Qualification Before Expansion: Auditable Synthetic-Only Class Addition

Synthetic data can make a new class trainable before any real example exists, but trainability does not show that the expanded classifier will recognize the class in deployment. We formulate pre-deployment qualification: deciding, from historical classes and candidate synthetic data alone, whether the evidence supports admitting a class. Qualification-Guided Synthetic Class Expansion (Q-SCE) replays each historical class as a pseudo-missing candidate under the same information budget, tests whether candidate-observable descriptors predict revealed real recall better than a historical-mean reference, and abstains when they do not. We froze one instantiation, a four-descriptor Ridge qualifier with task, support, and retention gates and an empirical lower-score tier, before opening any candidate target data on DomainNet Clipart→Real (45 candidates) and Stable-Diffusion→Caltech-256 (101). Candidate-recall MAE falls from .259 to .224 and from .138 to .111. Admission gains are small. On Caltech the policy admits 78 classes with one failure, a detector-p10≥ .50 screen does as well, and CLIP zero-shot already succeeds on 99 classes; on DomainNet the policy admits one class and abstains on the rest. The results support replay-based qualification with frozen abstention. They do not show a better ranker than simple scores or a guarantee on the failure rate among admitted classes.