Skip to content

Latest commit

 

History

History
317 lines (211 loc) · 20.9 KB

case-studies-page.md

File metadata and controls

317 lines (211 loc) · 20.9 KB

Case Studies Page

Attacks on machine learning (ML) systems are being developed and released with increased regularity. Historically, attacks against ML systems have been performed in a controlled academic settings, but as these case-studies demonstrate, attacks are being seen in-the-wild. In production settings ML systems are trained on personally identifiable information (PII), trusted to make critical decisions with little oversight, and have little to no logging and alerting attached to their use. The case-studies were selected because of the impact to production ML systems, and each demonstrates one of the following characteristics.

  1. Range of Attacks: evasion, poisoning, model replication and exploiting traditional software flaws.
  2. Range of Personas: Average user, Security researchers, ML Researchers and Fully equipped Red team.
  3. Range of ML Paradigms: Attacks on MLaaS, ML models hosted on cloud, hosted on-premise, ML models on edge.
  4. Range of Use case: Attacks on ML systems used in both "security-sensitive" applications like cybersecurity and non-security-sensitive applications like chatbots.

Evasion of Deep Learning detector for malware C&C traffic

Summary of Incident: Palo Alto Networks Security AI research team tested a deep learning model for malware command and control (C&C) traffic detection in HTTP traffic. Based on the publicly available paper by Le et al. [1] (open source intelligence), we built a model that was trained on a similar dataset as our production model and had performance similar to it. Then we crafted adversarial samples and queried the model and adjusted the adversarial sample accordingly till the model was evaded.

Mapping to Adversarial Threat Matrix:

  • The team trained the model on ~ 33 million benign and ~ 27 million malicious HTTP packet headers
  • Evaluation showed a true positive rate of ~ 99% and false positive rate of ~0.01%, on average
  • Testing the model with a HTTP packet header from known malware command and control traffic samples was detected as malicious with high confidence (> 99%).
  • The attackers crafted evasion samples by removing fields from packet header which are typically not used for C&C communication (e.g. cache-control, connection, etc.)
  • With the crafted samples the attackers performed online evasion of the ML based spyware detection model. The crafted packets were identified as benign with >80% confidence.
  • This evaluation demonstrates that adversaries are able to bypass advanced ML detection techniques, by crafting samples that are misclassified by an ML model.

Reported by:

  • Palo Alto Networks (Network Security AI Research Team)

Source:

  • [1] Le, Hung, et al. "URLNet: Learning a URL representation with deep learning for malicious URL detection." arXiv preprint arXiv:1802.03162 (2018).

Botnet Domain Generation Algorithm (DGA) Detection Evasion

Summary of Incident: Palo Alto Networks Security AI research team was able to bypass a Convolutional Neural Network (CNN)-based botnet Domain Generation Algorithm (DGA) detection [1] by domain name mutations. It is a generic domain mutation technique which can evade most ML-based DGA detection modules, and can also be used for testing against all DGA detection products in the security industry.

Mapping to Adversarial Threat Matrix:

  • DGA detection is a widely used technique to detect botnets in academia and industry. 
  • The researchers look into a publicly available CNN-based DGA detection model [1] and tested against a well-known DGA generated domain name data sets, which includes ~50 million domain names from 64 botnet DGA families.
  • The CNN-based DGA detection model shows more than 70% detection accuracy on 16 (~25%) botnet DGA families.
  • On the DGA generated domain names from 16 botnet DGA families, we developed a generic mutation technique that requires a minimum number of mutations, but achieves a very high evasion rate.
  • Experiment results show that, after only one string is inserted once to the DGA generated domain names, the detection rate of all 16 botnet DGA families can drop to less than 25% detection accuracy.
  • The mutation technique can evade almost all DGA detections, not limited to CNN-based DGA detection shown in this example. If the attackers add it on top of the existing DGA, most of the DGA detections might fail.
  • The generic mutation techniques can also be used to test the effectiveness and robustness of all DGA detection methods developed by security companies in the industry before it is deployed to the production environment.

Reported by:

  • Palo Alto Networks (Network Security AI Research Team)

Source:

  • [1] Yu, Bin, Jie Pan, Jiaming Hu, Anderson Nascimento, and Martine De Cock. "Character level based detection of DGA domain names." In 2018 International Joint Conference on Neural Networks (IJCNN), pp. 1-8. IEEE, 2018. Source code is available from Github: https://github.com/matthoffman/degas

VirusTotal Poisoning

Summary of Incident: An increase in reports of a certain ransomware family that was out of the ordinary was noticed. In investigating the case, it was observed that many samples of that particular ransomware family were submitted through a popular Virus-Sharing platform within a short amount of time. Further investigation revealed that based on string similarity, the samples were all equivalent, and based on code similarity they were between 98 and 74 percent similar. Interestingly enough, the compile time was the same for all the samples. After more digging, the discovery was made that someone used 'metame' a metamorphic code manipulating tool to manipulate the original file towards mutant variants. The variants wouldn't always be executable but still classified as the same ransomware family.

