Large Language Models (LLMs) such as GPT and similar architectures have revolutionized artificial intelligence by enabling machines to understand and generate human-like text. However, these models are inherently statistical predictors rather than real-time reasoning systems, leading to fundamental limitations in accessing up-to-date information and verifying factual accuracy. This issue is particularly critical in high-stakes domains such as cryptocurrency markets, decentralized finance (DeFi), and autonomous AI agents, where real-time, verifiable, and tamper-proof information is essential for decision-making.In this paper, we introduce AI Oracle, a novel framework that integrates blockchain-powered oracles with LLMs and autonomous agents to ensure real-time access to cryptographically verified knowledge. We compare AI Oracle with both standalone LLMs and retrieval-based systems using the Model Context Protocol (MCP), highlighting significant advantages in factual reliability, adversarial robustness, and interpretability. AI Oracle combines decentralized consensus, immutable storage, and cryptographic attestation to equip AI agents with enhanced resistance to manipulation, hallucination, and misinformation.Beyond architectural improvements, we explore the broader applicability of AI Oracle across domains that require provable correctness and trustâranging from real-world asset (RWA) tokenization to autonomous agent coordination and decentralized governance. By positioning AI Oracle as a trust-minimized epistemic infrastructure, we propose a new paradigm in AI systems: the fusion of decentralized trust with autonomous reasoning, enabling agents to operate with resilience, transparency, and embedded verifiability across dynamic environments.
Florian Spychiger, Sabrina WollenschlÀger, Matthias Hafner, Nicolas Oderbolz
Decentralized autonomous organizations (DAOs) have gained popularity over the last few years. Many projects use a DAO for community-based decisions and use a token to enable governance processes and foster participation. The setup of these tokens varies from DAO to DAO. While there are some general tokenomics frameworks, there is no DAO-specific framework including designs of multiple tokens. In this short paper, we aim at exploring the development of such DAO tokenomics framework. To unravel requirements and benefits of such a framework, we conduct interviews with six experts from the Swiss blockchain ecosystem. Switzerland is at the forefront of blockchain development and therefore well suited to serve as an exploration ground. Our results show that a DAO tokenomics framework needs to provide clear guidance while still being flexible to diverse project needs. It may bring along economic gains coupled with a risk reduction and an innovation boost for Switzerland. These benefits could be generalized to other jurisdictions making the development of a DAO tokenomics framework worthwhile.
ABSTRACT The rise of mining pools in Blockchain networks has improved reward distribution but introduced critical challenges related to centralization and malicious miner activity, which threaten the integrity of decentralized consensus. Addressing this gap, this paper proposes the Reputationâbased Consensus Protocol (RCP), a novel framework designed to enhance trust and security in mining pools by incorporating a transparent and dynamic reputation system. Unlike traditional consensus algorithms like Proof of Work (PoW) and Proof of Stake (PoS), which do not differentiate between trustworthy and malicious participants, RCP evaluates miners based on a multiâdimensional scoring mechanism, including historical reputation, willingness reputation, and indirect feedback reputation. This targeted approach allows the network to prioritize reputable miners for block creation, thereby mitigating attacks and improving consensus reliability. By integrating RCP with modular Blockchain frameworks such as Hyperledger, this protocol not only strengthens miner accountability but also sets the foundation for more secure and trustworthy decentralized networks. The proposed model has the potential to redefine mining pool operations and significantly contribute to the evolution of secure Blockchain consensus protocols.
Caixiang Fan, Amirhossein Sohrabbeig, Petr MusıÌlek
Blockchain-based peer-to-peer energy trading enables individuals to directly share renewable energy using Internet of Things technologies. However, it faces significant challenges related to privacy, scalability, and the integration of advanced artificial intelligence. To address these issues, this article proposes zkPET, a secure and intelligent peer-to-peer energy trading framework. zkPET integrates machine learning and blockchain with advanced cryptographic techniques of zero-knowledge machine learning to protect user data while enabling intelligent decision making. In the zkPET framework, the computationally intensive operations of various machine learning models are executed off-chain, and only succinct cryptographic proofs of these computations are uploaded to the blockchain for verification and recording. In addition, a time-series clustering approach is incorporated into federated learning to enhance both inference accuracy and the efficiency of proof generation. Experimental validation using the zero-knowledge proof tool EZKL and a real-world electricity dataset demonstrates the feasibility and effectiveness of zkPET. The results underscore its potential to significantly improve privacy, scalability, and computational efficiency in decentralized energy trading, contributing to the advancement of secure and intelligent energy markets.
