Gerardo Iuliano, Daniele Carangelo, Carmine T. Calabrese, Dario Di Nucci
Mutation testing is a technique to assess the effectiveness of test suites by introducing artificial faults into programs. Although mutation testing plugins are available for many platforms and languages, none is currently available for Remix-IDE, the most widely used Integrated Development Environment for the entire contract development journey, used by users of all knowledge levels, and serves as a learning lab for teaching and experimenting with Ethereum. The quality and security of smart contracts are crucial in blockchain systems, as even minor issues can result in substantial financial losses. This paper proposes MuSe, a mutation testing plugin for the Remix-IDE. MuSe includes traditional, Solidity-specific, and security-oriented mutation operators. Its integration into the Remix-IDE eliminates the need for additional setup and lowers the entry barrier. As a result, developers and researchers can immediately leverage mutation testing to assess the effectiveness of their test suites and identify potential issues in smart contracts. We provide a demo video showing MuSe: https://www.youtube.com/watch?v=MIFk9exTDu0 and its repository: https://github.com/GerardoIuliano/MuSe-Remix-Plugin.
In decentralized web applications, users face an inherent conflict between public verifiability and personal privacy. To participate in regulated on-chain services, users must currently disclose sensitive identity documents to centralized intermediaries, permanently linking real-world identities to public transaction histories. This binary choice between total privacy loss or total exclusion strips users of agency and exposes them to persistent surveillance. In this work, we introduce a Selective Disclosure Framework designed to restore user sovereignty by decoupling eligibility verification from identity revelation. We present ZK-Compliance, a prototype that leverages browser-based zero-knowledge proofs to shift the interaction model, enabling users to prove specific attributes (e.g., "I am over 18") locally without revealing the underlying data. We implement a user-governed Grant, Verify, Revoke lifecycle that transforms the user's mental model of compliance from a permanent data handover into a dynamic, revocable authorization session. Our evaluation shows that client-side proof generation takes under 200ms, enabling a seamless interactive experience on commodity hardware. This work provides early evidence that regulatory compliance need not come at the cost of user privacy or autonomy.
Shikah J. Alsunaidi, Hamoud Aljamaan, Mohammad Hammoudeh
Smart Contract (SC) vulnerabilities are programming errors or design flaws that can lead to financial loss or functional failure, making accurate detection essential. Although Machine Learning (ML) is widely applied to SC vulnerability detection, existing datasets are often small, imbalanced, inconsistently labeled, or nonstandardized, and frequently rely on limited feature representations that do not account for different contract lifecycle stages, restricting generalization and degrading benchmark reliability. This study introduces DIVE, a multi-label dataset that addresses these structural and feature-level limitations. DIVE includes 22,330 real-world SCs deployed between 2016 and 2024, and spanning major Solidity compiler versions, annotated for eight vulnerability types aligned with the Decentralized Application Security Project (DASP) Top 10 taxonomy. It provides 221 pre-deployment and 176 post-deployment features and employs a standardized multi-tool labeling pipeline based on Power-based voting and post-hoc filtering, which corrected 14.3% false positives in DoS and 24.9% in Time Manipulation. Unlike prior datasets, DIVE offers two lifecycle-specific feature sets and an open-source framework enabling reproducible benchmarking and periodic reconstruction aligned with evolving vulnerability patterns.