Mapping to Adversarial Threat Matrix:

  • The actor used a malware sample from a prevalent ransomware family as a start to create ‘mutant’ variants.
  • The actor uploaded ‘mutant’ samples to the platform.
  • Several vendors started to classify the files as the ransomware family even though most of them won’t run.
  • The ‘mutant‘ samples poisoned the dataset the ML model(s) use to identify and classify this ransomware family.

Reported by:

  • Christiaan Beek (@ChristiaanBeek) - McAfee ATR Team

Source:

  • McAfee Advanced Threat Research

Bypassing Cylance's AI Malware Detection

Summary of Incident: Researchers at Skylight were able to create a universal bypass string that when appended to a malicious file evades detection by Cylance's AI Malware detector.

Mapping to Adversarial Threat Matrix :

  • The researchers read publicly available information and enabled verbose logging to understand the inner workings of the ML model, particularly around reputation scoring.
  • The researchers reverse-engineered the ML model to understand which attributes provided what level of positive or negative reputation. Along the way, they discovered a secondary model which was an override for the first model. Positive assessments from the second model overrode the decision of the core ML model.
  • Using this knowledge, the researchers fused attributes of known good files with malware. Due to the secondary model overriding the primary, the researchers were effectively able to bypass the ML model.

Cylance

Reported by: Research and work by Adi Ashkenazy, Shahar Zini, and SkyLight Cyber team. Notified to us by Ken Luu (@devianz_)

Source:


Camera Hijack Attack on Facial Recognition System

Summary of Incident: This type of attack can break through the traditional live detection model and cause the misuse of face recognition.

Mapping to Adversarial Threat Matrix:

  • The attackers bought customized low-end mobile phones, customized android ROMs, a specific virtual camera application, identity information and face photos.
  • The attackers used software to turn static photos into videos, adding realistic effects such as blinking eyes. Then the attackers use the purchased low-end mobile phone to import the generated video into the virtual camera app.
  • The attackers registered an account with the victim's identity information. In the verification phase, the face recognition system called the camera API, but because the system was hooked or rooted, the video stream given to the face recognition system was actually provided by the virtual camera app.
  • The attackers successfully evaded the face recognition system and impersonated the victim.

Reported by:

  • Henry Xuef, Ant Group AISEC Team

Source:

  • Ant Group AISEC Team

Attack on Machine Translation Service - Google Translate, Bing Translator, and Systran Translate

Summary of Incident: 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 an "imitation model" with near-production, state-of-the-art translation quality. Beyond demonstrating that IP can be stolen from a black-box system, they used the imitation 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.

Mapping to Adversarial Threat Matrix:

  • Using published research papers, the researchers gathered similar datasets and model architectures that these translation services used
  • They abuse a public facing application to query the model and produce machine translated sentence pairs as training data
  • Using these translated sentence pairs, researchers trained a substitute model (model replication)
  • The replicated models were used to construct offline adversarial examples that successfully transferred to an online evasion attack

Reported by:

  • Work by Eric Wallace, Mitchell Stern, Dawn Song and reported by Kenny Song (@helloksong)

Source:


ClearviewAI Misconfiguration

Summary of Incident: Clearview AI's source code repository, though password protected, was misconfigured to allow an arbitrary user to register an account. This allowed an external researcher to gain access to a private code repository that contained Clearview AI production credentials, keys to cloud storage buckets containing 70K video samples, and copies of its applications and Slack tokens. With access to training data, a bad-actor has the ability to cause an arbitrary misclassification in the deployed model.

Mapping to Adversarial Threat Matrix :

  • In this scenario, a security researcher gained initial access to via a "Valid Account" that was created through a misconfiguration. No Adversarial ML techniques were used.
  • These kinds of attacks illustrate that any attempt to secure ML system should be on top of "traditional" good cybersecurity hygiene such as locking down the system with least privileges, multi-factor authentication and monitoring and auditing.

ClearviewAI

Reported by:

  • Mossab Hussein (@mossab_hussein)

Source:


GPT-2 Model Replication

Summary of Incident: : OpenAI built GPT-2, a powerful natural language model and adopted a staged-release process to incrementally release 1.5 Billion parameter model. Before the 1.5B parameter model could be released by OpenAI eventually, two ML researchers replicated the model and released it to the public. Note this is an example of model replication NOT model model extraction. Here, the attacker is able to recover a functionally equivalent model but generally with lower fidelity than the original model, perhaps to do reconnaissance (See ProofPoint attack). In Model extraction, the fidelity of the model is comparable to the original, victim model.

Mapping to Adversarial Threat Matrix :

  • Using public documentation about GPT-2, ML researchers gathered similar datasets used during the original GPT-2 training.
  • Next, they used a different publicly available NLP model (called Grover) and modified Grover's objective function to reflect GPT-2's objective function.
  • The researchers then trained the modified Grover on the dataset they curated, using Grover's initial hyperparameters, which resulted in their replicated model.