Samreen Khan, Suguna Balusamy, Mariya Princy Antony Saviour, Satish Bojjawar · 6 authors
Blockchain Technology (BT) is a promising approach for building scalable & reliable open Decision Support Systems and Artificial Intelligence (AI) could be integrated with BT to present a better approach towards the same. Transparency, trust, data security, and scalability are traditionally regarded as the major deficiencies of DSS models which may impede their efficacy in critical decision-making environments. Seamlessly integrating blockchain with AI-powered DSS can help with data integrity, as well as prevent manipulation and inconsistencies while improving trust with many different stakeholders, because all transactions on a blockchain are transparent immutable, owing to blockchain's decentralized ledger system. The paper investigates a hybrid Blockchain-AI framework that can overcome the main drawbacks of existing DSS systems. This means that AI predictive analytics and machine learning models can match massive datasets for decision making, while blockchain guarantees the integrity, traceability, and decentralization of the data's inputs and outputs. Smart contracts: automate decision processes & ensure tamper-proof execution for defined set of rules & policies In addition, blockchain consensus mechanisms (e.g., Proof-of-Stake, Byzantine Fault Tolerance) can facilitate transparency and verifiability, thus improving conflict-avoiding distributed decision-making. To enable privacy-preserving collaboration between organizations, we introduce a new Decentralized AI Decision Support System (Dai-DSS) framework that integrates federated learning and a distributed ledger. It enables real-time data sharing while preserving data sovereignty, which helps organizations comply with regulatory standards such as GDRP and HIPAA. We are also looking at various optimization techniques such as off-chain scaling solutions (sidechains, Layer-2 protocols, etc.) that improve system efficiency while still keeping the system decentralized. Our research shows that by comparing with the traditional DSS models, the Blockchain-AI integration strengthens system resilience, improves the elimination of single point failure and facilitates open, trustless decision-making processes. Practical applications and the benefits of the proposed model are demonstrated through case studies in healthcare, finance, and supply chain management. Lastly, we address the challenges including computational overhead, interoperability and regulatory challenges, and suggest future research avenues on AI-augmented consensus algorithms and quantum-resistant cryptography. The research demonstrates the ability of Blockchain-AI synergy to revolutionize decision-making systems through decentralized, clean, and scalable operating frameworks across industries.
This research examines the incorporation of Artificial Intelligence (AI) in blockchain consensus algorithms, presenting an extensive overview of current improvements and anticipated effects. We conducted a thorough examination of a diverse array of academic sources, encompassing a broad spectrum of AI methodologies, such as machine learning, deep learning, and reinforcement learning, that have been applied to blockchain consensus mechanisms. The study highlights critical areas where AI can bolster blockchain performance, including enhancing effectiveness, dependability, and flexibility. Despite the promising benefits that AI integration offers, it also presents complexities and potential security risks, including data centralization and increased computational power requirements. In this analysis, we review the risks and examine the proposed mitigation strategies from existing studies, such as federated learning to preserve data privacy, secure multi-party computation to protect sensitive data, and decentralized AI marketplaces to distribute AI resources fairly. This study makes a significant contribution to the field by emphasizing the dual potential of AI to both improve and challenge blockchain systems. By advocating for balanced approaches that prioritize decentralization and security, our findings aim to provide direction for future research and practical applications in this multidisciplinary field.
