Integration of smart contracts and artificial intelligence using cryptographic oracles⋆ Denys Virovets1,†, Sergiy Obushnyi1,†, Bohdan Zhurakovskyi2,†, Pavlo Skladannyi1,∗,† and Volodymyr Sokolov1,† 1 Borys Grinchenko Kyiv Metropolitan University, 18/2 Bulvarno-Kudriavska str., 04053 Kyiv, Ukraine 2 National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute,” 37 Peremogy ave., 03056 Kyiv, Ukraine Abstract Artificial Intelligence (AI) and Distributed Ledger Technology (DLT) together address complex tasks by optimizing and automating business processes and creating innovative new products. Despite their shared digital nature, integrating these two technologies is a challenging process that requires sophisticated solutions. AI relies on large amounts of data and computational power, which are difficult to provide within distributed ledgers. However, the integration of DLT with AI, particularly its interaction with smart contracts, is made possible through the use of an intermediary data exchange and transfer mechanism known as an oracle. This paper analyzes various methods of smart contract interaction with distributed ledgers and hypothesizes the existence of decentralized AI technology. By exploring the methods and techniques for using oracles to facilitate AI and blockchain interaction, we can assess new opportunities for the decentralized economy arising from their combination with AI services and models, and predict the emergence of new decentralized products enhanced by AI technologies. Keywords blockchain, cryptography, smart contracts, artificial intelligence, AI, web3, crypto-economy, decentralization, oracle, decentralized artificial intelligence 1 1. Introduction Homomorphic Encryption, Zero-Knowledge Proofs, Multi- Party Computation, and Quantum-Resistant 1.1. Smart contracts and cryptography: Cryptography. ensuring security in decentralized For users of smart contracts, cryptography represents systems a commitment scheme that allows for the selection of execution conditions in secret, with the ability to reveal Smart contracts are traditionally defined as programs that these conditions later [7]. This is made possible by using operate on blockchain technology [1], requiring a Zero-Knowledge Proof (ZKP) technology, which enables decentralized virtual machine capable of programming transaction verification without identifying participants and data processing [2]. The defining characteristic of or data. Cryptographic methods used for smart contracts, smart contracts is that, once deployed on the virtual including Homomorphic Encryption, Zero-Knowledge machine, they execute autonomously according to the Proofs, Multi-Party Computation, and Quantum-Resistant program’s instructions, without control by any user. The Cryptography, can be employed in projects focused on execution of smart contracts can result in the creation of decentralized identity, decentralized finance, and medical new types of digital assets, allowing for full or partial research where privacy is paramount. In this way, management by the user of the smart contract. These cryptography ensures secure communication between characteristics of smart contracts enable the development parties and guarantees the security of data involved in the of economic and financial digital systems [3] that utilize operation of smart contracts. various other digital technologies for data collection [4], The Promising and Complex Integration of AI analysis [5], and asset management [6]. Trust between Technologies in Smart Contracts participants in smart contracts is established through Among the most promising and complex technologies cryptographic algorithms and protocols that ensure to be applied within smart contracts are AI technologies, secure communication and data protection. This security which are associated with large volumes of data and is achieved through the application of various advanced tools for their analysis. The concept of AI cryptographic methods in smart contracts, such as CQPC-2024: Classic, Quantum, and Post-Quantum Cryptography, August 0000-0003-4934-8377 (D. Virovets); 0000-0001-6936-955X 6, 2024, Kyiv, Ukraine (S. Obushnyi); 0000-0003-3990-5205 (B. Zhurakovskyi); 0000-0002-7775- ∗ Corresponding author. 6039 (P. Skladannyi); 0000-0002-9349-7946 (V. Sokolov) † These authors contributed equally. © 2024 Copyright for this paper by its authors. Use permitted under d.virovets.asp@kubg.edu.ua (D. Virovets); s.obushnyi@kubg.edu.ua Creative Commons License Attribution 4.0 International (CC BY 4.0). (S. Obushnyi); zhurakovskybiyu@tk.kpi.ua (B. Zhurakovskyi); p.skladannyi@kubg.edu.ua (P. Skladannyi); v.sokolov@kubg.edu.ua (V. Sokolov) CEUR Workshop ceur-ws.org ISSN 1613-0073 39 Proceedings encompasses a variety of data analysis and management 1.2. The role of distributed ledger technologies, ranging from simple machine imitations of technology in static and dynamic human intelligence, as envisioned by Alan Turing, to deep models learning technologies for training neural networks. Modern AI technologies can utilize models for Natural Distributed ledger technology (DLT) has demonstrated its Language Processing (NLP), Computer Vision, Expert effectiveness in static models for accounting financial Systems, Genetic Algorithms, and Multi-agent Systems, transactions, managing reputation systems, handling