APromptRiskDBThreat intelligence atlas
AI Security Technique

Black-Box Transfer - AI Security Technique

In Black-Box Transfer attacks, the adversary uses one or more proxy models (trained via Create Proxy AI Model or Train Proxy via Replication) they have full access to and are representative of the target model. The adversary uses White-Box Optimization on the proxy models to generate adversarial examples. If the set of proxy models...

AI Security Techniquedemonstrated

Record summary

A quick snapshot of what this page covers.

Tactics0Attacker goals connected to this method.
Mitigations4Defenses that may help against this attack.
AI risks0Research-backed risks connected to this topic.

Attack context

How this AI attack works in practice.

In Black-Box Transfer attacks, the adversary uses one or more proxy models (trained via Create Proxy AI Model or Train Proxy via Replication) they have full access to and are representative of the target model. The adversary uses White-Box Optimization on the proxy models to generate adversarial examples. If the set of proxy models are close enough to the target model, the adversarial example should generalize from one to another. This means that an attack that works for the proxy models will likely then work for the target model. If the adversary has AI Model Inference API Access, they may use Verify Attack to confirm the attack is working and incorporate that information into their training process.

ATLAS ID
AML.T0043.002
Priority score
62
Maturity: demonstrated

Mitigations

Defenses that may help against this attack.

AML.M0015 - Adversarial Input Detection

Data PreparationML Model Engineering+3 more
LifecycleData Preparation + 4 moreCategoryTechnical - ML

Incorporate adversarial input detection to block malicious inputs at inference time.

AML.M0010 - Input Restoration

Data PreparationML Model Evaluation+2 more
LifecycleData Preparation + 3 moreCategoryTechnical - ML

Input restoration can help remediate adversarial inputs.

AML.M0003 - Model Hardening

Data PreparationML Model Engineering
LifecycleData Preparation + 1 moreCategoryTechnical - ML

Hardened models are more robust to adversarial inputs.

AML.M0006 - Use Ensemble Methods

ML Model Engineering
LifecycleML Model EngineeringCategoryTechnical - ML

Using an ensemble of models increases the difficulty of crafting effective adversarial data and improves overall robustness.

Case studies

Examples from public reports and exercises.

Confusing Antimalware Neural Networks

exercise
Date2021-06-23

Cloud storage and computations have become popular platforms for deploying ML malware detectors. In such cases, the features for models are built on users' systems and then sent to cybersecurity company servers. The Kaspersky ML research team explored this gray-box scenario and showed that feature knowledge is enough for an adversarial attack on ML models.

They attacked one of Kaspersky's antimalware ML models without white-box access to it and successfully evaded detection for most of the adversarially modified malware files.

Attack on Machine Translation Services

exercise
Date2020-04-30

Machine translation services (such as Google Translate, Bing Translator, and Systran Translate) provide public-facing UIs and APIs. A research group at UC Berkeley utilized these public endpoints to create a replicated model with near-production state-of-the-art translation quality. Beyond demonstrating that IP can be functionally stolen from a black-box system, they used the replicated model to successfully transfer adversarial examples to the real production services. These adversarial inputs successfully cause targeted word flips, vulgar outputs, and dropped sentences on Google Translate and Systran Translate websites.

ProofPoint Evasion

exercise
Date2019-09-09

Proof Pudding (CVE-2019-20634) is a code repository that describes how ML researchers evaded ProofPoint's email protection system by first building a copy-cat email protection ML model, and using the insights to bypass the live system. More specifically, the insights allowed researchers to craft malicious emails that received preferable scores, going undetected by the system. Each word in an email is scored numerically based on multiple variables and if the overall score of the email is too low, ProofPoint will output an error, labeling it as SPAM.

Source

Where this page information comes from.