Solana is an emerging blockchain platform, recognized for its high throughput and low transaction costs, positioning it as a preferred infrastructure for Decentralized Finance (DeFi), Non-Fungible Tokens (NFTs), and other Web 3.0 applications. In the Solana ecosystem, transaction initiators submit various instructions to interact with a diverse range of Solana smart contracts, among which are decentralized exchanges (DEXs) that utilize automated market makers (AMMs), allowing users to trade cryptocurrencies directly on the blockchain without the need for intermediaries. Despite the high throughput and low transaction costs of Solana, the advantages have exposed Solana to bot spamming for financial exploitation, resulting in the prevalence of failed transactions and network congestion. Prior work on Solana has mainly focused on the evaluation of the performance of the Solana blockchain, particularly scalability and transaction throughput, as well as on the improvement of smart contract security, leaving a gap in understanding the characteristics and implications of failed transactions on Solana. To address this gap, we conducted a large-scale empirical study of failed transactions on Solana, using a curated dataset of over 1.5 billion failed transactions across more than 72 million blocks. Specifically, we first characterized the failed transactions in terms of their initiators, failure-triggering programs, and temporal patterns, and compared their block positions and transaction costs with those of successful transactions. We then categorized the failed transactions by the error messages in their error logs, and investigated how specific programs and transaction initiators are associated with these errors. We find that transaction failure rates on Solana exhibit recurring daily patterns, and demonstrate a strong positive correlation with the volume of failed transactions, with bots on Solana experiencing a high transaction failure rate of 58.43%. We identify ten distinct error types in the error logs of failed transactions, with price or profit not met and invalid status errors accounting for 67.18% of all failed transactions. AMMs primarily experience invalid status errors among failed transactions, while DEX aggregators are more commonly affected by price or profit not met errors. Among transaction initiators, bots encounter a broader range of errors due to their high-frequency trading and complex interactions with smart contracts. In contrast, human users experience a more limited range of errors. Based on our findings, we provide recommendations to mitigate transaction failures on Solana and outline future research directions.
Ali Al Maqousi, Ammar Almomani, Ahmad AlâQerem, Mouhammd Alkasassbeh
Blockchain has emerged as a distributed ledger mechanism that enables decentralized recordkeeping and transaction validation with applications in various fields. Recent research has highlighted its potential for managing urban water and energy systems, especially given increasing urbanization and climate-related resource pressures. This chapter explores theoretical dimensions of blockchain for integrated urban water and energy management. It examines the conceptual linkages of blockchain-based smart contracts, distributed consensus, and tamper-proof data exchange with the operational and strategic needs of urban water and energy stakeholders. It discusses the roles of trust, security, and data transparency in facilitating stakeholder cooperation in resource allocation and highlights resilience benefits for water supply networks and energy grids.
Blockchain has been broadly practiced in different markets. It is decentralized, unchangeable, and transparent. Our essay summarizes its practices in three key industries including finance, healthcare, as well as supply chain management. In the former, it benefits efficiency in payment and settlement, preventing greenwashing and optimizing carbon trading. In the middle, it helps in digital health check management, clinical trial transparency, and insurance claim simplification. Blockchain also brings product traceability and multi-party collaboration, while facilitating managing flow. Our work uncovered the common obstacles blockchain practices confronted. Future direction standing on newest research heats and multi-subjects are identified. Hopefully, we can plant theoretical bases and practical guidance in blockchain's coming development and general practices.
Sarthak Nimje, Rushab Taneja, Om Baviskar, Rachana Patil
Educational institutions face significant challenges with event attendance verification, including manual document validation, fraud risks, and delayed approval processes. This study introduces ElizaEdu, a novel decentralized AI agent system utilizing Ethereum blockchain and ElizaOS to automate and secure attendance verification workflows for academic events. The proposed system integrates autonomous AI agents to handle document validation, approval processes, and ERP integration, while utilizing blockchain technology for immutable record-keeping. The system employs four specialized agents: RequestBot for initial verification, VerifyBot for teacher validation, ApproveBot for department head confirmation, and ERPBot for automatic attendance updates. Through a 3-month pilot implementation with 120 students and 15 faculty members, ElizaEdu demonstrated an 85% reduction in verification time, complete elimination of document fraud, and 84% decrease in administrative workload. The system achieves 97.3% accuracy in document validation and 100% data integrity through blockchain verification. This study presents the architecture, implementation details, and evaluation results, demonstrating ElizaEduâs effectiveness in transforming attendance management in educational institutions.
Federated Machine Learning (FML) is an unconventional method that performs decentralized analysis of financial data without the need for sensitive data to be uploaded for secure model training that works across distributed platforms. In this paper, we explored the feasibility of applying FML to the cloud for financial institutions, which ultimately satisfies major privacy-preserving and compliance requirements. We discuss the unique challenges brought up by decentralized settings, including issues with data heterogeneity, communication efficiency, and convergence. As a solution, we present a federated learning framework that enables collaborative training under a cloud infrastructure while ensuring that private data does not leave the local institutions. This is to improve performance, maximize the use of resources, increase speed and scalability of analytical actions in the finance sector. The experimental results demonstrate the efficacy of the proposed system in delivering trustworthy and secure financial predictions, paving the way for considerable improvements in decentralized machine learning for the financial industry.