which enhance and improve AI models. agreements, and confirming ownership rights. In contrast, AI algorithms require dynamic data updates to create and Attempts to integrate smart contracts with AI lead to maintain training models. Training models based on smart several challenges due to their inherent complexity, which contracts cannot guarantee accurate predictions, partly due require resolution [8]. The primary issue with smart to the lack of support for fixed-point arithmetic and contracts is the technical difficulty of storing and directly differences in computational architectures. One potential utilizing large amounts of data in distributed ledgers. It is solution is to apply the Naive Bayes algorithm in smart well known that AI can significantly contribute to the contracts, which requires probability calculations using development of decentralized products and asset floating-point numbers based on Gaussian probability [11]. management by preprocessing and analyzing data AI, which relies on a vast array of tools for model through normalization and cleansing before submitting it training and extensive databases, cannot operate within a to the blockchain structure, thereby reducing the blockchain environment in its current form due to the excessive load on the ledger. Additionally, machine resource demands needed to perform the necessary learning techniques have been developed that can be used computations and model creation. To integrate the to build and refine smart contract code, including through decentralized protocol environment with AI, several the use of NLP [9]. AI enhances the adaptability of smart solutions have been proposed, which will be discussed contracts by incorporating logic, neural graphs, and below. neural networks through its integration into the smart One approach to organizing the interaction between AI contract code or external usage to verify and ensure and smart contracts is through the use of specific contract integrity. Tools such as deep learning mechanisms known as oracles. Oracles enable the frameworks like TensorFlow are considered promising integration of AI systems with smart contracts, offering a for their integration into smart contracts [10]. range of possibilities [12]. The traditional form of an oracle that facilitates this interaction is depicted in Fig. 1 Figure 1: Traditional interaction of AI with smart contracts via an oracle 1.3. Oracles as mediators between smart contracts cannot access data outside the blockchain blockchain and AI environment. The primary function of a blockchain oracle is to send requests, verify, and authenticate external data Oracles are third-party services with smart contracts that sources, and deliver this data to the user’s smart contract act as intermediaries between blockchain and external [8]. The table below outlines the functions of oracles that systems, allowing smart contracts to access off-chain data, facilitate the collection and provision of data for use in including data from AI models. By design, blockchains and smart contracts and AI models. 40 1.4. Oracle functions and their decentralized oracles. Each oracle in this system gathers characteristics data from independent sources and compares it to ensure its accuracy [8]. These intelligent smart contracts, capable of Table 1 responding to monitored conditions, acquire the Oracle functions in data provision for intelligent smart characteristics of dynamic smart contracts. They can contracts autonomously make decisions after analyzing the Oracle Functions Function Characteristics information and then send commands and queries to other Data Collection for AI – Collects and processes systems [15]. In these interactions, AI can be effectively external data for centralized used to execute logic for monitoring states and events and and decentralized AI models. making appropriate decisions. – Transfers external data to AI AI models may involve using services to collect algorithms. meteorological, agricultural, seismic, and other real-world Data Verification – Ensures the reliability and authenticity of data, data, which is then structured for subsequent monitoring preventing manipulation and and analysis [16]. Distributed ledgers alone cannot perform fraud. these tasks without special tools. On the other hand, smart – Verifies data sources, reducing contracts can manage interactions with AI models. For risks associated with example, the SingularityNET project represents a inaccurate or falsified data. Integration with AI – Transfers and integrates decentralized platform that allows the creation and sharing analytical predictive data from of AI models and their monetization in a marketplace based AI into smart contracts. on the ERC20 standard within the Ethereum network. Similar solutions are offered by projects such as Namahe, To implement a smart contract, it may be necessary to Neuromation, TraDove, AdHive, ATN, Cortex, and NAM obtain information from AI based on the analysis of external [17]. The table below lists some services that offer solutions data, such as prices and exchange rates, sensor data, sports for integrating AI into smart contracts. event results, flight information, insurance claims, and so The literature identifies three main methods for AI forth. This can be achieved through an intelligent oracle, interaction