The Ethical Dimensions of ICT In The Digital Transformation Of Cities

Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 96 THE ETHICAL DIMENSIONS OF ICT IN THE DIGITAL TRANSFORMATION OF CITIES Kristina Isakzay, Martina Bogdanova, Maria Lachova, Daniela Ilieva Law and Internet Foundation, Sofia, Bulgaria Abstract This paper examines the ethical dimensions of integrating Information and Communication Technology (ICT), specifically digital twin technology, in the digital transformation of cities. As digital twins— enhanced by artificial intelligence for data analysis and decision-making—become increasingly prevalent in urban planning, it is crucial to address the ethical challenges they pose to ensure sustainable and inclusive urban development. Using the UP2030 Project in Istanbul as a case study, the paper explores key ethical issues such as data privacy, security, and biases in AI models. A comprehensive ethical analysis is conducted alongside a multi-faceted methodology that includes a literature review, legal framework analysis, and an in-depth examination of project-specific documentation. The findings underscore the need for stringent data protection measures, equitable representation across demographic groups, and careful management of AI-driven decision-making systems to prevent the erosion of human oversight. The paper argues that integrating AI-enhanced digital twin technology ethically into urban planning is essential for addressing energy consumption, climate resilience, and mobility challenges. Istanbul’s case demonstrates how technology, when governed fairly, can drive positive change, achieving goals like carbon neutrality by 2050 while setting a global example. This approach not only ensures responsible urban development but also serves as a model for the sustainable and ethical evolution of smart cities worldwide. Keywords: Urbanization, Digital Twin, ICT, Ethical Considerations, Smart Cities, Istanbul, UP2030 Project 1. INTRODUCTION Urbanization is progressing at an unprecedented rate, with 68% of the world population expected to reside in urban areas by 2050. 95% of the urban expansion is expected to take place in the developing world, putting tremendous pressure on these regions to cope with the new human development challenges. (United Nations Department of Economic and Social Affairs, 2018) As cities are responsible for approximately 75% of global emissions, they are central to meeting the 2030 and 2050 emission reduction targets of the European Green Deal, set out to respond to the Paris Agreement. (United Nations Environment Programme, n.d.) The Mission for climate-neutral and smart cities calls for even faster action by cities to meet neutrality. The current scenario requires cities to find ways to manage new challenges. Cities worldwide have started to look for solutions which enable high-quality urban services with long-term positive effects on the economy. (Pierce, et al., 2017) Many of the new approaches related to urban services have been based on harnessing technologies, including ICT helping to create what some call “smart cities.” (Albino, et al., 2015) The smart city is a growing phenomenon where the ultimate goal is to increase city resilience, sustainability and quality of life through smart innovations. ( Pierce, et al., 2017) The smart city concept includes most technology-based innovations that tackle planning, development, operation, or management of urban activities. (Moggi & Dameri, 2021) Among other things, it can be utilized to catalyse efficient resource allocation, provide security, improve operational efficiency and facilitate human activities. (Mohammadi & Taylor, 2017) The idea is that vast amounts of data are being produced from the movement and behaviour of people and communities together with the infrastructure in highly Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 97 connected cities. ( Pierce, et al., 2017) This data can be utilized to maximize services (Mohammadi & Taylor, 2017) provided in the top layer of the city. (Pierce, et al., 2017) 1.1. Digital twin concept as part of thedigital transformation Further technology that might contribute towards smart city goals is the digital twin. A digital twin is a digital representation of a physical asset based on timely data that allows for analysis, insight, decisionmaking, sense-making and interventions in the physical world. (El-Agamy, et al., 2024) Digital twins involve developing technology, but also designing processes and tools that are run by people, enable collaboration, support human decision-makers and create positive outcomes for society. Digital twins scale up and improve through an ongoing cycle of prototyping, deployment and reflection. Digital twins bridge the physical world around us with a world of digital data and models, which represent the state of the asset, how it is performing and potentially predict its future state, enabling better interventions in the physical world. (El-Agamy, et al., 2024) What distinguishes a digital twin from any other digital model is its connection to the physical twin. Based on data from the physical asset or system, a digital twin unlocks value principally by supporting improved decision-making, which creates the opportunity for positive feedback into the physical twin. (El-Agamy, et al., 2024) 1.2. UP2030 Digital Twins in Istanbul: A Use Case Istanbul, a rapidly growing metropolis with a population nearing 16 million, faces significant urban challenges, particularly as it contends with the exacerbating effects of climate change. (UP2030 Project, n.d.) The city is increasingly susceptible to severe heat waves, which are intensified by the inefficiencies in the design and energy performance of its building stock, especially within economically disadvantaged neighborhoods. These inefficiencies contribute to elevated levels of energy consumption and greenhouse gas emissions. Furthermore, persistent issues such as traffic congestion and overcrowded public transportation systems strained Istanbul's urban mobility infrastructure. (UP2030 Project, n.d.) In response to these multifaceted challenges, Istanbul has committed to the C40 "Deadline 2020" initiative, which mandates a trajectory towards carbon