Blockchain technology has emerged as a disruptive paradigm for secure and transparent data exchange; however, it continues to face significant challenges in scalability, interoperability, and security. Fragmentation across blockchain networks restricts seamless data integration, while traditional consensus mechanisms such as Proof of Work and Proof of Stake impose high computational costs and latency. To address these limitations, this article proposes an Intractive Blockchain structure to AI that takes advantage of deep learning and light consensus mechanisms to improve performance and safety. The proposed structure introduces three main contributions: (i) a model of detection of deep learning vulnerabilities that identifies real-time intelligent contract weaknesses to reduce application failures; (ii) a lightweight consensus protocol inspired by Byzantine failure tolerance (BFT) to minimize latency and improve the transfer rate, ensuring safe authentication; and (iii) a cross -chain interoperability layer that facilitates the perfect data exchange between heterogeneous blockchain networks. Experimental assessment of TensorFlow Hyperledger tissue shows that the proposed model improves the accuracy of vulnerabilities detection by up to 96 %, reaches a 23 % reduction in latency and increases the transfer rate by 18 % compared to conventional approaches. This research highlights the potential of AI-Empowered blockchain systems for scalable, secure and interpreter applications in financial, health and public services.
One of the most promising technology of blockchain still suffers from multiple attacks. Attack detection is handled by various technologies to enhance the security. Hackers moto is to steal the digital currency here the discussion is of the digital currency Ethereum which also has suffered from the famous attack Re-entrancy attack. Though the usage of smart contact is highly believed by users, making the function to be called again and again leads to this kind of attack. Re-entrancy attacks pose a persistent threat to Ethereum smart contracts, enabling malicious users to repeatedly call vulnerable functions before the state variables are correctly updated. While traditional approaches rely heavily on symbolic execution and SMT solvers for vulnerability detection, these methods suffer from scalability and precision limitations. A novel hybrid framework that integrates dynamic execution tracing with AI-driven classification models, such as Quantum Neural Networks (QNN), for the robust detection of re-entrancy attacks. The system leverages taint analysis and runtime monitoring to generate behavioral features, which are then used to train and evaluate machine learning models. Experiments conducted on benchmark datasets demonstrate superior accuracy and reduced false positives when compared to conventional static analysis techniques, showcasing the effectiveness of combining dynamic and AI-based strategies for smart contract security.
Blockchain technology has revolutionized the management of Electronic Health Records by enhancing healthcare efficiency, security, and immutability. To achieve an enhanced version of the healthcare framework, we utilize Algorand’s Pure Proof of Stake (P-POS) consensus algorithm, which further ensures decentralized, scalable, and tamper-proof data storage, thereby addressing key issues in EHR systems. This study examines the development of P-POS in healthcare, highlighting its benefits in enhancing patient privacy, interoperability, and data integrity. The conventional healthcare framework suffers from issues such as data breaches, unauthorized access, and inefficiencies in data storage. However, using P-POS may solve these problems, and it acts as a game-changer for next-generation EHR systems. Further, this paper discusses P-POS uses, advantages, and difficulties, establishing P-POS as a crucial facilitator of safe and effective healthcare ecosystems in the future. Finally, the discussion focuses on how this integration can be achieved within a healthcare framework.