with traditional smart contracts: Edge AI, AI- which acts as an on-chain agent providing information in centric smart contracts, and Swarm Intelligence. The table response to queries to the AI model [13]. The intelligent below presents these methods along with their oracle enables smart contracts to utilize AI analytics and characteristics, as well as other theoretically possible ways forecasts for automated decision-making and contract AI could interact with smart contracts. The concept of execution, while also ensuring the verification of the Swarm Intelligence Smart Contracts, based on swarm accuracy of the information provided to the AI models [14]. intelligence principles for decision-making and task An example of such an intelligent oracle is the execution, appears particularly interesting as it integrates Chainlink project, which represents a system of well with the logic of smart contracts. Table 2 Classification of smart contracts by methods of interaction with AI (compiled based on the source [10]) Method of AI Interaction Characteristics with Smart Contracts Edge AI Smart Contracts Smart contracts that combine AI technologies with edge computing capabilities, enabling data processing and contract execution closer to the data source. This approach reduces latency and enhances efficiency. Key features include: – Local data processing. – Reduced communication delays. – Improved overall performance. AI-Centric Smart Contracts Smart contracts that integrate AI capabilities with automated contract execution on the blockchain. This allows for the creation of intelligent contracts that can analyze data, make complex decisions, and automatically execute actions based on the results. Key features include: – Data analysis capabilities. – Decision-making based on AI insights. – Automated contract execution. Swarm Intelligence Smart Smart contracts that utilize swarm intelligence principles for decision-making and task execution. Contracts Swarm intelligence refers to the collective behavior of decentralized, self-organizing systems, typically observed in natural systems such as ant colonies, bee swarms, or bird flocks. In the context of smart contracts, this means using multiple agents interacting with each other and their environment to achieve a common goal or perform tasks. Key features include: – Collective decision-making. – Decentralized coordination. – Task execution based on swarm dynamics. 41 1.5. Direct integration of AI into smart consideration, as training AI models on the blockchain is contracts currently a rather cumbersome process [8]. However, some projects have achieved notable success by adapting Another technology is the direct integration of AI into networks to leverage AI capabilities, including creating smart contracts without using an intermediary like oracles, specialized networks with their consensus algorithms, which eliminates the trust issue. This approach involves such as Proof of Intelligent Mining (PoIM) in Matrix and utilizing ASIC-resistant frameworks through specialized Delegated Proof of Stake (DPoS) in Cortex. The PoIM consensus algorithms and standards for ensuring consensus algorithm, developed for intelligent mining operational compatibility between infrastructure and deep and operating a decentralized network based on machine learning tools. An example of such technology is the Cortex learning technology, increases the number of project, which employs the Material Representation Tool transactions per second and facilitates the use of (MRT) technology compatibility standard and the Cuckoo Cycle consensus mechanism. intelligent models within the network. Embedding artificial intelligence directly into smart contracts in an on-chain mode requires careful Figure 2: Generalized Structure of Data Transmission in Smart Contracts Using an AI Oracle The oracle structure for interacting with smart contracts handling large volumes of data. The Matrix platform, on the involves the smart contract itself, a tool for retrieving data other hand, employs the Proof of Intelligent Mining (PoIM) from an AI model, a database, and a dynamic data source for algorithm, which leverages AI for optimizing mining the model. Upon meeting certain conditions, the smart processes and consensus tasks. Matrix provides components contract sends a request to the oracle, which in turn returns for semantic and syntactic analysis, security verification of data from the AI model to the smart contract. Popular smart contracts, and identification of issues in transaction Oracle services that can work with AI models include models [8]. These platforms’ solutions enable the creation Oraclize (Provable), Town Crier, Reality K, Witnet, and of decentralized AI models for implementation in DAO Chainlink, among others. An example of using Oraclize is smart contracts to optimize decision-making and manage the Etherisc project, which uses an oracle to access flight participant relationships. delay data to automate insurance payouts in case of flight An example of an AI oracle is the GainForest project, delays. Another project, Ethersquares, implements logic for where a smart contract is used to distribute bets based on sports betting, where users can verify the accuracy of data about deforestation status using artificial intelligence received data through the oracle [17]. [18]. The oracle automatically analyzes and evaluates The Cortex and Matrix platforms offer solutions for satellite images and detects deforestation issues using integrating AI models into the blockchain and subsequently remote sensing algorithms. Based on the collected and using them in smart contracts. Users can incorporate the analyzed information, bets are redistributed among proposed Cortex AI system into existing smart contracts. In participants. Fig. 3 illustrates the operation of such an the Cortex project, this is achieved through the Cuckoo oracle. Cycle consensus algorithm, known for its high efficiency in 42 Figure 3: Machine-powered incentive system in GainForest (compiled based on the source [19]) The blockchain-based platform for artificial intelligence smart contracts allows to make and execute decisions systems, SingularityNET, provides the capability to automatically based on AI [8]. Other mentioned services integrate autonomous agents into smart contracts, offer methods for integrating smart contracts with AI connecting them to data exchange channels and AI systems models for use in operations or provide mechanisms for that interact with modules [8]. This technology enables collaboration, where AI models are applied to manage smart contracts to interact with real-world sensors and incentive systems, reward distribution, and evaluate devices, obtaining information simultaneously with AI- participation in joint projects. driven decisions on subsequent actions. The structure of Table 3 Classification of smart contracts by AI application areas Smart Contracts AI Characteristics Interaction Technology Federated Learning Smart contracts utilizing federated learning for training AI models on decentralized data without Smart Contracts centralizing it in one location. Data remains on local devices, and models are trained locally, exchanging only model updates. Reinforcement Learning Smart contracts incorporating reinforcement learning algorithms for decision-making. RL models Smart Contracts are trained based on rewards and penalties for actions, optimizing decision-making strategies. Predictive Analytics Smart contracts incorporating reinforcement learning algorithms for decision-making. RL models Smart Contracts are trained based on rewards and penalties for actions, optimizing decision-making strategies. Natural Language Implementation of natural language processing technologies for understanding and analyzing Processing (NLP) Smart text data. Smart contracts can interact with users through text interfaces, analyze documents, and Contracts automatically execute contract terms based on textual information. Computer Vision Smart Use of computer vision technologies for analyzing images and video. Smart contracts can make Contracts decisions based on visual data, such as automatically detecting defects on a production line. Multi-agent Systems Employment of multi-agent systems where agents interact to achieve a common goal. Each agent Smart Contracts can operate autonomously while exchanging information with other agents, facilitating the creation of complex and adaptive systems. Autonomous Negotiation Application of AI for automatic negotiation and agreement formation between parties. Smart Smart Contracts contracts can autonomously negotiate terms based on predefined criteria and the interests of the parties. 43 Table 4 AI services integrated with blockchain and their possible uses in smart contracts (compiled based on the source [8, 19–21]) Service General Description and Functions Usage in Smart Contracts Cortex Decentralized Artificial Intelligence Platform that supports Analysis of decentralized data for decision- AI smart contracts and AI inference. making and process automation Matrix AI Intelligent contracts service that combines smart contract Using AI in smart contracts for management and Network capabilities with AI elements, allowing for handling large interactions volumes of transactions and integrating AI services into dApps. SingularityNET Decentralized AI marketplace and AI Publisher, running Data collection and analysis, process automation on blockchain with Distributed Atomspace (DAS) to represent and store primary knowledge for AI agents, and it encapsulates any computational results achieved during their execution Namahe Responsible Supply Chain based on Blockchain & Artificial Organizing production and supply of goods and Intelligence for managing the supply chain using smart services contracts and AI on the Hyperledger Sawtooth platform. Fetch.AI Decentralized platform for creating commercial models Using autonomous economic agents for process combining AI and smart contracts based on Neo automation and asset management. blockchain technology. Numerai Hedge fund platform based on artificial intelligence for AI model competitions in the Numerai predicting financial markets and creating AI models. Tournament, fund management, and strategic investment decisions. Eligma AI-powered chatbot for asset value prediction. Process automation for asset management, decentralized loyalty programs for communities, and data analysis for decision-making. Peculium Analytical platform using AI and machine learning for Developing automated investment strategies for analyzing market trends and decision-making. asset management. DeepBrain AI computational