Visa cut two thousand six hundred jobs, seven percent of its global workforce. The reduction concentrated in technology and product. The official explanation talks about operational efficiency and strategic reorganization. The internal communication — the kind that doesn't chase headlines — acknowledges that AI adoption is accelerating the transformation of work inside the company. That's the detail that matters.
Corporate efficiency means reorganizing human, technical, and financial resources to produce the same output with fewer inputs, because talent remains the highest cost in any tech organization. Visa processes transactions at planetary scale because it automated first what others still did by hand. Now it's applying that same logic to its own structure.
The dominant thesis holds that these cuts are part of a normal business cycle: organizations adjust headcount as markets and priorities shift. The argument contains genuine truth. Some layoffs respond to regulation, mergers, or falling demand. Treating every cut as automatic proof of technological replacement would be dishonest.
Companies avoid naming AI openly for clear reasons of reputational and legal risk. None of them wants to be the first to declare that a model now does the work of its product engineers. Silence reduces uncomfortable questions about accountability and transition. It makes sense from their perspective.
What happens when the same pattern shows up across dozens of companies at once? Microsoft, IBM, and SAP executed significant cuts in recent quarters while expanding their generative AI capabilities and reporting growth. These are not isolated cases. Oracle laid off thirty thousand people after a record quarter while Larry Ellison publicly stated that AI already writes corporate code. Conglomerates tied to Sam Altman and Elon Musk added up to one hundred forty-two thousand layoffs while investing seven hundred billion dollars in AI infrastructure and promoting universal basic income as the remedy.
That question exposes the conflict of interest: the same entities causing the displacement later propose the social solutions on their own terms. The language repeats with surgical precision — efficiency, reorganization, strategic focus — but no press release states that generative AI systems replace functions because they're cheaper and scale better. The rhetorical smoothness conceals power dynamics that turn out rough for whoever loses the job.
Unemployment hurts two to three times more than inflation, according to decades of well-being research. Not just because of the economic loss: it breaks belonging and the sense of contribution. Marie Jahoda, Paul Lazarsfeld, and Hans Zeisel documented this in the Marienthal study nearly a century ago. When a community loses its main source of work, time, identity, and social bonds come apart. Knowing the real cause, however painful, allows the loss to be processed with dignity. Denying it amplifies the damage.
This omission isn't carelessness. It's a deliberate narrative-management strategy while corporations write the actual rules for implementing AI within their labor structures. Naming the cause would open debates about funded retraining and shared responsibility. The silence preserves the regulatory vacuum they themselves help maintain.
This connects to the findings in The Generosity in the Doorway, which explores why decentralized coordination — without intermediaries dominating the flow of information — is a precondition for any real alternative. The organizations doing the automating are rarely the ones funding the social remedy on fair terms.
I'm still not sure whether requiring companies to explicitly name AI as the cause would change the practice beyond the symbolic. It could turn into pure theater. Still, the current opacity serves a concrete function: it keeps society from perceiving the structural scale of the transformation happening simultaneously across entire sectors.
The historical record of the industrial revolution shows something similar. Factory owners also avoided naming the new machines as the direct cause of mass unemployment, preferring to talk about "productive improvements." Stones don't lie, and the parallel confirms that narrative management was always part of the mechanism.
In the same quarter, Visa reported revenue growth. This isn't a company in crisis laying people off to survive. It's a highly profitable company laying people off to become even more profitable, with technology that — according to its own internal memos — already performs the work those two thousand six hundred people used to do. Efficiency isn't being denied. What's being denied is at whose expense it's achieved.
What would happen if society demanded that technological causes be named without euphemism before accepting that this transformation be presented as inevitable?