AI agents that execute tasks via tool calls frequently hallucinate results - fabricating tool executions, misstating output counts, or presenting inferences as facts. Recent approaches to verifiable AI inference rely on zero-knowledge proofs, which provide cryptographic guarantees but impose minutes of proving time per query, making them impractical for interactive agents. We propose NabaOS, a lightweight verification framework inspired by Indian epistemology (Nyaya Shastra), which classifies every claim in an LLM response by its epistemic source (pramana): direct tool output (pratyaksha), inference (anumana), external testimony (shabda), absence (abhava), or ungrounded opinion. Our runtime generates HMAC-signed tool execution receipts that the LLM cannot forge, then cross-references claims against these receipts to detect hallucinations in real time. We evaluate on NyayaVerifyBench, a new benchmark of 1,800 agent response scenarios across four languages with injected hallucinations of six types. NabaOS detects 94.2% of fabricated tool references, 87.6% of count misstatements, and 91.3% of false absence claims, with <15ms verification overhead per response. For deep delegation (agents performing multi-step web tasks), our cross-checking protocol catches 78.4% of URL fabrications via independent re-fetching. We compare against five approaches: zkLLM (cryptographic proofs, 180s/query), TOPLOC (locality-sensitive hashing), SPEX (sampling-based proof of execution), tensor commitments, and self-consistency checking. NabaOS achieves the best cost-latency-coverage trade-off for interactive agents: 94.2% coverage at <15ms versus zkLLM's near-perfect coverage at 180,000ms. For interactive agents, practical receipt-based verification provides better cost-benefit than cryptographic proofs, and epistemic classification gives users actionable trust signals rather than binary judgments.
Blockchain has gained significant attention in recent years, with smart contracts enabling automated and trustless financial interactions such as decentralized exchanges, tokenized assets, and on-chain governance. Because these programs often control assets of substantial value, a large body of research has focused on detecting security vulnerabilities in smart contracts. However, beyond security, understanding the actual behavior of a contract remains challenging, particularly when source code is unavailable. This work addresses the identification of semantic properties, defined as behavior-level characteristics that describe a contract's purpose based on its state changes and interactions. Detecting such properties can support applications such as regulatory analysis of relevant contracts and the simplification of contract logic by filtering semantically uninformative boilerplate code. The work focuses on static analysis approaches utilizing Datalog.The methodology first includes a systematic literature review to identify existing approaches for the static analysis of Ethereum bytecode using Datalog, as well as related work on semantic properties. The review indicates that Gigahorse is currently the most prominent tool in this category; consequently, it was selected as the basis for the following analysis.Based on this foundation, several function-level properties are defined, including authenticated functions as well as different types of setter and getter functions. In addition, a contract-level property representing a simple token contract is defined. Detection mechanisms for these properties are implemented in Datalog and subsequently evaluated. The results indicate that Gigahorse is generally well-suited for detecting such semantic properties, although practical limitations were encountered, particularly in the form of limited or missing documentation.
<b>Abstract.</b>Smart contract vulnerabilities have led to losses exceeding billions of US dollars in the decentralised finance (DeFi) ecosystem. Existing detection tools based on symbolic execution and static analysis, while precise, are computationally expensive and often impractical for large-scale screening. In this work, we propose a lightweight machine learning approach that operates directly on compiled EVM bytecode, requiring neither source code nor contract ABI. We design a feature engineering pipeline that extracts 65 security-oriented numerical features from disassembled bytecode instructions, covering reentrancy patterns, arithmetic overflow indicators, gas-based denial-of-service risks, access control anomalies, and environmental dependencies. Using a dataset of 117,091 real-world Ethereum smart contracts labelled by the Slither static analyser, we evaluate four classifiers—Logistic Regression, Decision Tree, Random Forest, and XGBoost—under stratified 5-fold cross-validation. XGBoost, optimised via Bayesian hyperparameter search (Optuna, 50 trials), achieves an F1-score of 0.947 on cross-validation and 93% accuracy on a held-out validation set, with 0.97 recall for vulnerable contracts and 0.85 recall for safe contracts. We additionally benchmark text-based opcode sequence representations and find that hand-crafted numerical features substantially outperform n-gram vectorisation approaches.<br>Code and materials (GitHub): https://github.com/SergeySolovyev/Machine-Learning-Based-Vulnerability-DetectionDate: 26 Feb 2026. Version: v1.