BACKGROUND: Kenya’s public tertiary healthcare is facing persistent quality of healthcare challenges characterized by acute shortage of healthcare workers, frequent industrial unrest, broken-down healthcare facilities, and erratic supply of essential commodities. To address these systemic challenges the government introduced the asset lease financing (ALF) mechanism aimed at strengthen tertiary hospitals through modern medical equipment and technologies. However, the effect of ALF on quality remains highly debated and controversial. This study examined the effect and constraints of ALF in improving quality of healthcare within Kenya’s tertiary hospitals. METHODS: A convergent parallel mixed-methods design was employed with quantitative data collected from 145 hospital managers, staff and patients. Descriptive statistics were used to summarize participants characteristics and indicators of study variables. Ordinary least square regression was then used to estimate the effect of ALF on quality of tertiary healthcare, controlling for existing traditional funding. Complementary qualitative insights were gathered from 26 policymakers, hospital managers, and health financing experts through semi-structured interviews and analyzed using thematic analysis to identify patterns in strengths and constraints. Integration of findings happened through triangulation to enhance interpretation and understanding. RESULTS: Analysis showed that asset lease financing had a significant positive effect on quality of tertiary healthcare (β = 0.587, p < 0.01), explaining 26% of the variance. When traditional funding was controlled, ALF remained significant (β = 0.495, p < 0.01), with the model explaining 33% of the variance. Respondents attributed this to improved access to advanced diagnostic and therapeutic equipment, as well as expanded service capacity. However, descriptive summaries and qualitative perspectives revealed several constraints limiting ALF optimal effect in improving tertiary healthcare quality in Kenya. Stakeholders noted high recurrent costs, under-utilized assets, weak contract negotiation, and top-down procurement processes that limited hospital autonomy and contribution. Operational gaps, including inadequate training and delayed maintenance, further constrained ALF effect on quality. CONCLUSIONS: ALF has the potential to enhance quality of healthcare and technological capacity in Kenya’s tertiary hospitals, but its effects are contingent on robust governance, effective contract design, and alignment with institutional capacity which seem lacking in the Kenyan context. Without these safeguards, current leasing arrangements risk becoming fiscally unsustainable with little quality enhancement. Policymakers should strengthen transparency, decentralize decision-making, and incorporate performance-based provisions into leasing contracts to maximize ALF effect in enhancing quality of care.
We present ZK IR, a novel 32-bit instruction set architecture (ISA) specifically designed for efficient zero-knowledge proof generation using STARK protocols. Unlike existing zkVMs that adapt general-purpose ISAs like RISC-V, ZK IR is designed from first principles to minimize proving overhead while maintaining compatibility with modern compiler toolchains. Our key contribution is a rigorous analysis demonstrating that a pure 32-bit register architecture with software-based multi-precision arithmetic outperforms designs with wider registers or specialized field arithmetic units. We achieve approximately 2× reduction in constraint count compared to naive approaches. ZK IR uses the Baby Bear field (31-bit prime) with Plonky3 for proving, and provides an LLVM-based compiler infrastructure enabling developers to write ZK applications in Rust, C, and C++.
Certificate authentication in online systems is required to ensure integrity and authenticity and prevent forgery. Traditional blockchain-based approaches work with double-chain architecture without any privacy-preservation capability or pack whole certificates into a single chain and incur substantial storage overhead. In this study, we introduce a light-weight dual-blockchain architecture with an external verification and audit side-chain and an inner chain for offline storing certificates. Zero-Knowledge Succinct Non-Interactive Arguments of Knowledge (Zero Knowledge Proofs) are employed to sign hashes without revealing sensitive information, and threshold signatures are employed to sign certificates. Compared to the traditional single-chain and double-chain architecture, the proposed system realizes up to 30% lower latency and 25% higher throughput based on experimental results on 1,000–100,000 certificate dataset. These results indicate the efficiency, scalability, and privacy-preserving feature of the proposed solution, which can be applied to large-scale applications for certificate management.
This research examines how emerging forms of digital sovereignty, decentralized infrastructures, and anticipatory AI governance are reshaping nationhood in the algorithmic age. Drawing on the conceptual framework of Algorithmic Nations (Calzada 2018) and incorporating new empirical insights from embedded action research (2022–2025), the study analyses the Basque Country as a paradigmatic case of a “small stateless nation” navigating the global reconfiguration of power between states, corporations, and communities. The presentation synthesizes three competing post-Westphalian paradigms—Network States (Srinivasan 2022), Network Sovereignties (De Filippi 2024), and Algorithmic Nations (Calzada 2018)—as shown in the comparative table on page 19, highlighting their differing assumptions regarding governance, identity, participation, and technological control. Building on the diagnostic indicators of Europe’s digital dependence (page 10) and the transition from Gaia-X to EuroStack (page 11), the study evaluates the strategic implications of digital public infrastructures, data cooperatives, federated architectures, and Web3 ecosystems for stateless nations. Through comparative analysis of the Global North (e.g., Scotland, Quebec, Flanders), the Global South (e.g., Kurdistan, Sámi, Tamil, Amazigh), and the Basque Country (pages 16–17), the work demonstrates how communities with diverse geopolitical constraints can articulate forms of AI sovereignty grounded in rights-based, culturally rooted, and community-driven governance. The Basque case illustrates how fragmented digital systems (.eus, EJIE/Izenpe, Osakidetza, MUBIL, etc.) can evolve toward an interoperable, multi-scalar technopolitical architecture, aligning linguistic, territorial, and infrastructural dimensions. The analysis argues that AI-driven infrastructures, data governance, and decentralized architectures are not merely technical layers but emerging geopolitical terrains where stateless, indigenous, diasporic, and minority nations can renegotiate autonomy. The concept of Algorithmic Nations provides a framework for understanding how community sovereignty can be built through data commons, federated systems, and anticipatory governance, particularly in multilingual and culturally distinct territories such as the Basque Country. Overall, the study contributes to debates on global digital governance, digital sovereignty, and the future of nationhood by proposing that algorithmic infrastructures are becoming central to political organization. It calls for democratic, inclusive, and community-oriented models of AI governance capable of avoiding techno-authoritarianism, Big Tech dependency, and “sovereignty washing,” while enabling emancipatory, culturally anchored, and future-oriented forms of collective self-determination.
