AI Comes for the If Statement: Specialized Deciders Cut Classification Cost ~100x

Tomer Tunguz 谈 AI 优化 if-then 判断：专用决策器把分类成本降近百倍

Tomasz Tunguz tested Jev and SemIf on 98 production emails: classification accuracy jumped from 47% to over 80%, cost dropped to $0.0004 per call—76x to 209x cheaper than frontier models. These deciders skip text generation, run attention once, and output choice probabilities in hundreds of milliseconds. Tunguz sees this as a bifurcation: frontier models for discovery, specialized models for production, with if-then as the first optimized programming primitive.

Why it matters: Tunguz ran a real production test with concrete accuracy and cost numbers—not just trend talk. The piece flags a meaningful fork in AI infra: general-purpose generators vs. specialized deciders. Score stays at 78 because both tools are brand-new with no large-scale validation ...