Process attestation verifies human authorship by collecting behavioral biometric evidence, including keystroke dynamics, typing patterns, and editing behavior, during the creative process. However, the very data needed to prove authenticity can reveal intimate details about an author's cognitive state, health conditions, and identity, constituting sensitive biometric data under GDPR Article 9. We resolve this privacy-attestation paradox using zero-knowledge proofs. We present ZK-PoP, a construction that allows a verifier to confirm that (a) sequential work function chains were computed correctly, (b) behavioral feature vectors fall within human population distributions, and (c) content evolution is consistent with incremental human editing, all without learning the underlying behavioral data, exact timing, or intermediate content. Our construction uses Groth16 proofs over arithmetic circuits with Pedersen commitments and Bulletproof range proofs. We prove that ZK-PoP is computationally zero-knowledge, computationally sound, and achieves unlinkability across sessions. Evaluation shows proof generation in under 30 seconds for a 1-hour writing session, with 192-byte proofs verifiable in 8.2 ms, while incurring less than 5% accuracy loss in simulation at practical privacy levels (epsilon >= 1.0) compared to non-private baselines.
Muhammad Sannan Khaliq, Love Allen Chijioke Ahakonye, Jae Min Lee, Dong-Seong Kim
With the rapid proliferation of blockchain technology and smart contracts in consumer IoT systems, ensuring digital trust and security remains a persistent challenge due to vulnerability diversity, data scarcity, and limited on-chain auditability. To address these issues, this paper presents PureChain, a scalable trust framework that integrates semi-supervised deep learning for smart contract vulnerability detection with blockchain-based audit logging for transparent security assurance. The framework employs a lightweight Conv1D neural architecture trained via iterative pseudo-labeling and a Mean Teacher strategy, effectively leveraging tens of thousands of unlabeled contracts to minimize annotation requirements. Experimental evaluation shows that PureChain achieves macro F1-scores exceeding 99% and significant improvements in recall for rare vulnerabilities, outperforming both supervised and state-of-the-art baselines. The integration of blockchain ensures all detection events are immutably recorded, enabling accountable and verifiable device operation. These results demonstrate that adaptive semi-supervised learning, combined with on-chain transparency, provides a robust and efficient foundation for secure smart contract monitoring across IoT and edge environments, with future work targeting more adaptive pseudo-labeling, semi supervised learning (SSL) techniques and cross-chain generalization.
Verifying that a compiled binary originates from its claimed source code is a fundamental security requirement, called source code provenance. Achieving verifiable source code provenance in practice remains challenging. The most popular technique, called reproducible builds, requires difficult matching and reexecution of build toolchains and environments. We propose a novel approach to verifiable provenance based on compiling software with zero-knowledge virtual machines (zkVMs). By executing a compiler within a zkVM, our system produces both the compiled output and a cryptographic proof attesting that the compilation was performed on the claimed source code with the claimed compiler. We implement a proof-of-concept implementation using the RISC Zero zkVM and the ChibiCC C compiler, and evaluate it on 200 synthetic programs as well as 31 OpenSSL and 21 libsodium source files. Our results show that zk-compilation is applicable to real-world software and provides strong security guarantees: all adversarial tests targeting compiler substitution, source tampering, output manipulation, and replay attacks are successfully blocked.
Smart contract security mainly aims to discover vulnerabilities in contract code based on the assumption that malicious behavior emerges from exploitable execution paths. This assumption is increasingly misaligned with modern blockchain systems, where off-chain automated agents continuously observe on-chain state and act autonomously. In this paper, we argue that smart contracts now act as public signalling mechanisms, capable of encoding command-and-control (C2) primitives through events and persistent storage variables. Off-chain agents, including LLM-based agents, convert benign-looking state changes into coordinated, high-impact actions across protocols and chains, semantically interpreting these signals. This kind of behaviour does not require code-level exploitation and thus can be built to evade existing analysis methods by construction. We explore how such a kind of coordination enables delayed activation, selective targeting, and deception, and explain why contract-centric security models fall short of such risks. We conclude by examining the main challenges and the directions for securing agent-mediated blockchain ecosystems.