neutrality by 2050. (IPA Istanbul, 2022) As part of this commitment, the city participates in the UP2030 Project, which seeks to refine urban planning methodologies through the integration of advanced computational technologies. Central to this effort is the use of digital twin technology, which creates detailed virtual replicas of physical urban environments. These digital twins allow city planners to simulate, monitor, and optimize various aspects of urban infrastructure in real-time. The integration of artificial intelligence (AI) is crucial in enhancing the capabilities of these digital twins. AI algorithms, particularly those based on deep learning, process vast amounts of data collected from sensors and other sources to generate accurate predictions and insights. For instance, AI-driven models such as Recurrent Neural Networks, Graph Neural Networks, and Multi-layer Perceptrons are employed to forecast hourly energy consumption and electricity generation in buildings, as well as to analyse urban mobility patterns in targeted neighbourhoods (UP2030 Project, n.d.) The UP2030 initiative prioritizes three key areas: 1. Decarbonization of buildings and transport systems: This includes the strategic integration of photovoltaic systems within both residential and public urban spaces. 2. Sustainable transportation: The initiative emphasizes the deployment of e-bikes powered entirely by renewable energy, ensuring that these solutions are accessible and equitable. 3. Assessment of social impacts: The project will rigorously evaluate the social benefits, particularly in reducing energy poverty and addressing health risks associated with extreme heat. (UP2030 Project, n.d.) 1.3. Ethical dimensions of the digital twin concept in Istanbul as a use case The implementation of the digital twin presents substantial potential for addressing urban challenges, yet it simultaneously brings forth significant ethical considerations. (El-Agamy, et al., 2024) In the Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 98 context of Istanbul’s UP2030 Project, these ethical dimensions are salient. The application of digital twins to develop positive energy neighbourhoods and optimize urban mobility necessitates meticulous management to ensure data privacy and security. Given the city's diverse population and complex socioeconomic landscape (Istanbul Development Agency, 2015), there exists a pronounced risk of bias and discrimination if the models fail to adequately represent all community segments. (El-Agamy, et al., 2024) Crucial to the ethical deployment of the digital twin technology are considerations of accountability and transparency. While digital twins are sophisticated representations of the physical world, the decisions they inform are derived from complex data processing and predictive models that can influence realworld outcomes. This means that the algorithms and data used in digital twins must be transparent and the decision-making processes accountable to ensure that stakeholders trust the technology. (Helbing & Argota Sánchez-Vaquerizo, 2022) Clear documentation and communication regarding data collection, usage, protection, and decision-making processes are imperative for responsible and inclusive technology deployment. (Sanchez, et al., 2024) Moreover, while digital twins are designed to reflect and manage the physical world, there is a risk that the insights and decisions they generate could be granted undue authority, potentially overriding human judgment. This can occur if stakeholders overly rely on the digital twin's recommendations without sufficient human oversight, leading to decisions that may be inaccurate, biased, or even vulnerable to manipulation or hacking. Such scenarios underscore a significant ethical challenge, as the technology could inadvertently prioritize its data-driven outputs over human values and nuanced decision-making. (Helbing & Argota Sánchez-Vaquerizo, 2022) By proactively addressing these ethical dimensions, Istanbul can not only achieve its carbon-neutral objectives but also establish a benchmark for the responsible and inclusive implementation of digital twin technology in urban environments. 1.4. Thesis This paper explores how the ethical integration of digital twin technology into urban planning can transform a city’s approach to tackling energy consumption, climate resilience, and mobility challenges, using the UP2030 Project pilot city Istanbul as a use case. Addressing ethical concerns such as data privacy, bias prevention, and broader societal impacts is crucial for cities to benefit from digital twin technology. By focusing on these ethical dimensions, cities can pave the way for a sustainable and inclusive urban future. Istanbul’s case study shows how technology can drive positive change, underscoring the importance of ethical governance. Successfully implementing digital twins in Istanbul aims not only for carbon neutrality by 2050 but also to set a global example. This demonstrates that combining technological innovation with ethical considerations can build resilient and fair urban environments, serving as a model for the responsible and sustainable development of smart cities worldwide. 1.5. Research Gap and Contribution While existing research extensively covers the technical aspects and potential benefits of digital twin technology in urban planning, there is a noticeable gap in addressing the ethical implications of its deployment. Most studies focus on the efficiency and performance improvements offered by digital twins, often overlooking critical ethical concerns such as data privacy, security, and bias in AI models. This study diverges from this trend by providing a comprehensive examination of these ethical dimensions, using the UP2030 Project in Istanbul as a case study. By doing so, it aims to fill the gap in the literature regarding the socio-ethical impacts of digital twin technology in urban settings. The research highlights the potential risks and proposes a framework for ethical integration, ensuring that the benefits of digital transformation are realized in a manner that is inclusive and sustainable. This focus on ethical considerations positions this study as a necessary complement to the predominantly technical discourse in the current body of research. Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 99 2. MATERIALS AND METHODS 2.1. Materials The materials utilized in this study encompass a wide array of sources and documents, which are essential for a thorough examination of the ethical dimensions of digital twin technology in urban transformation. The key materials are as follows: 2.1.1. Scientific Literature A thorough review of existing academic literature was conducted to gather theoretical insights and empirical data on digital twin technology, smart city concepts, and the ethical implications of ICT integration in urban environments. This literature review encompassed peer-reviewed journal articles, conference papers, and relevant books, which provided foundational knowledge and context for the study, particularly in relation to ethical concerns such as data privacy, security, and algorithmic bias. 2.1.2. Ethical Frameworks and Theoretical Sources To specifically address the ethical dimensions of ICT and digital twin technologies, the study incorporated materials that provide ethical frameworks and theoretical perspectives relevant to these technologies. This included sources on applied ethics and international standards on ICT ethics. These materials were critical in forming the basis for the ethical analysis conducted in the study. 2.1.3. Legal Acts and Regulatory Documents An analysis of pertinent legal frameworks and regulations was carried out to understand the compliance requirements and ethical guidelines governing data privacy, security and ethical deployment of AI systems and machine learning (ML) based technologies. 2.1.4. Project Documentation from UP2030 Detailed documentation from the UP2030 Project in Istanbul was critically analyzed. This included technical specifications, implementation strategies, and progress reports, which provided concrete examples of how digital twin technology was deployed in a real-world urban setting. These documents were not only used to understand the technical aspects but also to identify and assess ethical challenges specific to the project's implementation. 2.2. Methods The methodology employed in this study was multifaceted, combining various analytical techniques to comprehensively address the research questions. The primary methods are detailed below: 2.2.1. Literature Review An extensive review of scientific literature was conducted to build a theoretical framework for the study. This review involved systematic searches in academic databases using keywords related to digital twins, smart cities, data privacy, ethics, and ICT. The selected literature was critically appraised to identify prevailing theories, methodologies, and findings relevant to both the technical and ethical dimensions of the study’s focus. 2.2.2. Ethical analysis The study employed an ethical analysis methodology specifically to examine the ethical implications of digital twin technologies within the context of urban planning. This involved applying established ethical frameworks to the issues identified in the literature review and project documentation. The analysis focused on key ethical concerns such as data privacy, security, algorithmic bias, transparency, and accountability. 2.2.3. Legal Analysis The legal analysis involved a detailed examination of EU regulations governing data privacy, security and ethical deployment of AI and ML-based technologies. Key legal documents such as the GDPR Regulation (EU) 2016/679 (GDPR), Regulation (EU) 2023/2854 (Data Act), Regulation (EU) Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 100 2024/1689 (AI Act) and international ICT standards were scrutinized to understand their implications for digital twin technology. This legal analysis was integrated with the ethical analysis to provide a comprehensive understanding of the regulatory and ethical landscape in which digital twins operate. 2.2.4. Case Study Approach The UP2030 Project in Istanbul served as the primary case study for this research. A case study approach was chosen to provide an in-depth understanding of the implementation and ethical challenges of digital twin technology in a specific urban context. The case study involved a detailed analysis of project documentation. This approach facilitated a comprehensive examination of the practical and ethical dimensions of digital twin technology in urban planning. 2.2.5. Methodological limitations and mitigations While this study employs a multifaceted methodology, including a literature review, ethical analysis, legal analysis, and a detailed case study, there are inherent limitations to these approaches. One limitation is the reliance on existing project documentation and publicly available data, which may not capture all nuances and real-time developments of the UP2030 Project. To address this, efforts were made to cross-reference multiple sources of data to ensure a more comprehensive understanding. Another limitation is the potential bias in literature and legal documents, which might reflect prevailing viewpoints rather than a balanced perspective. This was addressed by incorporating a diverse range of sources and critically evaluating contrasting viewpoints to ensure a thorough analysis. Additionally, while the case study approach offers detailed contextual insights, its findings may not be universally applicable. This limitation is acknowledged, and recommendations for further research in different urban contexts are provided to enhance the generalizability of the study. 