platform with access to a distributed Analyzing large volumes of data, modeling and Chain (DBC) network with unlimited scalability. Analyzing large forecasting, integrating AI models into business volumes of data, modeling and forecasting, integrating AI processes, and sharing computational resources. models into business processes, and sharing computational resources. Neural Neural network algorithms for AI for forecasting, analysis, Forecasting future events, decision-making, risk and asset management. management, and business process optimization BurstIQ The platform for simplifying the collection, evaluation, and Managing medical data, storing confidential analysis of data with a consensus mechanism at the participant information, providing decentralized storage level. medical services, and managing medical records. LifeGraph Machine learning platform for generating relational maps Collecting, monitoring, and analyzing mental of verified data elements collected from “consent health and behavioral data, solutions for contracts”. personnel, and resource management. Vytalyx Decentralized database of clinical and non-clinical data Real-time biometric monitoring of participants, managed by AI. and personalized medical solutions. Neuromation Ecosystem for creating AI models using distributed Providing computational resources for training computing and incentive models. AI models. Synapse AI Machine learning model training platform and creation of Decentralized data exchange and management, autonomous agents with data sharing service. creating autonomous agents for process optimization. StyleGAN2 The platform for generating and managing digital art Automated solutions for profit distribution, using AI and Generative Adversarial Networks (GAN). trend prediction NFT management, and art generation. NFTGAN NFT Art Generation platform using GAN neural network Creating and managing NFTs, content architecture, combining two generators and monetization, and participant engagement, and discriminators to generate texts, images, and sounds. managing intellectual property rights. 44 Kojii.ai AI platform to enhance public perception of artworks, Intelligent chatbots for participants, organizing train artists, and gather information for AI models. communications with communities, investment decisions, and creating recommendations for project participation. PrimeIntellect Service for training large AI models using distributed Tools for forecasting and analyzing market data, resources, reducing costs, and democratizing AI asset and risk management strategies, and development. business process automation. Nous Incentive system using tokens for participation in AI Tools for asset management and large data Research model creation. volumes, cybersecurity protection. DanKu Service for creating and accessing machine learning Collaborative development of AI models for models with an incentive system and evaluation using token-based rewards. smart contracts. The interaction between AI and smart contracts raises Standardization of smart contracts is crucial for uniform several issues, including questions about AI’s task typing for AI models to handle various data classes, responsibility regarding the consequences of executing analyze their performance, and create necessary a contract by an autonomous agent or providing false or integration modules for AI. Through these modules, incorrect information for the execution of a smart smart contracts can interact with each other and contract [22]. Due to these risks, the use of AI in DAOs organize intelligent decentralized systems with self- brings up the need for standardization of smart contracts learning and environmental adaptation capabilities. and crypto-assets for their compatibility with AI models. Additionally, the standardization of intelligent smart Known token standards such as ERC-20, ERC-721, and contracts will ensure the automated and autonomous ERC-1155 are used in economic models or reward development of intelligent systems. Inter-network systems created by AI. AI can also create unique content integration and data exchange with the physical world and release it as NFTs. The idea of more global use of AI should be managed by oracles adapted to intelligent in smart contracts has led to the development of specific networks, which presents new challenges. The use of smart contract standards designed to facilitate their different types of AI, with varying structures and integration with AI. Below are some of these standards complexities, requires finding new integration solutions. along with their characteristics. 