Ziyue Wang, Jiangshan Yu, Kaihua Qin, Dawn Song · 6 authors
Decentralized Finance (DeFi) has turned blockchains into financial infrastructure, allowing anyone to trade, lend, and build protocols without intermediaries, but this openness exposes pools of value controlled by code. Within five years, the DeFi ecosystem has lost over 15.75B USD to reported exploits. Many exploits arise from permissionless opportunities that any participant can trigger using only public state and standard interfaces, which we call Anyone-Can-Take (ACT) opportunities. Despite on-chain transparency, postmortem analysis remains slow and manual: investigations start from limited evidence, sometimes only a single transaction hash, and must reconstruct the exploit lifecycle by recovering related transactions, contract code, and state dependencies. We present TxRay, a Large Language Model (LLM) agentic postmortem system that uses tool calls to reconstruct live ACT attacks from limited evidence. Starting from one or more seed transactions, TxRay recovers the exploit lifecycle, derives an evidence-backed root cause, and generates a runnable, self-contained Proof of Concept (PoC) that deterministically reproduces the incident. TxRay self-checks postmortems by encoding incident-specific semantic oracles as executable assertions. To evaluate PoC correctness and quality, we develop PoCEvaluator, an independent agentic execution-and-review evaluator. On 114 incidents from DeFiHackLabs, TxRay produces an expert-aligned root cause and an executable PoC for 105 incidents, achieving 92.11% end-to-end reproduction. Under PoCEvaluator, 98.1% of TxRay PoCs avoid hard-coding attacker addresses, a +22.9pp lift over DeFiHackLabs. In a live deployment, TxRay delivers validated root causes in 40 minutes and PoCs in 59 minutes at median latency. TxRay's oracle-validated PoCs enable attack imitation, improving coverage by 15.6% and 65.5% over STING and APE.
Weihong Wang, Yana Dimova, Victor Vansteenkiste, Tom Van Goethem · 5 authors
Cryptocurrency wallets are the primary interface for managing pseudonymous blockchain addresses, viewing balances, and interacting with Web3 applications. Although users typically assume that their addresses remain independent of each other unless intentionally revealed, modern wallets routinely communicate with both blockchain infrastructure and decentralized applications (dApps), generating network-side and web-side signals that may undermine this assumption. In this paper, we identify and formalize five privacy threats that arise directly from wallets interacting with the network and the web browser. Using large-scale dynamic measurements of 85 of the most popular Chrome Web Store browser-extension wallets (representing 35.16 million users), we observe that routine remote procedure call (RPC) operations leak structural links between a user's addresses; that the majority of Ethereum wallets implement permission revocation inconsistently and continue to expose previously revoked addresses across sessions; and that many wallets inject their provider interfaces into cross-origin iframes, enabling passive cross-site tracking beyond dApps and potentially real-world identity deanonymization without user interaction. Taken together, our results show that these wallet behaviors leak sensitive information that can be used to link multiple addresses to the same user, track wallet users across sessions and sites, and connect their browsing activity to their on-chain wealth. We discuss practical mitigations and show that many of these threats can be substantially reduced through improved wallet implementation, stronger privacy considerations in ecosystem standards, and stricter controls over provider exposure. Our results highlight the need for standardized, privacy-preserving wallet architectures and provide actionable guidance for strengthening user privacy in the emerging Web3 ecosystem.
Sundara Srivathsan M, Lighittha P. R., Prithivraj S., R. Suganya · 5 authors
Web3 platforms face a critical challenge: once unsafe content is minted on-chain, it becomes immutable and irrevocable. Traditional NSFW classifiers operate off-chain without cryptographic guarantees, leaving blockchain ecosystems vulnerable to harmful content. We present VisionGuard, a unified moderation framework that integrates cost-sensitive AI decision-making with blockchain-based enforcement. Our system combines calibrated NSFW classification, abstention-based triage for uncertain cases, perceptual hashing for near-duplicate detection, and on-chain k-of-n quorum attestation using EIP-712 signatures. We establish formal guarantees for: (i) Bayes-optimal cost-sensitive thresholds minimizing asymmetric error costs, (ii) optimal abstention intervals for human review, (iii) monotone false-negative reduction under classifier-pHash fusion, (iv) quorum compromise bounds, and (v) end-to-end unsafe-mint probability. Empirical validation on a zero-shot NSFW task demonstrates 82% accuracy (AUC =0.88), with the Bayes-optimal threshold (τ∗=0.1) reducing expected cost to 27,520 versus 54,942 at the F1-optimal threshold—a 50% improvement. Calibrated abstention further lowers harm (cost =10,649.5), while a 3-of-5 quorum with oracle compromise p=0.1 yields break probability Pbreak<1%. Together, VisionGuard bridges decision theory, adversarial robustness, and cryptographic enforcement, providing the first provably safe AI moderation pathway for blockchain content.