Behavioral finance has emerged as a critical framework for understanding market dynamics beyond traditional rational agent models. This research presents a comprehensive multimodal approach to behavioral finance analysis, integrating market data, macroeconomic indicators, news sentiment, cryptocurrency metrics, Web3 analytics, GitHub development activity, and social sentiment to test five advanced hypotheses regarding behavioral pattern identification and market anomaly detection. The study employs an ultra-comprehensive data pipeline processing 30,400 samples across seven distinct data sources, generating 91 engineered features representing behavioral biases, investment patterns, and market psychology. Advanced machine learning techniques including Principal Component Analysis, t-Distributed Stochastic Neighbor Embedding, Variational Autoencoders, K-Means, Hierarchical Clustering, DBSCAN, Isolation Forest, One-Class SVM, and Elliptic Envelope are applied to identify behavioral structures and detect anomalies. Statistical validation through chi-square tests, ANOVA, Granger causality analysis, and lagged correlation studies demonstrates that three of five hypotheses (60%) achieve statistical significance at p < 0.05. Key findings reveal that behavioral structures exist and correspond to canonical biases (chi-square = 3406.780, p < 0.001), cluster assignments maintain moderate stability across market regimes (Jaccard similarity = 0.300), and sentiment and macroeconomic factors exhibit 65 significant causal relationships with behavioral patterns. However, multimodal data integration does not uniformly improve clustering quality (Silhouette score decrease of 0.116), and cluster-conditioned anomaly detection fails to outperform global methods (F1-score decrease of 0.017). These findings contribute to behavioral finance theory while providing practical applications for investment management, fraud detection, and regulatory compliance.
The exponential growth in automobile ownership has intensified global carbon emissions, underscoring the urgent need for sustainable transport solutions. Electric Vehicles (EVs) represent a cleaner alternative; however, their adoption in India remains limited. This paper proposes a blockchain-enabled carbon credit trading platform designed to incentivise EV adoption by converting verified charging activities into carbon credit tokens. Through smart contracts and decentralized verification, the system ensures transparency, accountability, and equitable distribution of credits. Two innovative algorithms underpin the model: the Carbon Credit Generation Algorithm, which standardizes token issuance, and the Voting Algorithm, which validates charging data through a consensus-based mechanism. Experimental results demonstrate that the proposed framework achieves up to 500 transactions per second with energy consumption as low as 0.01 kWh per transaction, significantly outperforming Proof-of-Work and Proof-of-Stake consensus models. The findings confirm that blockchain-driven mechanisms can effectively link green energy use to measurable rewards, fostering trust, scalability, and active participation in carbon-neutral mobility ecosystems.
The effectiveness of distributed ledger technology (DLTs), in particular blockchain, in improving the security and administration of massive distributed systems made up of independent agents is examined in this research. It looks at how these technologies are making the leap from lab settings to practical robotics uses. Recently, solutions that are resilient to sporadic connectivity and provide scalable frameworks for managed or permissioned networks have been presented, despite persistent concerns over scalability and practical implementation. While highlighting the numerous applications in which DLTs facilitate multi-robot systems, the paper also points out that the majority of research on open, permissionless blockchains is only applicable to a small number of robotics applications. As an alternative, the emphasis is now on permissioned blockchains and creative DLT architectures, which are propelling adoption in more conventional and industrial sectors.
