Full-Stack AI Is a Lease
(14)
Overview
Full-stack AI vendors offer speed and simplicity, the same pitch as a car lease. For a regulated institution, the exit clause matters more than the monthly payment.
Year
2026
Industry
Financial Services / Governance-Heavy Multinationals
Challenge
A regulated institution evaluating its AI strategy usually gets pitched the same story: buy the full stack, model, infrastructure, and interface, from one vendor, and move fast. It's a reasonable pitch. It's also a lease. A lease gets you into the car today, with less friction than buying outright. The dealer sets the terms, handles the maintenance, and the arrangement works well right up until it doesn't, the day the leasing company changes its pricing, restricts a feature, gets acquired, or, in AI's case, suddenly can't serve your jurisdiction because of an export control decision that has nothing to do with you. At that point you're not upgrading. You're renegotiating your entire operation from zero leverage, because you never owned any piece of what you built. Buying a car with standard, swappable parts works differently. If one supplier stops making a part, you replace that part. You don't replace the car.
Impact
For financial institutions specifically, this isn't a hypothetical, it's a named regulatory category. The EU's DORA framework requires financial institutions to avoid over-reliance on any single critical ICT provider and to maintain a documented exit strategy. Swiss regulatory guidance on outsourcing and operational resilience runs on the same logic. A bank going full-stack with one AI vendor isn't buying safety, it's building the exact concentration risk its own regulator has already flagged as a supervisory concern. That's the piece the leasing pitch leaves out: the buyer with the most to lose from a bad exit is usually the buyer being sold hardest on not needing one.