Abstract Existing accountability mechanisms based on audit logs, blockchains, and distributed ledgers attempt to prove compliance by recording what happened—transactions, identities, timestamps, or state transitions. While tamper-resistant, these approaches inherently generate behavioural traces that conflict with data-minimisation requirements and enable correlation, profiling, and surveillance. This creates a structural accountability–privacy paradox in modern regulatory frameworks such as the GDPR, which simultaneously require demonstrable compliance (Art. 5(2)) and strict limitation of personal data collection (Art. 5(1)(c)). This paper introduces Ledger-Anchored Validation Receipts (LAVR), a cryptographic mechanism that resolves this paradox by proving that enforcement occurred, rather than recording the underlying activity. LAVR operates exclusively at system finality boundaries—points where actions become irreversible, such as database commits, cross-border data exports, AI output release, payment settlement, or content publication. At each finality point, a hardware-adjacent enforcement gate evaluates a regulatory validation predicate derived from applicable legal constraints. Upon decision (Allow or Deny), the gate generates a minimal, fixed-size receipt that contains only enforcement metadata and explicitly excludes payloads, identities, transaction details, routing information, and fine-grained timestamps. The operational workflow consists of four steps: (1) an execution request reaches a finality boundary; (2) a trusted validation context evaluates scope, purpose, and control predicates; (3) a non-identifying receipt is cryptographically generated; and (4) a hash commitment of the receipt is anchored to a tamper-evident ledger. Unlike blockchain transaction logs, the ledger anchors only receipt commitments—not events—preventing the formation of transaction graphs or behavioural histories. Implemented using commodity hardware primitives (Trusted Execution Environments, SmartNICs, secure elements, and kernel-level finality hooks), LAVR enables accountability without surveillance across GDPR, AI Act, eIDAS, CBDC, NIS2, and Democracy Shield contexts. The approach demonstrates that regulatory simplification and stronger enforcement can be achieved through execution-time cryptographic proof rather than expanded monitoring or deregulation.
This paper introduces a cybersecurity framework that combines a deception-based ransomware detection system, called the Intrusion and Ransomware Detection System for Cloud (IRDS4C), with a blockchain-enabled Cyber Threat Intelligence platform (CTIB). The framework aims to improve the detection, reporting, and sharing of ransomware threats in cloud environments. IRDS4C uses deception techniques such as honeypots, honeytokens, pretender network paths, and decoy applications to identify ransomware behavior within cloud systems. Tests on 53 Windows-based ransomware samples from seven families showed an ordinary detection time of about 12 s, often quicker than tralatitious methods like file hashing or entropy analysis. These detection results are currently limited to Windows-based ransomware environments, and do not yet cover Linux, containerized, or hypervisor-level ransomware. Detected threats are formatted using STIX/TAXII standards and firmly shared through CTIB. CTIB applies a hybrid blockchain consensus of Proof of Stake (PoS) and Proof of Work (PoW) to ensure data integrity and protection from tampering. Security analysis shows that an attacker would need to control over 71% of the network to compromise the system. CTIB also improves trust, accuracy, and participation in intelligence sharing, while smart contracts control access to erogenous data. In a local prototype deployment (Hardhat devnet + FastAPI/Uvicorn), CTIB achieved 74.93–125.92 CTI submissions/min, The number of attempts or requests in each test was 100 with median end-to-end latency 455.55–724.99 ms (p95: 577.68–1364.17 ms) across PoW difficulty profiles (difficulty_bits = 8–16).