2. Conclusion Table 5 Smart contract standards in the Ethereum environment, The integration of smart contracts and AI undoubtedly adapted for interaction with AI offers significant advantages, as blockchain provides the potential for cryptographic storage of large volumes of Standard Characteristics data and its use for training AI models, securely ERC-2362 Standard for decentralized oracles, which transmitting data between digital systems, and can be used to obtain external data for implementing economic model logic. Despite the smart contracts, including data from AI or complexities involved in integrating smart contracts and forecasting models. ERC-3001 Standardizes the integration of AI, methods and solutions for such integration already decentralized oracles, allowing the exist. In this context, the need for standardizing the incorporation of external data from AI to methods of linking smart contracts and AI arises to provide more accurate and informed enable the mutual integration of various solutions decisions. (services). The assertion that “If Smart Contracts Are the ERC- Standards are used for proof of work Body of the Digital Deal Era, Artificial Intelligence is the 8000+ (PoW) and other consensus mechanisms, Mind” aptly describes the importance of intelligent which can be utilized for managing oracles and cryptography in ensuring the interaction resources and computations with AI. between systems. ERC-7007 A specifically designed standard for verified AI-generated content. Allows for the tokenization of AI-generated content References with verification of source and generation [1] V. Zhebka, et al., Methodology for Choosing a parameters [23]. Consensus Algorithm for Blockchain Technology, ERC-7231 Standard for tokens of verified AI- in: Digital Economy Concepts and Technologies, generated content, aimed at improving data sovereignty and user control over vol. 3665 (2024) 106–113. digital identities. Integrates Web2 and [2] S. Obushnyi, et al., Ensuring Data Security in the Web3 identities into a single non-fungible Peer-to-Peer Economic System of the DAO, in: token (NFT), enabling users to own, Cybersecurity Providing in Information and manage, and monetize their online Telecommunication Systems II, vol. 3187 (2021) presence across various platforms. 284–292. [3] D. Virovets, et al., Soul Bound Token as Digital Twins in Peer-to-Peer Economic Systems, in: 45 Digital Economy Concepts and Technologies, vol. International Joint Conference on Artificial 3665 (2024) 97–105. Intelligence. Doctoral Consortium (2022) 5851– [4] R. Marusenko, V. Sokolov, I. Bogachuk, Method of 5852. doi: 10.24963/ijcai.2022/828. Obtaining Data from Open Scientific Sources and [17] V. Lopes, L. A. Alexandre, An Overview of Social Engineering Attack Simulation, Advances Blockchain Integration with Robotics and in Artificial Systems for Logistics Engineering, vol. Artificial Intelligence (2018). doi: 135 (2022) 583–594. doi: 10.1007/978-3-031-04809- 10.48550/arXiv.1810.00329. 8_53. [18] AI Smart Contracts. The Age of A.I. Powered [5] F. Kipchuk, et al., Assessing Approaches of IT Incentives (2017). URL: Infrastructure Audit, in: IEEE 8th International https://medium.com/gainforest/predictive-smart- Conference on Problems of Infocommunications, contracts-dc15b9986d8c Science and Technology (2021) 213–217. doi: [19] From Imagination to Reality: 6 Companies 10.1109/picst54195.2021.9772181. Building the Future with AI and Blockchain (2023). [6] Aimeta Bridging Communities, AI and Metaverse URL: https://www.linkedin.com/pulse/from- Projects, Whitepaper, Version 1.2 (2023). imagination-reality-6-companies-building-future- [7] M. C. Reis, Extending Smart Contracts Security ai-blockchain-yusgf Through Cryptographic Protocols, Faculdade de [20] S. Shahriar, K. Hayawi NFTGAN: Non-Fungible Computação da Universidade Federal de Token Art Generation Using Generative Uberlândia, Uberlândia, Brasil (2020). Adversarial Networks. (2021). doi: [8] I. Reshi, et al., AI-Powered Smart Contracts: The 10.48550/arXiv.2112.10577. Dawn of Web 4, TechRxiv (2023). doi: [21] A. Anupama, S. Deekshith, S. Karthik, Crypto AI: 10.36227/techrxiv.221894 38.v1. Digital Nostalgic Art Generation Using GAN and [9] S. Badruddoja, et al., Making Smart Contracts Creation of NFT Using Blockchain, J. Emerging Smarter, (2021) 1–3. doi: 10.1109/ICBC51069. Technol. Innov. Res. 9(7) (2014) 217–220. 2021.9461148. [22] M. Sako, Contracting for Artificial Intelligence. [10] M. Sai Mohan, T. Swamy, V. Chandu Reddy, Communications of the ACM 66 (2023) 20–23. doi: Strengthening Smart Contracts: An AI-Driven 10.1145/3584707. Security Exploration. Global Journal of Computer [23] C. So, et al., ERC-7007: Verifiable AI-Generated Science and Technology, 23(D2) (2023) 57–67. Content Token. An ERC-721 extension for [11] S. Badruddoja, et al., Making Smart Contracts verifiable AI-generated content tokens using Zero- Predict and Scale, Fourth International Conference Knowledge and Optimistic Machine Learning on Blockchain Computing and Applications Techniques (2023). URL: (BCCA) (2022) 127–134. doi: 10.1109/BCCA55292. https://eips.ethereum.org/EIPS/eip-7007 2022.9922480. [12] V. Papadouli, V. Papakonstantinou, A Preliminary Study on Artificial Intelligence Oracles and Smart Contracts: A Legal Approach to the Interaction of Two Novel Technological Breakthroughs, Computer Law & Security Review 51 (2023). [13] Oracle. Gain Access to Data Stored Outside of the Blockchain. URL: https://github.com/fravoll/solidity- patterns/blob/master/docs/oracle.md [14] Blockchain Oracles: Bridging External Data Sources with AI Smart Contracts (2024). URL: https://medium.com/sample-dcentai- blog/blockchain-oracles-bridging-external-data- sources-with-ai-smart-contracts-e7fffac0c323 [15] S. Terzi, et al., Blockchain 3.0 Smart Contracts in E-Government 3.0 Applications (2019). doi: 10.48550/arXiv.1910.06092. [16] S. D. Facchini, Decentralized Autonomous Organizations and Multi-agent Systems for Artificial Intelligence Applications and Data Analysis, Proceedings of the Thirty-First 46