Threat intelligence / Technique

Train Proxy via Replication

Adversaries may replicate a private model. By repeatedly querying the victim's [AI Model Inference API Access](/techniques/AML.T0040), the adversary can collect the target model's inferences into a dataset. The inferences are used as labels for training a separate model offline that will mimic the behavior and performance of the target model. A replicated model that closely mimics the target model is a valuable resource in staging the attack. The adversary can use the replicated model to [Craft Adversarial Data](/techniques/AML.T0043) for various purposes (e.g. [Evade AI Model](/techniques/AML.T0015), [Spamming AI System with Chaff Data](/techniques/AML.T0046)).

Technique

TypeTechnique

External ID—

First seen2026-08-14

Last seen2026-08-14

Confidence

—

Adversaries may replicate a private model. By repeatedly querying the victim's [AI Model Inference API Access](/techniques/AML.T0040), the adversary can collect the target model's inferences into a dataset. The inferences are used as labels for training a separate model offline that will mimic the behavior and performance of the target model. A replicated model that closely mimics the target model is a valuable resource in staging the attack. The adversary can use the replicated model to [Craft Adversarial Data](/techniques/AML.T0043) for various purposes (e.g. [Evade AI Model](/techniques/AML.T0015), [Spamming AI System with Chaff Data](/techniques/AML.T0046)).

6 recorded

Relationships

mitigates

from AI Telemetry Logging

Mitigation

mitigates

from Limit AI Service Query Volume and Rate

Mitigation

mitigates

from Predictive AI Output Obfuscation

Mitigation

subtechnique-of

to Create Proxy AI Model

Technique

uses

from Attack on Machine Translation Services

Campaign

uses

from ProofPoint Evasion

Campaign

1 external

References

mitre-atlas

AML.T0005.001