This paper presents a decentralized application (dApp) for warranty management built on the Ethereum blockchain using smart contracts. The solution addresses key challenges in conventional warranty systems, such as lack of transparency, difficulty in transferring ownership, and reliance on centralized control. By leveraging Solidity-based smart contracts, the proposed framework establishes a transparent and immutable warranty registry that supports registration, verification, and ownership transfer of warranties. A React-based frontend, integrated with MetaMask, provides an intuitive interface for warranty-related operations. Comprehensive testing and deployment in real-world scenarios demonstrate the system’s ability to manage the complete warranty lifecycle while enhancing security and transparency. This work contributes to the growing body of research on blockchain-enabled supply chain and asset management by offering a practical framework for decentralized warranty systems in production environments.
Jiaxi Liu, Lin Sun, Tianyu Kang, Di Wu · 7 authors
Federated Learning (FL) enables model training on distributed devices while preserving data privacy. However, malicious clients can submit fabricated model updates to fraudulently obtain training rewards, a behavior known as free-rider attacks. Existing detection-based solutions analyze anomalies in model updates but lack direct evidence of local training, making it fail to fully prevent free-riders. To address this limitation, we propose zkVFL, a verifiable FL framework leveraging Zero-Knowledge Proofs (ZKP) to ensure the integrity of local training while preserving privacy. To reduce the computational overhead of proof generation in ZKP, zkVFL introduces two novel techniques: (i) anomaly-aware client sampling to selectively perform ZKP verification and (ii) A recursive ZKP protocol (ReMPoT), incorporating a pruning-based layer selection technique, reduces proof generation costs. Experimental results demonstrate that zkVFL improves the accuracy and convergence of FL training under free-rider attacks while significantly reducing the computational and memory overhead of proof generation on resource-constrained devices.
Permissionless blockchains have evolved beyond cryptocurrency into foundations for Web3 applications, decentralized finance (DeFi), and digital asset ownership, yet this rapid expansion has intensified privacy vulnerabilities. This study provides a comprehensive review of recent trends, emerging privacy threats, and mitigation strategies in permissionless blockchain ecosystems. We examine six developments reshaping the landscape: meme coin proliferation on high-throughput networks, real-world asset tokenization linking on-chain activity to regulated identities, perpetual derivatives exposing trading strategies, institutional adoption concentrating holdings under regulatory oversight, prediction markets creating permanent records of beliefs, and blockchain–AI integration enabling both privacy-preserving analytics and advanced deanonymization. Through this work and forensic analysis of documented incidents, we analyze seven critical privacy threats grounded in verifiable 2024–2025 transaction data: dust attacks, private key management failures, transaction linking, remote procedure call exposure, maximal extractable value extraction, signature hijacking, and smart contract vulnerabilities. Blockchain exploits reached $2.36 billion in 2024 and $2.47 billion in the first half of 2025, with over 80% attributed to compromised private keys and signature vulnerabilities. We evaluate privacy-enhancing technologies, including zero-knowledge proofs, ring signatures, and stealth addresses, identifying the gap between academic proposals and production deployment. We further propose a Secure Development Lifecycle framework incorporating measurable security controls validated against incident data. This work bridges the disconnect between privacy research and industrial practice by synthesizing current trends, providing insights, documenting real-world threats with forensic evidence, and providing actionable insights for both researchers advancing privacy-preserving techniques and developers building secure blockchain applications.