Henrique Lin, JoĂŁo Santos, Tiago Dias, Miguel Correia
The real estate sector plays a vital role in today's economy and society. However, the current system for managing real estate transactions remains heavily reliant on manual document handling and verification processes, which are often inefficient and vulnerable to fraud, underscoring the need for innovative solutions. This position paper proposes a system that integrates Optical Character Recognition (OCR), Natural Language Processing (NLP), and Verifiable Credentials (VCs) to automate document extraction, verification, and management within real estate transactions. Key goals include (1) a comprehensive workflow to transform diverse document formats into standardized VCs and (2) an automated data matching mechanism to identify inconsistencies and potential fraud indicators. The approach involves using the potential of blockchain and Web3 technologies as a decentralized trust layer to improve data integrity and transparency. This solution holds significant promise for streamlining real estate processes, fostering trust among stakeholders, and establishing a scalable framework for secure and efficient digital transactions.
The decentralized, open-source architecture of blockchain technology, exemplified by the Ethereum platform, has transformed online transactions by enabling secure and transparent exchanges. However, this architecture also exposes the network to various security threats that cyber attackers can exploit. Detecting suspicious behaviors in account on the Ethereum blockchain can help mitigate attacks, including phishing, Ponzi schemes, eclipse attacks, Sybil attacks, and distributed denial of service (DDoS) incidents. The proposed system introduces an ensemble stacking model combining Random Forest (RF), eXtreme Gradient Boosting (XGBoost), and a neural network (NN) to detect potential threats within the Ethereum platform. The ensemble model is fine-tuned using Bayesian optimization to enhance predictive accuracy, while explainable artificial intelligence (XAI) tools-SHAP, LIME, and ELI5-provide interpretable feature insights, improving transparency in model predictions. The dataset used comprises 9,841 Ethereum transactions across 52 initial fields (reduced to 17 relevant features), encompassing both legitimate and fraudulent records. The experimental findings demonstrate that the proposed model achieves a superior accuracy of 99.6%, outperforming that of other cutting-edge methods. These findings demonstrate that the XAI-enabled ensemble stacking model offers a highly effective, interpretable solution for blockchain security, strengthening trust and reliability within the Ethereum ecosystem.
As cryptocurrency transactions continue to grow, detecting scams within transaction records remains a critical challenge. These transactions can be represented as dynamic graphs, where Neural Network Convolution (NNConv) models are widely used for detection. However, NNConv models suffer from model decay due to evolving transaction patterns, the introduction of new users, and the emergence of adversarial techniques designed to evade detection. To address this issue, we propose an automated, periodic hyperparameter optimization method based on proximal policy optimization (PPO), a reinforcement learning algorithm designed for dynamic environments. By leveraging PPOâs stable policy updates and efficient exploration strategies, our approach continuously refines hyperparameters to sustain model performance without frequent retraining. We evaluate the proposed method on a large-scale cryptocurrency transaction dataset containing 2,973,489 nodes and 13,551,303 edges. The results demonstrate that our method achieves an F1 score of 0.9478, outperforming existing graph-based approaches. These findings validate the effectiveness of PPO-based optimization in mitigating model decay and ensuring robust cryptocurrency scam detection.
H.C. Zhang, Shike Li, Shike Li, Hang Bao · 6 authors
The rapid development of blockchain technology has driven the widespread application of decentralized applications (DApps) across various fields. However, DApps cannot directly access external data and rely on oracles to interact with off-chain data. As a bridge between blockchain and external data sources, oracles pose potential risks of malicious behavior, which may inject incorrect or harmful data, leading to trust and security issues. Additionally, with the surge in data requests, the disparity in oracle trustworthiness and costs has increased, making the dynamic selection of the most suitable oracle for each request a critical challenge. To address these issues, this paper proposes a Trust-Aware and Cost-Optimized Blockchain Oracle Selection Model with Deep Reinforcement Learning (TCO-DRL). The model incorporates a comprehensive trust management mechanism to evaluate oracle reputation from multiple dimensions and employs an improved sliding time window to monitor reputation changes in real time, enhancing resistance to malicious attacks. Moreover, TCO-DRL uses deep reinforcement learning algorithms to dynamically adapt to fluctuations in oracle reputation, ensuring the selection of high-reputation oracles while optimizing node selection, thereby reducing costs without compromising data quality. We implemented and validated TCO- DRL on Ethereum. Experimental results show that, compared to existing methods, TCO-DRL reduces the allocation rate to malicious oracles by more than 39.10% and saves over 12.00% in costs. Furthermore, simulated experiments on various malicious attacks further validate the robustness and effectiveness of TCO-DRL
Mohammad Shahab Sepehri, Asal Mehradfar, Mahdi Soltanolkotabi, Salman Avestimehr
Predicting Bitcoin price remains a challenging problem due to the high volatility and complex non-linear dynamics of cryptocurrency markets. Traditional time-series models, such as ARIMA and GARCH, and recurrent neural networks, like LSTMs, have been widely applied to this task but struggle to capture the regime shifts and long-range dependencies inherent in the data. In this work, we propose CryptoMamba, a novel Mamba-based State Space Model (SSM) architecture designed to effectively capture long-range dependencies in financial time-series data. Our experiments show that CryptoMamba not only provides more accurate predictions but also offers enhanced generalizability across different market conditions, surpassing the limitations of previous models. Coupled with trading algorithms for real-world scenarios, CryptoMamba demonstrates its practical utility by translating accurate forecasts into financial outcomes. Our findings signal a huge advantage for SSMs in stock and cryptocurrency price forecasting tasks.