This technical disclosure describes a system and method for detecting vulnerabilities in smart contracts using semantic vector embeddings that compare code patterns against a curated database of historical blockchain exploits. Unlike traditional code similarity approaches that match syntactically similar code, the disclosed system performs exploit-centric similarity matching, identifying code that is semantically similar to previously exploited patterns even when syntactically different. The system incorporates attack mode classification, economic impact estimation, and cross-contract analysis to provide comprehensive vulnerability assessment. This document is published as a defensive publication to establish prior art and prevent third parties from obtaining patent protection for similar approaches while preserving trade secret protection for specific implementation details.
Many Ethereum smart contracts rely on block attributes such as block.timestamp or blockhash to generate random numbers for applications like lotteries and games. However, these values are predictable and miner-manipulable, creating the Bad Randomness vulnerability (SWC-120) that has led to real-world exploits. Current detection tools identify only simple patterns and fail to verify whether protective modifiers actually guard vulnerable code. A major obstacle to improving these tools is the lack of large, accurately labeled datasets. This paper presents a benchmark dataset of 1,752 Ethereum smart contracts with validated Bad Randomness vulnerabilities. We developed a five-phase methodology comprising keyword filtering, pattern matching with 58 regular expressions, risk classification, function-level validation, and context analysis. The function-level validation revealed that 49% of contracts initially classified as protected were actually exploitable because modifiers were applied to different functions than those containing vulnerabilities. We classify contracts into four risk levels based on exploitability: HIGH_RISK (no protection), MEDIUM_RISK (miner-exploitable only), LOW_RISK (owner-exploitable only), and SAFE (using Chainlink VRF or commit-reveal). Our dataset is 51 times larger than RNVulDet and the first to provide function-level validation and risk stratification. Evaluation of Slither and Mythril revealed significant detection gaps, as both tools identified none of the vulnerable contracts in our sample, indicating limitations in handling complex randomness patterns. The dataset and validation scripts are publicly available to support future research in smart contract security.
Smart contracts are increasingly being adopted in modern supply chain (SC) systems, offering a transformative shift from traditional centralized models to decentralized, automated, and trustless processes. However, vulnerabilities in the chain of smart contract represent a critical threat to the reliability and security of supply chain ecosystems, often arising from intricate logical flaws, unintended inter-contract interactions, or improper handling of user input—issues that remain difficult to uncover through conventional testing or manual auditing. This paper introduces an innovative deep learning–based methodology that transforms smart contract bytecode into image representations, enabling precise and efficient classification of diverse vulnerability patterns. Unlike existing approaches that rely on source code availability or dynamic execution, the proposed framework operates independently of source code and circumvents the limitations inherent to dynamic analysis, thereby offering a versatile and system-agnostic solution. Experimental evaluations conducted on a newly curated dataset collected from multiple publicly available repositories demonstrate the robustness of the proposed method, achieving 92.24% accuracy and an 89.06% F1-score. Beyond its strong empirical performance, the framework ensures reproducibility, data transparency, and adaptability across heterogeneous blockchain environments. Collectively, these contributions establish a comprehensive and accessible foundation for enhancing the detection, mitigation, and overall resilience of blockchain smart contracts.
Blockchain Technology Applications and Security
Physical Unclonable Functions (PUFs) and Hardware Security
Xinyu Hou, Yang Lu, Rabimba Karanjai, Lei Xu · 5 authors
Ransomware is still one of the most serious cybersecurity threats. Victims often pay but fail to regain access to their data, while also facing the danger of losing data privacy. These uncertainties heavily shape the attacker-victim dynamics in decision-making. In this paper, we introduce and analyze zkRansomware. This new ransomware model integrates zero-knowledge proofs to enable verifiable data recovery and uses smart contracts to enforce multi-round payments while mitigating the risk of data disclosure and privacy loss. We show that zkRansomware is technically feasible using existing cryptographic and blockchain tools and, perhaps counterintuitively, can align incentives between the attacker and the victim. Finally, we develop a theoretical decision-making framework for zkRansomware that distinguishes it from known ransomware decision models and discusses its implications for ransomware risk analysis and response decision support.