Nov 28, 2025·2025 IEEE 1st International Conference on Smart Innovations in Systems, Infrastructure, Mechanical, Power, AI and Computing Technologies (SISIMPACT)
Blockchain and Decentralized Finance (DeFi) are transforming the financial world by redefining how financial transactions and services operate. This paper explores the convergence of these technologies, emphasizing their potential to disintermediate traditional banking, enhance financial inclusivity, and increase transparency and efficiency. While these innovations hold significant promise, they are also met with challenges, including regulatory uncertainty, security vulnerabilities, and technological limitations. This study analyzes the underlying technologies, key platforms, and comparative frameworks that highlight how DeFi is challenging conventional financial systems, while suggesting avenues for sustainable adoption and global impact
Denis Wapukha Walumbe, Gabriel Kamau, Jane Wanjiru Njuki
With the rising integration of blockchain in critical domains such as healthcare, designing efficient, lightweight, and privacy-preserving consensus mechanisms remain a significant challenge.Existing Proof-of-Stake (PoS) implementations often incur high computational and communication overhead, making them unsuitable for telemedicine systems.This study proposed LightweightPoS, a novel voting mechanism designed for this environment.The proposed mechanism incorporates a cluster-based voting to minimize message complexity, Byzantine Agreement protocol for robust fault tolerance and cryptographic sortition to ensure fairness and privacy.This implementation slashes global communication, reducing message complexity by over 95% compared to traditional PoS models.The study evaluated the proposed and baseline mechanisms through simulations using real-time telemedicine data sensors.The results demonstrated that the proposed mechanism consistently achieved sub-10ms latency, high transaction throughput (up to 2400 TPS) and low energy consumption (~0.002kWh per round).It significantly outperformed baseline mechanism like Algorand and Ouroboros.Furthermore, the system included an effective Byzantine node detection, ensuring reliability under adversarial conditions.This work contributes a practical consensus voting mechanism that balances privacy and regulatory compliance.It provides a robust foundation for deploying blockchain technology in privacy-sensitive telemedicine applications.
The proliferation of Web3 and Internet of Things (IoT) applications generates unprecedented volumes of real-time data streams, demanding secure and efficient subscription mechanisms that uphold data sovereignty. While decentralized architectures are the logical paradigm to ensure this sovereignty, a prominent class of existing schemes suffers from critical vulnerabilities—notably revocation attacks and prohibitive communication overhead—that severely hinder their practical deployment in large-scale environments. This paper introduces SegSub, a novel decentralized data subscription scheme specifically designed to significantly enhance both security and efficiency. SegSub's core innovations include the Segmented Dual-Key Regression with Binary Hash Trees (SDKR-BHT) mechanism, which partitions key regression chains into isolated segments to effectively contain potential data leakage and optimize token management and a strategic user grouping policy that localizes key updates, thereby substantially reducing system-wide communication overhead during revocation events. We formally quantify security improvements using a proposed security index and demonstrate a configurable trade-off between security and efficiency. Theoretical analysis and extensive experimental results validate that SegSub's security index is inversely proportional to segment length while communication efficiency is directly proportional. Furthermore, our grouping policy significantly reduces communication costs in large-scale scenarios through optimal group sizing. SegSub offers a robust and adaptable foundation for sovereignty-preserving data subscription services in Web3, empowering system designers with precise control over the critical security-efficiency balance to meet diverse deployment requirements.
Giovanni Maria Cristiano, Salvatore D'Antonio, Jonah Giglio, Giovanni Mazzeo · 5 authors
The growing scalability demand of public Blockchains led to the rise of Layer-2 solutions, such as Rollups. Rollups improve transaction throughput by processing operations off-chain and posting the results on-chain. A critical component in Rollups is the Sequencer, responsible for receiving, ordering and batching transactions before they are submitted to the Layer-1 blockchain. While essential, the centralized nature of the Sequencer makes it vulnerable to attacks, such as censorship, transaction manipulation and tampering. To enhance its security, there are solutions in the literature that shield the Sequencer inside a Trusted Execution Environment (TEE). However, the attestation of TEEs introduces additional centralization, which is in contrast with the core Blockchain principle. In this paper, we propose a TEE-secured Sequencer equipped with a decentralized attestation mechanism. We outline the design and implementation of our solution, covering the system architecture, TEE integration, and the decentralization of the attestation process. Additionally, we present an experimental evaluation conducted on a realistic Rollup testnet. Our results show that this approach strengthens Sequencer integrity without sacrificing compatibility or deployability in existing Layer-2 architectures.
