Middle Management Gets Absolutely Wrecked: Dorsey's Unhinged Claim That AI Can Do Your Job Better Than You

Jack Dorsey thinks middle managers should probably start updating their LinkedIn profiles—or, you know, learn to code. In a new essay titled "From Hierarchy to Intelligence" co-authored with Sequoia Capital managing partner Roelof Botha, the Block founder argues that his company's decision to cut approximately 4,000 of its more than 10,000 employees wasn't just another boring cost-cutting exercise—it was apparently a vision quest to replace middle management with our robot overlords.

The essay's core thesis hits harder than a 3x leverage degen: corporate hierarchy exists to solve one fundamental problem—routing information through organizations too large for any single person to manage. Managers aggregate context from below, relay directives from above, and keep teams aligned. According to Dorsey and Botha, AI can now handle these functions continuously and at scale, making the messenger permanently redundant. Sorry, Karen from Product—your bridge-burning skills have officially been automated.

The proposed replacement involves two AI-driven "world models" that sound suspiciously like Skynet with a fintech coat of paint. One aggregates internal data from code, decisions, workflows, and performance metrics to create a living picture of company operations—essentially absorbing the institutional knowledge managers traditionally carried in their heads. The second maps customer and merchant behavior using transaction data from Cash App and Square. Together, they're basically building a corporate brain that never sleeps, never takes vacation, and definitely never brings donuts to the Monday meeting.

These models feed what Block calls an "intelligence layer" that dynamically composes financial products based on market demand. Rather than building from fixed roadmaps written by humans who probably just wanted to ship the feature and go home, the essay proposes breaking Block's business into modular capabilities: payments, lending, card issuance, and payroll. When the system spots a need—say, a merchant facing seasonal cash flow issues—it assembles a solution from existing capabilities. When it can't, that gap defines what gets built next. It's like LEGO for banking, except the instruction manual is written by an AI that has never experienced the joy of stepping on a plastic brick at 2 AM.

Dorsey told Wired in early March that the restructuring was triggered by a capability shift he observed in December, citing improvements in Anthropic's Opus 4.6 and OpenAI's Codex 5.3, which he said could now operate effectively in large codebases. Not everyone is convinced, obviously. Current and former Block employees told The Guardian that roughly 95% of AI-generated code changes still require human modification, and that AI tools can't yet lead in regulated areas like banking and money transfers—because apparently even our robot friends draw the line at money laundering compliance.