Stealing AI Model Weights Through Covert Communication Channels

• URL: http://arxiv.org/abs/2510.00151v1
• Date: Tue, 30 Sep 2025 18:21:41 GMT
• Title: Stealing AI Model Weights Through Covert Communication Channels
• Abstract summary: We present a novel attack targeting wireless devices equipped with AI hardware accelerators.<n>In the first phase, the victim's device is compromised with a hardware Trojan.<n>In the second phase, the adversary uses a nearby wireless device to intercept the victim's transmission frames.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: AI models are often regarded as valuable intellectual property due to the high cost of their development, the competitive advantage they provide, and the proprietary techniques involved in their creation. As a result, AI model stealing attacks pose a serious concern for AI model providers. In this work, we present a novel attack targeting wireless devices equipped with AI hardware accelerators. The attack unfolds in two phases. In the first phase, the victim's device is compromised with a hardware Trojan (HT) designed to covertly leak model weights through a hidden communication channel, without the victim realizing it. In the second phase, the adversary uses a nearby wireless device to intercept the victim's transmission frames during normal operation and incrementally reconstruct the complete weight matrix. The proposed attack is agnostic to both the AI model architecture and the hardware accelerator used. We validate our approach through a hardware-based demonstration involving four diverse AI models of varying types and sizes. We detail the design of the HT and the covert channel, highlighting their stealthy nature. Additionally, we analyze the impact of bit error rates on the reception and propose an error mitigation technique. The effectiveness of the attack is evaluated based on the accuracy of the reconstructed models with stolen weights and the time required to extract them. Finally, we explore potential defense mechanisms.

Related papers

• Attacking AI Accelerators by Leveraging Arithmetic Properties of Addition [0.0]
  We present a new hardware aging attack that exploits commutative properties of addition to disrupt the multiply-and-add operation.<n>By permuting the inputs of an adder, the attack preserves functional correctness while inducing unbalanced stress among transistors, accelerating delay degradation in the circuit.
  arXiv  Detail & Related papers  (2026-03-28T23:03:27Z)
• Cuckoo Attack: Stealthy and Persistent Attacks Against AI-IDE [64.47951172662745]
  Cuckoo Attack is a novel attack that achieves stealthy and persistent command execution by embedding malicious payloads into configuration files.<n>We formalize our attack paradigm into two stages, including initial infection and persistence.<n>We contribute seven actionable checkpoints for vendors to evaluate their product security.
  arXiv  Detail & Related papers  (2025-09-19T04:10:52Z)
• Attacking Attention of Foundation Models Disrupts Downstream Tasks [18.92561703051693]
  Foundation models are large models, trained on broad data that deliver high accuracy in many downstream tasks.<n>These models are vulnerable to adversarial attacks.<n>This paper studies the vulnerabilities of vision foundation models, focusing specifically on CLIP and ViTs.<n>We introduce a novel attack, targeting the structure of transformer-based architectures in a task-agnostic fashion.
  arXiv  Detail & Related papers  (2025-06-03T19:42:48Z)
• MISLEADER: Defending against Model Extraction with Ensembles of Distilled Models [56.09354775405601]
  Model extraction attacks aim to replicate the functionality of a black-box model through query access.<n>Most existing defenses presume that attacker queries have out-of-distribution (OOD) samples, enabling them to detect and disrupt suspicious inputs.<n>We propose MISLEADER, a novel defense strategy that does not rely on OOD assumptions.
  arXiv  Detail & Related papers  (2025-06-03T01:37:09Z)
• Vulnerabilities in AI-generated Image Detection: The Challenge of Adversarial Attacks [39.524974831780874]
  We show that adversarial attack is truly a real threat to AIGI detectors, because FPBA can deliver successful black-box attacks.<n>We name our method as Frequency-based Post-train Bayesian Attack, or FPBA.
  arXiv  Detail & Related papers  (2024-07-30T14:07:17Z)
• Disarming Steganography Attacks Inside Neural Network Models [4.750077838548593]
  We propose a zero-trust prevention strategy based on AI model attack disarm and reconstruction. We demonstrate a 100% prevention rate while the methods introduce a minimal decrease in model accuracy based on Qint8 and K-LRBP methods.
  arXiv  Detail & Related papers  (2023-09-06T15:18:35Z)
• Isolation and Induction: Training Robust Deep Neural Networks against Model Stealing Attacks [51.51023951695014]
  Existing model stealing defenses add deceptive perturbations to the victim's posterior probabilities to mislead the attackers. This paper proposes Isolation and Induction (InI), a novel and effective training framework for model stealing defenses. In contrast to adding perturbations over model predictions that harm the benign accuracy, we train models to produce uninformative outputs against stealing queries.
  arXiv  Detail & Related papers  (2023-08-02T05:54:01Z)
• AdIoTack: Quantifying and Refining Resilience of Decision Tree Ensemble Inference Models against Adversarial Volumetric Attacks on IoT Networks [1.1172382217477126]
  We present AdIoTack, a system that highlights vulnerabilities of decision trees against adversarial attacks. To assess the model for the worst-case scenario, AdIoTack performs white-box adversarial learning to launch successful volumetric attacks. We demonstrate how the model detects all non-adversarial volumetric attacks on IoT devices while missing many adversarial ones.
  arXiv  Detail & Related papers  (2022-03-18T08:18:03Z)
• The Feasibility and Inevitability of Stealth Attacks [63.14766152741211]
  We study new adversarial perturbations that enable an attacker to gain control over decisions in generic Artificial Intelligence systems. In contrast to adversarial data modification, the attack mechanism we consider here involves alterations to the AI system itself.
  arXiv  Detail & Related papers  (2021-06-26T10:50:07Z)
• An Empirical Review of Adversarial Defenses [0.913755431537592]
  Deep neural networks, which form the basis of such systems, are highly susceptible to a specific type of attack, called adversarial attacks. A hacker can, even with bare minimum computation, generate adversarial examples (images or data points that belong to another class, but consistently fool the model to get misclassified as genuine) and crumble the basis of such algorithms. We show two effective techniques, namely Dropout and Denoising Autoencoders, and show their success in preventing such attacks from fooling the model.
  arXiv  Detail & Related papers  (2020-12-10T09:34:41Z)
• Omni: Automated Ensemble with Unexpected Models against Adversarial Evasion Attack [35.0689225703137]
  A machine learning-based security detection model is susceptible to adversarial evasion attacks. We propose an approach called Omni to explore methods that create an ensemble of "unexpected models" In studies with five types of adversarial evasion attacks, we show Omni is a promising approach as a defense strategy.
  arXiv  Detail & Related papers  (2020-11-23T20:02:40Z)
• Online Alternate Generator against Adversarial Attacks [144.45529828523408]
  Deep learning models are notoriously sensitive to adversarial examples which are synthesized by adding quasi-perceptible noises on real images. We propose a portable defense method, online alternate generator, which does not need to access or modify the parameters of the target networks. The proposed method works by online synthesizing another image from scratch for an input image, instead of removing or destroying adversarial noises.
  arXiv  Detail & Related papers  (2020-09-17T07:11:16Z)
• On Adversarial Examples and Stealth Attacks in Artificial Intelligence Systems [62.997667081978825]
  We present a formal framework for assessing and analyzing two classes of malevolent action towards generic Artificial Intelligence (AI) systems. The first class involves adversarial examples and concerns the introduction of small perturbations of the input data that cause misclassification. The second class, introduced here for the first time and named stealth attacks, involves small perturbations to the AI system itself.
  arXiv  Detail & Related papers  (2020-04-09T10:56:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.