Visa cuts 2,600 jobs and the tell isn't AI — it's that product and engineering took the hit

Visa is preparing to cut about 2,600 workers, roughly 7% of its approximately 34,000 employees, the company confirmed to Fast Company after Bloomberg and others reported the plan. The cuts were disclosed internally in a memo from CEO Ryan McInerney, whose accuracy Visa also confirmed. Per that memo, the reductions fall primarily on product and technology teams, and McInerney framed them around a "deep conviction that we are doing what is right for Visa, our clients and our partners as we continue to focus on driving efficiency across the company in order to reinvest in our highest potential opportunities."

Read the attribution carefully, because it matters. The word "AI" in the coverage is largely framing and context — an industry-wide push toward AI-driven efficiency, plus a payments landscape being pressured by smaller, nimbler fintechs. The quoted memo language is about efficiency and reinvestment. Bloomberg reports the savings are headed into commercial and money-movement solutions, consumer payments, and value-added services including stablecoins. That is not a company saying software replaced its people. That is a company moving headcount budget from one bet to another.

The genuinely interesting detail is where the axe fell. The dominant narrative about AI and employment — one we have argued ourselves across nine sector analyses — is that back-office and administrative work is the most exposed, while judgment, relationships and physical presence hold up. Visa cut product and technology. Whatever else that is, it is not the back office. Either the exposure curve is broader than the consensus assumes and AI coding tools are compressing engineering org charts faster than expected, or — the reading we find more likely — Visa is reorganizing around stablecoins and new rails and simply needs different engineers than the ones it has.

Our reading: treat "AI layoffs" as a claim requiring evidence, not as a category. We have tracked how often the AI justification fails to survive contact with reality — more than a third of AI-attributed layoffs end in rehiring, with Ford reversing roughly 300 roles on quality grounds, the Commonwealth Bank of Australia walking back about 45 in customer service, and Klarna reversing course publicly. AI is a useful thing to gesture at when announcing cuts, because it reads to investors as forward-looking rather than defensive. Visa, to its credit, mostly didn't gesture. The label is being applied from outside.

What this does confirm is our banking and payments thesis in a harder form than we wrote it. The value in payments is migrating from headcount to software and from legacy rails to programmable money, and the incumbent with 34,000 employees and enormous fixed advantages still has to keep buying its way into the next platform shift. That transition is real, the disruption to 2,600 households is real today, and the abundance argument does nothing for anyone getting a severance package this quarter. The honest version is that reallocation like this is how capital finds the next thing — and the test of whether it worked is not the size of the cut but whether Visa is rehiring in twelve months.