Continued AI Scaling Requires Repeated Efficiency Doublings

TL;DR

This paper emphasizes that ongoing AI scaling depends on continuous efficiency improvements across hardware, algorithms, and systems, akin to Moore's Law, to sustain logical compute growth at manageable costs.

Contribution

It reframes AI scaling laws to focus on efficiency gains in physical resources, highlighting the necessity of repeated improvements for sustained progress.

Findings

Efficiency doublings are essential for continued AI scaling.

Recent trends suggest Moore-like or faster efficiency improvements.

Operational burden increases as diminishing returns set in.

Abstract

This paper argues that continued AI scaling requires repeated efficiency doublings. Classical AI scaling laws remain useful because they make progress predictable despite diminishing returns, but the compute variable in those laws is best read as logical compute, not as a record of one fixed physical implementation. Practical burden therefore depends on the efficiency with which physical resources realize that compute. Under that interpretation, diminishing returns mean rising operational burden, not merely a flatter curve. Sustained progress then requires recurrent gains in hardware, algorithms, and systems that keep additional logical compute feasible at acceptable cost. The relevant analogy is Moore's Law, understood less as a theorem than as an organizing expectation of repeated efficiency improvement. AI does not yet have a single agreed cadence for such gains, but recent evidence suggests trends that are at least Moore-like and sometimes faster. The paper's claim is therefore simple: if AI scaling is to remain active, repeated efficiency doublings are not optional. They are required.