-generated code changes still require human modification, and existing AI tools are not yet ready to take the lead in highly regulated domains such as banking and money transfers. Implications for fintech and crypto For crypto and fintech audiences, Block’s vision signals a future where product composability and automated decisioning become central — reminiscent of smart-contract-driven composability in Web3. But the implementation hurdles flagged by employees, and the regulatory realities around money movement, suggest the transition will be gradual and hybrid: human oversight where reliability and compliance matter, AI automation where scale and pattern recognition deliver value. Bottom line: Block is betting on AI to replace coordination and product-planning functions traditionally carried out by middle managers. The idea could reshape how financial products are built and teams are organized — but technical limits and regulatory constraints mean the replacement is likely to be evolutionary rather than instantaneous. Read more AI-generated news on: undefined/news