3. RESULTS 3.1. Understanding the digital twin technology specifications Just a few years ago, creating digital twins of dynamic, possibly even living systems, would have been considered science fiction and deemed impossible from a scientific standpoint. (Helbing & Argota Sánchez-Vaquerizo, 2022) However, many believe that this scenario has recently changed and will continue to evolve, especially with the enormous amounts of data generated by the Internet of Things (IoT), transmitted via light (LiFi) or other low-latency communication systems, processed by quantum computers, and learned by powerful AI systems. (Helbing & Argota Sánchez-Vaquerizo, 2022) To be precise, a digital twin consists of three crucial parts: 1. physical products 2. virtual products 3. the connections tying them A digital twin fully describes a potential or physically manufactured product from the micro-atomic to the macrogeometrical level. Digital twin technology deeply integrates hardware, software, and IoT technologies to enrich and improve virtual entities. (Deng, et al., 2021) 3.2. Understanding the digital twin technology specification in Istanbul The AI-powered toolset for the decarbonization of buildings and transport in Istanbul represents an advanced application of digital twin technology, designed to address the city's complex challenges related to energy consumption, climate resilience, and urban mobility. Digital twin technology serves as a sophisticated virtual representation of physical assets, such as buildings and transportation systems, continuously updated with real-time data collected from these environments. (Deng, et al., 2021) By integrating both physical and digital domains, this toolset not only aims to contribute to Istanbul's goal Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 101 of achieving carbon neutrality by 2050 but also aspires to set a global standard for sustainable urban management. (UP2030 Project, 2023) AI is integral to the functionality of digital twins, providing the analytical power necessary to translate raw data into actionable insights. While the digital twin offers a dynamic virtual model of the physical world, it is through AI that this model gains predictive and prescriptive capabilities. (Helbing & Argota Sánchez-Vaquerizo, 2022) AI algorithms, which process the continuous stream of data from IoT sensors and other inputs, are essential in modelling complex urban scenarios, forecasting energy needs, and optimizing infrastructure deployment (Lam & Wernholm, 2023). In the context of Istanbul's UP2030 project, AI-driven models such as Recurrent Neural Networks (RNNs), Graph Neural Networks (GNNs), and Multi-layer Perceptrons (MLPs) are employed to accurately predict energy consumption, estimate renewable energy generation, and optimize the placement of critical infrastructure like photovoltaic (PV) systems and e-scooter charging stations. This symbiotic relationship between digital twin technology and AI enables a more responsive and adaptive approach to urban planning, where data-driven decisions can be made in real-time (UP2030 Project, 2024) Stepping on the general methodology of describing the different parts of the digital twin technology solution such as: 1. physical products, 2. virtual products and 3. the connections tying them, the AIpowered toolset for the decarbonization of buildings and transport in Istanbul will be presented in the same manner. This approach will help in understanding the technology specifications, which is crucial for analyzing the ethical implications. (Deng, et al., 2021) To understand the operation of this integrated technology, it is crucial to explore the three main components of the digital twin: physical products, virtual products, and the connections that unite them. (Deng, et al., 2021) This framework not only elucidates the technical specifications but also provides a foundation for analyzing the ethical implications of deploying such technology in a vibrant and diverse urban context like Istanbul. 3.2.1. Physical product At the heart of this digital twin technology are the physical products, which in the case of the AI-powered toolset for the decarbonization of buildings and transport includes real-world buildings, PV and urban infrastructure such as e-scooter charging stations. These physical components are equipped with IoT sensors that collect real-time data on various parameters, such as energy consumption, environmental conditions, and mobility patterns. The physical products also encompass energy consumers like e-bikes and e-scooters, along with their charging stations, integrated into urban furniture to provide functionalities such as PV-supported charging and socializing areas. 3.2.2. Virtual product The virtual product is the digital twin itself—an intricate 3D model that represents the physical urban environment in detail. (UP2030 Project, 2024) This virtual environment is where AI algorithms operate, running simulations and making predictions based on the data collected from the physical products. (Lam & Wernholm, 2023) Tools such as Energy Plus, which models building energy consumption, and various AI-driven models like RNNs and GNNs, are utilized to simulate scenarios and optimize outcomes. For instance, the AI can forecast future energy demands based on predicted weather patterns or assess the potential for energy generation from PV systems under different conditions. This virtual component is essential for urbanscale modelling, enabling planners to visualize the impact of various interventions and make informed decisions that align with both sustainability goals and the well-being of residents. (UP2030 Project, 2024) 3.2.3. Connections tying them The connections between the physical and virtual products are facilitated by robust data integration platforms and communication networks. These connections are the lifeblood of the digital twin, allowing Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 102 for seamless data flow from the physical world to its digital counterpart and back. (UP2030 Project, 2024) IoT communication protocols ensure that data from sensors is efficiently transferred to central data hubs, where AI algorithms process the information to update the digital twin in real-time. Cloud computing platforms play a vital role in storing and managing this vast amount of data, ensuring that the digital twin remains an accurate reflection of its physical counterpart. AI-driven analysis of this data enables continuous optimization of urban systems, such as adjusting energy distribution in response to real-time demand or refining mobility patterns to reduce congestion. By enabling such dynamic interactions between the physical and virtual domains, the digital twin and AI together create a feedback loop that enhances urban resilience and adaptability. (UP2030 Project, 2024) 3.2.4. Key Features The AI-powered toolset for the decarbonisation of