Wenbin Wu, Kejiang Qian, Alexis Lui, Christopher Jack · 8 authors
We curate the DeXposure dataset, the first large-scale dataset for inter-protocol credit exposure in decentralized financial networks, covering global markets of 43.7 million entries across 4.3 thousand protocols, 602 blockchains, and 24.3 thousand tokens, from 2020 to 2025. A new measure, value-linked credit exposure between protocols, is defined as the inferred financial dependency relationships derived from changes in Total Value Locked (TVL). We develop a token-to-protocol model using DefiLlama metadata to infer inter-protocol credit exposure from the token's stock dynamics, as reported by the protocols. Based on the curated dataset, we develop three benchmarks for machine learning research with financial applications: (1) graph clustering for global network measurement, tracking the structural evolution of credit exposure networks, (2) vector autoregression for sector-level credit exposure dynamics during major shocks (Terra and FTX), and (3) temporal graph neural networks for dynamic link prediction on temporal graphs. From the analysis, we observe (1) a rapid growth of network volume, (2) a trend of concentration to key protocols, (3) a decline of network density (the ratio of actual connections to possible connections), and (4) distinct shock propagation across sectors, such as lending platforms, trading exchanges, and asset management protocols. The DeXposure dataset and code have been released publicly. We envision they will help with research and practice in machine learning as well as financial risk monitoring, policy analysis, DeFi market modeling, amongst others. The dataset also contributes to machine learning research by offering benchmarks for graph clustering, vector autoregression, and temporal graph analysis.
P. Bhuvaneshwari, A Krishnaveni, Harold Robinson, E. Golden Julie
The deep learning technique has emerged as an exemplary model for managing the Artificial Intelligence-based Blockchain framework with technological enhancements to guarantee reliable data through the consensus procedure. The deep learning-enabled blockchain transaction model has involved the development of security to solve the problems of confidentiality and data anonymity. The Hybrid techniques of the Blockchain with the Deep Learning technique are proposed to generate enhanced data durability and its propagation through the enhanced convolutional temporal network (EnCTN) for transaction analysis in a blockchain-enabled Auto Encoder technique. The sliding window extraction technique is used to extract information from a particular window size to evaluate the needed input values from the temporal series. The dilated Convolution is used to capture the long-range dependencies. The proposed technique is implemented in the Ethereum environment using Python, and experimental results show that it has produced an improved performance than the relevant technique in several performance parameters. The anomaly classification accuracy is improved than the relevant technique and it is evaluated using the NSL-KDD dataset. The proposed framework delivers an efficient solution for the real-world anomaly detection application while accurate discovery of temporal anomalies and computational efficiency is enhanced.
The agricultural food supply chain (FSC) is a critical infrastructure that demands robust traceability, transparency, and security to ensure food safety and consumer trust. However, current FSC systems suffer from third-party dependencies, data integrity concerns, limited traceability, and fragmented communication among stakeholders. To address these issues, we propose a novel blockchain-based solution featuring a hybrid architecture that efficiently balances on-chain security with off-chain scalability. Unlike conventional approaches, our design introduces specialized smart contracts tailored to agricultural workflows. The system is implemented using Ethereum smart contracts and the InterPlanetary File System (IPFS) to ensure end-toend traceability. Experimental evaluation demonstrates 50.41% lower CPU usage, 65.37% reduced latency, and 48.73% higher throughput compared to existing methods. Additionally, security analysis with Slither confirms the system's robustness. This work sets a new benchmark for building efficient, transparent, and secure food supply chains.
This paper presents an empirical investigation of textual and semantic cues for fake news detection using FAKES-XL, a multi-domain, multi-language benchmark with leak-proof splits. Current reports often conflate gains with source/topic leakage and rarely assess probability calibration, limiting deployability across sources and languages. The present study trained text-only, semantic-only, and fused models on five bundles spanning English, Spanish, German, Hindi, and Italian, with temporal/source-grouped, topic-disjoint, cross-lingual zero-shot, and entity-disjoint evaluations. The methodology incorporated precommitted textual features (n-grams, stylometry, readability) and semantic signals (contextual embeddings, discourse, knowledge and retrieval-based evidence), applied post-hoc calibration, and quantified uncertainty via stratified bootstrap. Outcomes included Macro F1, Area Under the Receiver Operating Characteristic (AUROC), Area Under the Precision-Recall Curve (AUPRC), and Expected Calibration Error (ECE), with per-source and per-language scorecards and latency profiling under deployment constraints ($<=50 ~\text{ms}$on GPU;$<=120 ~\text{ms}$on CPU). While numeric results are not reported here, the analysis quantified the marginal value of each cue family, ablated discourse/knowledge/retrieval components, and produced calibrated thresholds tuned on validation and frozen on test. The contributions are a controlled comparison under strict leakage guards and a calibration-first evaluation that informs threshold selection. These findings support practical moderation workflows by offering reproducible scorecards and deployment-ready operating points.