In this study, I propose a method for forecasting the next-day Bitcoin price range using a CART decision tree model, which integrates 124 high-dimensional technical indicators with Twitter-roBERTa sentiment analysis as the 125th feature to enhance prediction accuracy. The experiments utilize Bitcoin market data from the past six years (2019 to 2024) and approximately 58 million Twitter posts. The results demonstrate that the enhanced model, incorporating sentiment analysis, improves the average accuracy from 0.56 in the baseline modelâtrained solely on 124 technical indicatorsâto 0.62, with win rates increasing significantly by up to 45%. Sensitivity analysis further optimizes the sentiment feature weight, confirming the modelâs robustness, and provides an innovative perspective for cryptocurrency market prediction, with future applications extensible through multi-source data fusion.
ABSTRACT The burgeoning demand for blockchain technology in diverse sectors requires advanced optimization methods to improve the performance, security and privacy. However, today common blockchain mechanisms are effected by problems like suboptimal miner selection processes, susceptibility to abnormal transactions and types of attacks affecting nonânegligible parts of the ecosystem, performance bottlenecks and so forth, rendering them far from scalability and realâworld usage. This paper addresses the problem, by introducing a set of sophisticated methods that solve recent issues and enhances the robustness, scalability, confidentiality in blockchain networks. Firstly, we present âDeepMinerâ, a deep learningâbased solution that leverages historical blockchain data samples to infer optimal miner nodes. This method improves the block generation efficiency by optimizing miner node selection in realâtime, which is an essential addition to traditional random or otherwise static methods for selecting miners. Secondly, âAnoBlockâ which uses anomaly detection model to detect fraud in blockchain transactions using the statistical methods like Gaussian mixture models and isolation forests. Thirdly, âOptiChainâ uses data analytics to dynamically optimize blockchain performance by continuously evaluating live network metrics and the transaction throughout. Lastly, âPrivyChainâ which uses privacy preservation techniques such as zeroâknowledge proofs and homomorphic encryption to achieve transaction confidentiality while retaining blockchain transparency. Their solution addresses these issues with a dual approach to protect any sensitive transaction details from being leaked and make it feasible for computations over encrypted data, the result of which aligns blockchain technology with stringent privacy standards.
Bitcoin is the most valuable cryptocurrency and is renowned for its rapid and volatile price fluctuations in comparison to other currencies. This offers potential for the prediction of Bitcoin prices and has attracted the interest of researchers. Twitter (X) is one of the most widely used social media platforms. The aim of this study is to analyse the sentiment expressed in comments about bitcoin on the social media platform X using a variety of machine learning algorithms. A variety of machine learning techniques are used to classify user sentiment towards bitcoin. Moreover, the efficacy of standard bag-of-words and term frequency-inverse document frequency (TF-IDF) methods is evaluated in comparison with machine learning approaches for the purpose of expressing text as numerical vectors. Finally, a keyword ranking was performed to determine the importance of each sentiment in the development of cryptocurrencies. The bag-of-words and TF-IDF methods were used, which facilitate the representation of text-based data. The best result was obtained with the decision trees algorithm (98.74% accuracy) using the TF-IDF method. The bag-of-words method was found to produce better results in general.