buildings and transport in Istanbul comprises four main modules, each designed to address specific aspects of urban decarbonisation. • Module 1 supports the calculation of climate change's impact on building energy use and occupant comfort for the years 2020 and 2050. It also explores different building energy retrofit options using deep learning approaches. • Module 2 automates the calculation of electric energy generated by rooftop solar panels, enabling efficient and accurate energy forecasting. • Module 3 helps users determine the optimal locations of PV-integrated e-scooter charging stations using existing travel data, enhancing the infrastructure for electric mobility. • Module 4 conducts feasibility analyses for the integration of real-time control of micro-e-mobility (e-bike and/or e-scooter) and PV electricity generation co-deployment, ensuring effective and sustainable energy management in urban settings. These modules are interconnected through AI, which processes data from each component to ensure that the entire system functions cohesively and efficiently. (UP2030 Project, 2024) 3.2.5. Data requirements The toolset's functionality is supported by comprehensive data collection and management. • For Module 1 “Building Energy Use and Retrofit Options”, the required data includes geometric data such as building footprints, heights, and zone divisions; semantic data covering thermal transmittance of walls, roofs, and windows, window-to-wall ratios, number of occupants, boiler efficiency, infiltration rates, and more; and climate data involving dry bulb temperature, heating degree days, cooling degree days, and global horizontal radiation. (UP2030 Project, 2024) • For Module 2 “PV Energy Generation”, the necessary data encompasses panel-related data including panel efficiency, area, age, tilt, and azimuth; shading data with annual shading information; and climate data such as average dry bulb temperature, heating degree days, cooling degree days, and global horizontal radiation. (UP2030 Project, 2024) • Module 3 “E-Mobility Charging Stations” requires building-related data including coordinates and electricity consumption of buildings; PV-related data such as electricity production from PV systems; and mobility-related data including the starting and ending points of e-scooter journeys. (UP2030 Project, 2024) • For Module 4 “Real-Time Control of Micro-E-Mobility and PV Electricity Generation CoDeployment”, the data requirements include several categories. Mobility-related data encompasses real-time location information of e-bikes and e-scooters, battery levels, and usage patterns. Energyrelated data involves real-time energy production from PV systems, energy consumption rates of ebikes and e-scooters, and grid energy availability. Environmental data includes real-time weather conditions that affect PV energy generation. Operational data consists of control algorithms for coordinating energy distribution between PV systems and e-mobility charging stations, as well as Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 103 real-time feedback on system performance and user demand. These data inputs are essential for optimizing the integration and real-time management of micro-e-mobility and PV electricity generation. (UP2030 Project, 2024) 3.2.6. Outputs of the toolset The toolset provides various outputs to aid decision-making for carbon neutrality. • Module 1 calculates heating and cooling energy consumption, electrical energy consumption for lighting and equipment, and indoor overheating degrees, considering current conditions, retrofit scenarios, and climate change impacts for the years 2020 and 2050. • Module 2 determines the potential energy yield from installed PV systems on building rooftops, offering hourly, monthly, or annual resolutions. • Module 3 offers strategic placement of PV-powered e-mobility charging stations based on renewable energy availability and geographical distribution. • Module 4 provides feasibility analyses for the integration of real-time control of micro-e-mobility and PV electricity generation, outputting optimized control strategies for energy distribution and real-time performance metrics. (UP2030 Project, 2023) Fig. 1. Technical Specifications of the AI-Powered Toolset 3.2.7. Implementation and impact Currently, there is no existing installation of this solution in Istanbul. However, a pilot area has been selected with the Istanbul Metropolitan Municipality (IMM) to design and install PV-integrated urban furniture that supports various functions. The solution's Technology Readiness Level (TRL) is currently at level 3, indicating that it is in the early stages of development and testing. (UP2030 Project, 2023) OutputsData RequirementsFunctionModule •Energy consumption calculations •Retrofit scenario assessments of climate change on energy use •Geometric Data: building footprints, heights, zone division •Semantic Data: thermal properties, occupancy rates •Climate Data: temperature, heating/cooling degree days, radiation Energy Consumption and Retrofit Options Module 1 PV energy yield predictions (hourly, monthly, annually) •Panel-Related Data: efficiency, area, age, tilt, azimuth •Climate Data: temperature, radiation PV Energy Generation Calculation Module 2 Strategic placement of charging stations based on energy availability and geographical distribution •Building Data: coordinates, electricity consumption •PV Data: energy production •Mobility Data: start/end points of e-scooter journeys Optimal Placement of EScooter Charging Stations Module 3 Feasibility and optimization of integrated energy and mobility systems •Real-Time Mobility Data: location, battery levels, usage patterns •Energy Data: real-time PV production and consumption •Environmental Data: weather conditions Real-Time Control of MicroE-Mobility and PV Electricity Integration Module 4 Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 104 As the project progresses, AI will continue to play a central role in refining the digital twin, ensuring that it evolves into a robust and responsive tool capable of supporting Istanbul's ambitious sustainability goals. 