Skip to content
AImpact
IT EN
Safety Advanced Also known as: Estrazione del modello · Model stealing

Model extraction

An attack where an adversary repeatedly queries a model via API to reconstruct a functional copy of its weights or behavior.

ShareLinkedInX

In practice

A legal variant is distilling outputs of a frontier model to train a smaller one, banned by the terms of service of most providers. Mitigated with rate limits, watermarking, and fingerprint detection.

Related terms

Distillation Open weights vs open source

Seen in the wild

0 entries mentioning it

No archive entry mentions it explicitly. Appears in broader contexts.

← All terms