Firmware integrity is a foundational requirement for securing Cyber-Physical Systems (CPS), where malicious or compromised firmware can result in persistent backdoors, unauthorized control, or catastrophic system failures. Traditional verification mechanisms such as secure boot, digital signatures, and centralized hash databases are increasingly inadequate due to risks from insider threats and single points of failure. In this paper, we propose a decentralized firmware integrity verification framework built on the Ethereum blockchain, offering tamper-proof, transparent, and trustless validation. Our system stores SHA-256 hashes of firmware binaries within smart contracts deployed on the Ethereum Sepolia testnet, using Web3 and Infura for seamless on-chain interaction. A Python-based client tool computes firmware hashes and communicates with the blockchain to register and verify firmware authenticity in real-time. We implement and evaluate a fully functional prototype using real firmware samples, demonstrating successful contract deployment, hash registration, and integrity verification through live blockchain transactions. Experimental results confirm the reliability and low cost (in gas fees) of our approach, highlighting its practicality and scalability for real-world CPS applications. To enhance scalability and performance, we discuss extensions using Layer-2 rollups and off-chain storage via the InterPlanetary File System (IPFS). We also outline integration pathways with secure boot mechanisms, Trusted Platform Module (TPM)-based attestation, and zero-trust architectures. This work contributes a practical and extensible model for blockchain-based firmware verification, significantly strengthening the defense against firmware tampering and supply chain attacks in critical CPS environments.
Antonio Max L. B. Pereira, Dylan Paulin, Christine Hennebert
An intrusion into the operational network (OT) of a production site can cause serious damage by affecting productivity, reliability, and quality. The presence of embedded neural networks (NNs), such as classifiers, in physical devices opens the door to new attack vectors. Due to the stochastic behavior of the classifier and the difficulty of reproducing results, the Artificial Intelligence (AI) Act requires the NN’s behavior to be explainable. For this purpose, the platform HistoTrust enables tracing NN behavior, thanks to secure hardware components issuing attestations registered in a blockchain ledger. This solution helps to build trust between independent actors whose devices perform tasks in cooperation. This paper proposes going further by integrating a mechanism for detecting tampering of embedded NN, and using smart contracts executed on the blockchain to propagate the alert to the peer devices in a distributed manner. The use case of a bit-flip attack, targeting the weights of the NN model, is considered. This attack can be carried out by repeatedly injecting very small messages that can be missed by the Intrusion Detection System (IDS). Experiments are being conducted on the HistoTrust platform to demonstrate the feasibility of our distributed approach and to qualify the time required to detect intrusion and propagate the alert, in relation to the time it takes for the attack to impact decisions made by the AI. As a result, the blockchain may be a relevant technology to complement traditional IDS in order to face distributed attacks.
Smart contracts have become an integral part of most modern blockchain systems, playing a pivotal role in their operation. However, security vulnerabilities in smart contracts can lead to significant financial losses. Given the remarkable capabilities of Large Language Models (LLMs) in code understanding and analysis, it is worthwhile to explore how state-of-the-art LLMs can enhance the detection of smart contract vulnerabilities.In this paper, we propose a collaborative detection method that combines semantic analysis performed by LLMs with a lightweight deep learning model. Rather than relying on LLMs for direct vulnerability detection, we engineer tailored prompts that guide the model to perform multidimensional semantic analysis of smart contract code, yielding structured insights. Subsequently, we utilize a pre-trained CodeBERT model to encode these insights into dense feature vectors, which facilitate vulnerability detection via a lightweight, efficient deep learning classifier.Experiments on real-world smart contract datasets show that our approach achieves accuracy rates of 99.85%, 93.2%, 99.33%, and 98.13% for four types of vulnerabilities: reentrancy, timestamp dependency, unchecked external calls, and strict equality to Ether, respectively. These results demonstrate that the synergy between LLM-based semantic analysis and a lightweight detection model offers a novel and effective solution for automated smart contract auditing. This approach requires no expert knowledge, remains intuitive and easy to implement, and exhibits strong scalability.