3.3. Key ethical considerations in digital twin technology through the lens of Istanbul’s twin technology The AI-powered toolset for the decarbonisation of buildings and transport in Istanbul represents a sophisticated application of digital twin technology. This toolset provides comprehensive urban-scale modelling, simulation, and machine learning-based calculations to address the city's pressing energy consumption challenges, climate resilience, and mobility. By integrating physical products, virtual products, and their connections, this technology enables the collection and analysis of extensive data to optimise energy usage, predict renewable energy generation, and enhance urban mobility. (Deng, et al., 2021) However, while the benefits of this toolset are significant, its deployment raises key ethical considerations. These considerations are particularly critical given the specific technical specifications of the toolset, which involve detailed data collection from IoT sensors, advanced machine learning models, and robust data integration platforms. The primary ethical dimensions that arise include data privacy and security, potential bias and discrimination, and the implications for employment and societal well-being. (Deng, et al., 2021) 3.3.1. Data privacy and security ethical considerations The AI-powered toolset for the decarbonisation of buildings and transport in Istanbul involves extensive data collection from IoT sensors installed on buildings, PV systems, and mobility patterns. This includes geometric and semantic data of buildings, weather information, and utilization data of e-bikes and escooters. (UP2030 Project, 2024) Given the sensitive nature of this data, ensuring its privacy and security is paramount. The primary challenge lies in the sensitive data collection itself. The toolset gathers detailed information, including building footprints, heights, thermal properties, and occupancy rates, which are inherently sensitive and personal. Additionally, mobility data such as the start and end locations of e-bike and escooter journeys, times of use, and energy demand patterns contain personal information that requires stringent protection measures. (UP2030 Project, 2023) Each module within the toolset presents unique challenges concerning data privacy and security due to the type and sensitivity of the data collected. • Module 1 gathers geometric data (building footprints, heights, zone divisions) and semantic data (thermal properties, occupancy rates). (UP2030 Project, 2024) The detailed nature of this information makes it inherently sensitive, as it can reveal personal and private details about building occupants. • Module 2 involves the collection of panel-related data (efficiency, area, age, tilt, azimuth), shading data, and climate data. (UP2030 Project, 2024) While less personal, this data requires protection due to its volume and the potential insights it can offer into energy consumption patterns and locations. • Module 3 collects mobility-related data, such as real-time locations of e-scooters and e-bikes, battery levels, and usage patterns. (UP2030 Project, 2024) This data is highly sensitive as it can trace the movements and habits of individuals, necessitating stringent protection measures. • Module 4 gathers real-time operational data for managing the integration of micro-e-mobility and PV electricity generation, including real-time location data, energy production, consumption rates, and control algorithms. (UP2030 Project, 2024) The real-time nature and operational sensitivity of this data make its security paramount. Given the sensitive nature of the data the tool collects, which involves extensive data collection from IoT sensors, falls under the category of high-risk AI systems according to the AI Act. (The European Parliament and the Council of the European Union, 2024) Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 111 transparency in AI systems presents challenges. Many AI algorithms are proprietary, with their internal workings closely guarded by the developing companies. Even when algorithms are open for inspection, their complexity can render them incomprehensible to those without specialized knowledge. (Deng, et al., 2021) Moreover, AI-driven planning tools can be prohibitively expensive and require specialized expertise to operate, potentially centralizing decision-making power among those with access to these resources. This centralization risks creating a less inclusive planning process, marginalizing community members and smaller stakeholders from decisions that directly impact them. (Deng, et al., 2021) 4.5.2. Transparency and Accountability Ethical Considerations To address these issues, urban planners and planning organizations will need a multifaceted approach. One critical step is the development and enforcement of ethical guidelines for AI use in urban planning. These guidelines should emphasize accountability, requiring clear lines of responsibility for decisions made with the assistance of AI. (Deng, et al., 2021) They should also promote transparency, in terms of both the algorithms used and the data they operate on. Making these aspects of AI systems accessible and understandable to non-experts can help build trust and facilitate more inclusive decision-making processes. (Sanchez, et al., 2024) Addressing these concerns requires a concerted effort from all stakeholders involved in the urban planning process. By developing ethical guidelines, promoting transparency, ensuring accountability, and engaging with the community, urban planners can use new AI methods while mitigating its potential ethical pitfalls. (OECD, 2023) However, it is important to point out that this is true for all aspects of a planner’s responsibilities and is not limited to AI-based approaches. (Deng, et al., 2021) The use of AI algorithms in urban planning may be opaque, making it challenging for residents to understand how certain recommendations or decisions are made. A lack of transparency can erode trust in the planning process, particularly among marginalized communities that may already have a history of being excluded from decision-making. To address this concern, planners and organizations should prioritize transparency in their use of AI. Transparency should be emphasized throughout the planning process. (OECD, 