François Chollet: Scaling Is a Multiplier, Not the Core Method
The argument reframes the scaling debate: compute spending amplifies whatever generalization ability a method already has, so research effort should target methods with higher baseline generalization rather than relying on budget. Researchers choosing between architecture-driven and scaling-driven research agendas, and investors weighing AI infrastructure bets, can use this framing to evaluate long-term defensibility of current approaches.
François Chollet argued on X on 15 Jul 2026 that data and compute scaling act only as multipliers on an approach's generalization power. He said trillions of dollars in investment can make low-generalization approaches look adequate without making them the end state of AI. He advised researchers to work on methods that can scale, since lacking the multiplier makes competing impossible. He added that a cleverer approach will scale better than a brute-force one.
Data scaling and compute scaling are *multipliers* for the generalization power of an approach. Trillions of dollars in investment can make very inefficient, low-generalization approaches look adequate, but that doesn't make them the ultimate end-all-be-all of AI.
You should work on methods that *can* scale, because if you don't benefit from this multiplier you simply can't compete. But you should keep in mind scaling is only a multiplier: a cleverer approach will scale *better*.
