arXiv subjects

Chetan Pathade

Publications and source records attributed to Chetan Pathade.

5 recordsLinked to original sources

arXiv2026

Serverless AI Security: Attack Surface Analysis and Runtime Protection Mechanisms for FaaS-Based Machine Learning

Serverless computing has achieved widespread adoption, with over 70% of AWS organizations using serverless solutions [1]. Meanwhile, machine learning inference workloads increasingly migrate to Function-as-a-Service (FaaS) platforms for their scalability and cost-efficiency [2], [3], [4]. However, this convergence introduces critical security challenges, with recent reports showing a 220% increase in AI/ML vulnerabilities [5] and serverless computing's fragmented architecture raises new security concerns distinct from traditional cloud deployments [6], [7]. This paper presents the first comprehensive security analysis of machine learning workloads in serverless environments. We systematically characterize the attack surface across five categories: function-level vulnerabilities (cold start exploitation, dependency poisoning), model-specific threats (API-based extraction, adversarial inputs), infrastructure attacks (cross-function contamination, privilege escalation), supply chain risks (malicious layers, backdoored libraries), and IAM complexity (ephemeral nature, serverless functions). Through empirical assessments across AWS Lambda, Azure Functions, and Google Cloud Functions, we demonstrate real-world attack scenarios and quantify their security impact. We propose Serverless AI Shield (SAS), a multi-layered defense framework providing pre-deployment validation, runtime monitoring, and post-execution forensics. Our evaluation shows SAS achieves 94% detection rates while maintaining performance overhead below 9% for inference latency. We release an open-source security toolkit to enable practitioners to assess and harden their serverless AI deployments, advancing the field toward more resilient cloud-native machine learning systems.

cs.CR↗

arXiv2025

Invisible Injections: Exploiting Vision-Language Models Through Steganographic Prompt Embedding

Vision-language models (VLMs) have revolutionized multimodal AI applications but introduce novel security vulnerabilities that remain largely unexplored. We present the first comprehensive study of steganographic prompt injection attacks against VLMs, where malicious instructions are invisibly embedded within images using advanced steganographic techniques. Our approach demonstrates that current VLM architectures can inadvertently extract and execute hidden prompts during normal image processing, leading to covert behavioral manipulation. We develop a multi-domain embedding framework combining spatial, frequency, and neural steganographic methods, achieving an overall attack success rate of 24.3% (plus or minus 3.2%, 95% CI) across leading VLMs including GPT-4V, Claude, and LLaVA, with neural steganography methods reaching up to 31.8%, while maintaining reasonable visual imperceptibility (PSNR greater than 38 dB, SSIM greater than 0.94). Through systematic evaluation on 12 diverse datasets and 8 state-of-the-art models, we reveal moderate but meaningful vulnerabilities in current VLM architectures and propose effective countermeasures. Our findings have significant implications for VLM deployment in security-critical applications and highlight the need for proportionate multimodal AI security frameworks.

cs.CR↗

arXiv2025

Exposing Hidden Backdoors in NFT Smart Contracts: A Static Security Analysis of Rug Pull Patterns

The explosive growth of Non-Fungible Tokens (NFTs) has revolutionized digital ownership by enabling the creation, exchange, and monetization of unique assets on blockchain networks. However, this surge in popularity has also given rise to a disturbing trend: the emergence of rug pulls - fraudulent schemes where developers exploit trust and smart contract privileges to drain user funds or invalidate asset ownership. Central to many of these scams are hidden backdoors embedded within NFT smart contracts. Unlike unintentional bugs, these backdoors are deliberately coded and often obfuscated to bypass traditional audits and exploit investor confidence. In this paper, we present a large-scale static analysis of 49,940 verified NFT smart contracts using Slither, a static analysis framework, to uncover latent vulnerabilities commonly linked to rug pulls. We introduce a custom risk scoring model that classifies contracts into high, medium, or low risk tiers based on the presence and severity of rug pull indicators. Our dataset was derived from verified contracts on the Ethereum mainnet, and we generate multiple visualizations to highlight red flag clusters, issue prevalence, and co-occurrence of critical vulnerabilities. While we do not perform live exploits, our results reveal how malicious patterns often missed by simple reviews can be surfaced through static analysis at scale. We conclude by offering mitigation strategies for developers, marketplaces, and auditors to enhance smart contract security. By exposing how hidden backdoors manifest in real-world smart contracts, this work contributes a practical foundation for detecting and mitigating NFT rug pulls through scalable automated analysis.

cs.CR↗

arXiv2025

Securing Genomic Data Against Inference Attacks in Federated Learning Environments

Federated Learning (FL) offers a promising framework for collaboratively training machine learning models across decentralized genomic datasets without direct data sharing. While this approach preserves data locality, it remains susceptible to sophisticated inference attacks that can compromise individual privacy. In this study, we simulate a federated learning setup using synthetic genomic data and assess its vulnerability to three key attack vectors: Membership Inference Attack (MIA), Gradient-Based Membership Inference Attack, and Label Inference Attack (LIA). Our experiments reveal that Gradient-Based MIA achieves the highest effectiveness, with a precision of 0.79 and F1-score of 0.87, underscoring the risk posed by gradient exposure in federated updates. Additionally, we visualize comparative attack performance through radar plots and quantify model leakage across clients. The findings emphasize the inadequacy of na\"ive FL setups in safeguarding genomic privacy and motivate the development of more robust privacy-preserving mechanisms tailored to the unique sensitivity of genomic data.

cs.CR↗

arXiv2025

Red Teaming the Mind of the Machine: A Systematic Evaluation of Prompt Injection and Jailbreak Vulnerabilities in LLMs

Large Language Models (LLMs) are increasingly integrated into consumer and enterprise applications. Despite their capabilities, they remain susceptible to adversarial attacks such as prompt injection and jailbreaks that override alignment safeguards. This paper provides a systematic investigation of jailbreak strategies against various state-of-the-art LLMs. We categorize over 1,400 adversarial prompts, analyze their success against GPT-4, Claude 2, Mistral 7B, and Vicuna, and examine their generalizability and construction logic. We further propose layered mitigation strategies and recommend a hybrid red-teaming and sandboxing approach for robust LLM security.

cs.CR↗