GPT2_Replication

Reported by:

  • Vanya Cohen (@VanyaCohen)
  • Aaron Gokaslan (@SkyLi0n)
  • Ellie Pavlick
  • Stefanie Tellex

Source:


ProofPoint Evasion

Summary of Incident: : CVE-2019-20634 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 evade the live system.

Mapping to Adversarial Threat Matrix :

  • The researchers first gathered the scores from the Proofpoint's ML system used in email email headers.
  • Using these scores, the researchers replicated the ML mode by building a "shadow" aka copy-cat ML model.
  • Next, the ML researchers algorithmically found samples that this "offline" copy cat model.
  • Finally, these insights from the offline model allowed the researchers to create malicious emails that received preferable scores from the real ProofPoint email protection system, hence bypassing it.

PFPT_Evasion

Reported by:

  • Will Pearce (@moo_hax)
  • Nick Landers (@monoxgas)

Source:


Tay Poisoning

Summary of Incident: Microsoft created Tay, a twitter chatbot for 18- to 24- year-olds in the U.S. for entertainment purposes. Within 24 hours of its deployment, Tay had to be decommissioned because it tweeted reprehensible words.

Mapping to Adversarial Threat Matrix :

  • Tay bot used the interactions with its twitter users as training data to improve its conversations.
  • Average users of Twitter coordinated together with the intent of defacing Tay bot by exploiting this feedback loop.
  • As a result of this coordinated attack, Tay's training data was poisoned which led its conversation algorithms to generate more reprehensible material.

Tay_Poisoning

Source:


Microsoft - Azure Service

Summary of Incident: : The Azure Red Team and Azure Trustworthy ML team performed a red team exercise on an internal Azure service with the intention of disrupting its service.

Reported by: Microsoft Mapping to Adversarial Threat Matrix :

  • The team first performed reconnaissance to gather information about the target ML model.
  • Then, using a valid account the team found the model file of the target ML model and the necessary training data.
  • Using this, the red team performed an offline evasion attack by methodically searching for adversarial examples.
  • Via an exposed API interface, the team performed an online evasion attack by replaying the adversarial examples, which helped achieve this goal.
  • This operation had a combination of traditional ATT&CK enterprise techniques such as finding Valid account, and Executing code via an API -- all interleaved with adversarial ML specific steps such as offline and online evasion examples.

MS_Azure

Reported by:

  • Microsoft (Azure Trustworthy Machine Learning)

Source:

  • None

Microsoft - Edge AI

Summary of Incident: The Azure Red Team performed a red team exercise on a new Microsoft product designed for running AI workloads at the Edge.

Mapping to Adversarial Threat Matrix:

  • The team first performed reconnaissance to gather information about the target ML model.
  • Then, used a publicly available version of the ML model, started sending queries and analyzing the responses (inferences) from the ML model.
  • Using this, the red team created an automated system that continuously manipulated an original target image, that tricked the ML model into producing incorrect inferences, but the perturbations in the image were unnoticeable to the human eye.
  • Feeding this perturbed image, the red team was able to evade the ML model into misclassifying the input image.
  • This operation had one step in the traditional MITRE ATT&CK techniques to do reconnaissance on the ML model being used in the product, and then the rest of the techniques was to use Offline evasion, followed by online evasion of the targeted product.

alt_text

Reported by: Microsoft

Source: None


MITRE - Physical Adversarial Attack on Face Identification

Summary of Incident: MITRE’s AI Red Team demonstrated a physical-domain evasion attack on a commercial face identification service with the intention of inducing a targeted misclassification.

Mapping to Adversarial Threat Matrix:

  • The team first performed reconnaissance to gather information about the target ML model.
  • Using a valid account, the team identified the list of IDs targeted by the model.
  • The team developed a proxy model using open source data.
  • Using the proxy model, the red team optimized a physical domain patch-based attack using an expectation of transformations.
  • Via an exposed API interface, the team performed an online physical-domain evasion attack including the adversarial patch in the input stream which resulted in a targeted misclassification.
  • This operation had a combination of traditional ATT&CK enterprise techniques such as finding Valid account, and Executing code via an API – all interleaved with adversarial ML specific attacks.

mitre

Reported by: MITRE AI Red Team

Source: None


Contributing New Case Studies

We are especially excited for new case-studies! We look forward to contributions from both industry and academic researchers. Before submitting a case-study, consider that the attack:

  1. Exploits one or more vulnerabilities that compromises the confidentiality, integrity or availability of ML system.
  2. The attack was against a production, commercial ML system. This can be on MLaaS like Amazon, Microsoft Azure, Google Cloud AI, IBM Watson etc or ML systems embedded in client/edge.
  3. You have permission to share the information/published this research. Please follow the proper channels before reporting a new attack and make sure you are practicing responsible disclosure.

You can email [email protected] with summary of the incident, Adversarial ML Threat Matrix mapping and associated resources.