AI-Driven Cybersecurity and Anomaly Detection in Blockchain While the decentralized and open-source nature of blockchain provides inherent security, vulnerabilities can still exist. AI tools and models can significantly bolster cybersecurity by identifying unusual patterns, detecting threats in real-time, and automating responses to maintain network integrity, prevent fraud, and enhance overall resilience. AI can detect fraud in real-time, predict vulnerabilities, and automate smart contracts for improved efficiency. It strengthens security by identifying unusual patterns that may indicate potential threats or breaches. AI-powered anomaly detection, utilizing techniques such as Long Short-Term Memory networks, can continuously monitor multi-sensor data streams to detect malicious data injection and sensor malfunctions in real-time, recording alerts on a blockchain ledger for incorruptibility and authenticity. Machine learning algorithms can analyze vast amounts of blockchain address and transaction data to identify patterns indicative of malicious activity, such as deviations from typical patterns or known fraud signatures. This includes detecting double-spending, transaction spamming, or unusual transaction volumes. In Decentralized Finance, AI-powered fraud detection systems, employing machine learning and graph-based algorithms, can map complex wallet connections, detect high-risk addresses, and adapt to changing scammer tactics in real-time. This capability is critical for Anti-Money Laundering audits. The ability to freeze accounts, block transfers, or notify users instantly is a key benefit of real-time AI fraud detection in crypto, as transactions are often fast and irreversible. One of the use cases our group implemented was using a modified Smart-LLaMa model to determine the reliability rating of blockchain addresses. We used large language models for detecting vulnerabilities in closed-source Ethereum smart contracts. The model was fine-tuned on a collected dataset of operational codes to adapt to the semantics of compiled smart contracts. The method allows for assessing the reliability of addresses based on the technical content of contracts, eliminating dependence on the source code, which is an excellent tool for enhancing the security of decentralized applications amidst the growing number of attacks on blockchain. www.srcmeetings.com17International Conference on Artificial Intelligence and Cybersecurity (ICAIC 2025)November 27-28, 2025 (Virtual)Conference Proceedings
The rapid expansion of Indonesia’s digital financial ecosystem has significantly advanced financial inclusion and innovation through the growth of fintech platforms, digital payments, and crypto-asset adoption. However, this transformation introduces multifaceted risks, including cyber threats, data breaches, digital fraud, regulatory uncertainty, and money-laundering vulnerabilities associated with crypto-assets and decentralized finance. This study employs a systematic literature review to examine the challenges and innovations in digital financial risk control within Indonesia’s fintech and digital asset sectors. Findings indicate that effective risk mitigation relies heavily on regulatory coordination, advanced supervisory technology, consumer digital literacy, and robust data protection practices. RegTech and SupTech innovations powered by artificial intelligence support real-time risk monitoring and enhance compliance with global standards such as FATF recommendations. Nevertheless, successful digital financial governance also requires algorithmic accountability and ethical technology deployment. This study underscores that safeguarding stability, trust, and consumer protection is essential to achieving a secure and inclusive digital financial system while enabling responsible innovation.
The fundamental limitation of blockchain architecture lies not in cryptographic primitives or consensus mechanisms, but in a conceptual mistake: the bundling of state transitions with asset custody. Every distributed ledger since Bitcoin has conflated these two concerns, creating an artificial ceiling on performance that no amount of clever engineering can overcome. This paper presents Virtual Rollups, a post-blockchain architecture that achieves what was previously thought impossible—sub-millisecond finality with full self-custody—by recognizing that state and escrow need not travel together. We formalize the Virtual Rollup construction, prove its security properties under Byzantine conditions, and demonstrate how its unified liquidity layer solves the multi-chain fragmentation problem that plagues decentralized finance. The result is not merely an incremental improvement but a categorical leap: trading venues can now match centralized exchanges in performance while exceeding them in security.