2023)This means not only providing clear explanations of how algorithms work but also involving the community in algorithm development and decision-making. (Sanchez, et al., 2024) 4.6. Impact on general societal well-being 4.6.1. Ethical considerations regarding societal well-being The combination of digital twin technology and artificial intelligence (AI) introduces significant ethical considerations, particularly concerning the balance between automated decision-making and human judgment. Digital twins serve as detailed virtual representations of physical entities—such as buildings, infrastructure, or even entire cities. When AI is applied to these digital twins, it enhances their functionality by analyzing vast amounts of data to simulate scenarios, predict outcomes, and suggest optimized solutions. However, this efficiency poses the risk of these AI-enhanced digital twins being granted greater authority than human decision-makers. ( (Helbing & Argota Sánchez-Vaquerizo, 2022) One key concern is that the data-driven insights produced by AI within a digital twin might not fully capture the complexity of human intentions and social dynamics. This can result in decisions that, while technically sound, fail to align with the nuanced needs and values of individuals or communities. For instance, if there is a discrepancy between a person’s true intentions and the data represented by their digital twin, the system may prioritize the digital twin’s data, potentially overlooking the individual’s actual perspective. (Deng, et al., 2021) A societal digital twin could also be used to gauge how much pressure can be applied to individuals without provoking a revolution or to devise strategies to subdue majorities, undermine individual will, and impose policies misaligned with public sentiment. Such uses of mass surveillance would be highly invasive and undermine human rights. (Helbing & Argota Sánchez-Vaquerizo, 2022) Even in less extreme scenarios, the increasing reliance on AI and digital twins for managing societal functions could result in the inappropriate application of methodologies better suited for corporate or logistical operations. These approaches may oversimplify complex human factors, treating critical Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 112 aspects such as dignity and freedom as secondary concerns or "noise." This could erode the very strengths that allow societies to thrive—such as their capacity for adaptation, self-organization, and innovation. If left unchecked, these developments could lead to inhumane social structures, with the potential for a form of "technological totalitarianism," where control is prioritized over human welfare (Helbing & Argota Sánchez-Vaquerizo, 2022) 4.6.2. Ethical considerations regarding societal well-being mitigation To address these ethical challenges, it is recommended that rather than substituting individual decisions with automated machine determinations, digital assistance should be utilized to support decision-making processes. This approach would provide individuals and systems with beneficial opportunities. Instead of seeking a single optimal solution for a given objective (e.g., selecting the shortest route), decisionsupport systems should offer multiple high-performance solutions, each characterized by diverse qualities pertinent to various goals. (Xia, et al., 2023) Societal governance should prioritize human welfare. As individuals become increasingly integrated into socio-technical systems, it is evident that a purely technology-driven approach is insufficient. Social innovation is paramount to unlocking the full benefits of the digital age for all. A platform supporting genuine informational self-determination is urgently needed. Additionally, the traditional "war room" approach must be supplanted by a "peace room" approach, necessitating an interdisciplinary, ethical, and multi-perspective methodology. (Deng, et al., 2021) This new multi-stakeholder approach is essential for achieving better insights and fostering participatory resilience. By ensuring that digital twins serve as tools to aid human decision-making rather than replace it, and by fostering environments where diverse perspectives are considered and respected, the potential for technological totalitarianism can be mitigated. (Deng, et al., 2021) This approach emphasizes the importance of transparency, accountability, and inclusivity in the deployment of digital twin technologies, ultimately supporting a more humane and equitable society. (Helbing & Argota SánchezVaquerizo, 2022) 5. CONCLUSIONS This paper explores the transformative potential of ethically integrating digital twin technology into urban planning, using Istanbul's UP2030 Project as a case study. The AI-powered toolset for decarbonizing buildings and transport addresses critical challenges in energy consumption, climate resilience, and mobility. To fully leverage the benefits of digital twin technology, it is essential to address ethical concerns such as data privacy, bias prevention, and broader societal impacts. Ensuring data privacy and security is crucial due to extensive data collection from IoT sensors and AI systems. Robust security measures, such as encryption and regular audits, are essential. Additionally, addressing bias and discrimination is critical, as machine learning models can inherit biases from training data. This requires diverse data collection, bias mitigation techniques, and stakeholder engagement. The reliance on automated decision-making systems raises concerns about diminishing human oversight. Ensuring transparency, accountability, and maintaining human involvement in decisionmaking processes is vital. By addressing these ethical considerations, Istanbul aims to achieve carbon neutrality by 2050 and set a global benchmark for ethical digital twin implementation. This approach demonstrates that combining technological innovation with ethical governance can create resilient and equitable urban environments, serving as a model for sustainable smart city development worldwide. Looking ahead, the future of ICT in urban transformation, driven by AI, ML, IoT, and data analytics, holds immense promise. However, the ethical integration of these technologies is of absolute importance. Ensuring that advancements are inclusive, equitable, and transparent will be essential to embrace their full potential for urban improvement. Lessons from Istanbul's implementation can guide future projects, ensuring technological progress aligns with ethical responsibility. Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 113 ACKNOWLEDGEMENTS This paper was supported by the Horizon Europe Programme as part of the UP2030 Project. The financial assistance provided by the European Commission is gratefully acknowledged. Appreciation is also extended to the consortium colleagues whose expertise and insights significantly contributed to the study on the ethical dimensions of digital twin technology in urban planning. REFERENCES Lauradoux, C., Curelariu, T. & Lodie, A., 2023. Re-identification attacks and data protection law, s.l.: s.n. Pierce, P., Ricciardi, F. & Zardini,, A., 2017. Smart Cities as Organizational Fields: A Framework for Mapping Sustainability-Enabling Configurations. Sustainability, 9(9), pp. 1-21. Albino, V., Berardi, U. & Dangelico, R. M., 2015. Smart Cities: Definitions, Dimensions, Performance, and Initiatives. Journal of Urban Technology, 1(22), pp. 3-21. Batty, M., 2018. Digital Twins. Environment and Planning B: Urban Analytics and City Science, Volume 45, pp. 817-820. Berardi, A. V., Berardi, U. & Dangelico, R. M., 2021. Smart Cities: Dimensions, Performance, and Initiatives. Journal of Urban Technology. Brauneis, R. & Goodman, P. E., 2018. City, Algorithmic Transparency for the Smart. Yale Journal of Law and Technology, Issue 20, p. 103. Braun, T., Fung, C. B., Iqbal, F. & Shah, B., 2018. Security and privacy challenges in smart cities. Sustainable Cities and Society, pp. 499-507. Christofi, A. & Valcke, P., 2024. Data protection in smart cities. Balancing fundamental rights with public interests in a complex and multi-actor environment.. [Online] Available at: https://lirias.kuleuven.be/4154792&lang=en Dell Technologies, 2024. Urban Digital Twins: AI Comes to City Planning, s.l.: Forbes. Deng, T., Zhang, K. & Shen, J., 2021. A Systematic Review of a Digital Twin City: A New Pattern of Urban Governance toward Smart Cities. Journal of Management Science and Engineering, 6(2), pp. 125-134. Edwards, L., 2015. Privacy, Security and Data Protection in Smart Cities: a Critical EU Law Perspective, s.l.: CREATe. El-Agamy, F. R., Sayed, A. H., AL Akhatatneh, A. M. & Elhosseini, M., 2024. Comprehensive analysis of digital twins in smart cities: a 4200-paper bibliometric study. Artificial Intelligence Review. European Commission, n.d. The European Green Deal. [Online] Available at: https://ec.europa.eu/stories/european-green-deal/ European Parliament and Council of the European Union, 2016. [Online] Available at: https://eur-lex.europa.eu/eli/reg/2016/679/oj Gouvea, R., Kapelianis, D. & Kassicieh, S., 2017. Assessing the nexus of sustainability and information & communications technology. Technological Forecasting and Social Change, July.pp. 1-10. Hacker, P. & Neyer, J., 2023. Substantively smart cities: Participation, fundamental rights and temporality.. Internet Policy Review, Volume 1, pp. 1-30. Helbing, D. & Argota Sánchez-Vaquerizo, J., 2022. Digital Twins: Potentials, Ethical Issues, and Limitations (revised, final version). IPA Istanbul, 2022. Climate Vision of Istanbul. [Online] Available at: https://ipa.istanbul/en/wp-content/uploads/2021/11/Climate-Vision-of-Istanbul.pdf Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 114 Istanbul Development Agency, 2015. Istanbul Factsheet, s.l.: s.n. Kitchin, R., 2016. The ethics of smart cities and urban science. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 378(2083). Lam, Y. W. & Wernholm, E., 2023. Algorithmic Bias in City Scale Digital Twin Data Management Processes: A qualitative exploration of the pragmatic, ethical, and moral prioritization surrounding algorithmic bias in city scale digital twin projects, s.l.: Lund School of Economics and Management. Moggi, S. & Dameri, R. P., 2021. Circular Business Model Evolution: Stakeholder Matters for a SelfSufficient Ecosystem. Business Strategy and the Environment, 6(30). Mohammadi, N. & Taylor, J. E., 2017. s.l., IEEE Symposium Series on Computational Intelligence. MOSAIC Project, n.d. Cities Apply to Mission Cities Call. [Online] Available at: https://shorturl.at/eBlIl OECD, 2023. Smart City Data Governance: Challenges and the Way Forward, OECD: OECD Publishing. Oti-Sarpong, K., Bastidas, V., Nochta, T. & Wan, L., 2022. A Social Construction of Technology View for Understanding the Delivery of City-Scale Digital Twins. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, X-4(W3-2022), pp. 205-212. Pierce, P., Ricciardi, F. & Zardini, A., 2017. Smart Cities as Organizational Fields: A Framework for Mapping Sustainability-Enabling Configurations. Sustainability, 9(9), pp. 1-21. PwC, n.d. How digital twins can make smart cities better: Real-time simulations can create a bridge between physical and virtual worlds, s.l.: s.n. Sanchez, T. W., Brenman, M. & Ye, X., 2024. The Ethical Concerns of Artificial Intelligence in Urban Planning. Journal of the American Planning Association, pp. 1-3. Science Europe, 2024. Practical Guide to Supporting Diversity in Research Environments, s.l.: s.n. Smallwood, F. R., 2019. Information Governance: Concepts, Strategies and Best Practices. s.l.:John Wiley & Sons. Tetteh, E. K., Geng, E. H. & Hufman, M. D., 2023. Developing ethical standards for dissemination and implementation research: a roadmap for consensus and guidance, s.l.: s.n. The European Parliament and the Council of the European Union, 2019. Open Data Directive, s.l.: Official Journal of the European Union. The European Parliament and the Council of the European Union, 2023. Data Act, s.l.: s.n. The European Parliament and the Council of the European Union, 2024. AI Act, s.l.: Official Journal of the European Union. The European Parliament and the Council, 2016. NIS Directive, s.l.: Official Journal of the European Union. The Tech Robot, 2024. Data Security and Privacy in the Era of Digital Twins, s.l.: s.n. Tzachor, A. et al., 2022. Potential and Limitations of Digital Twins to Achieve the Sustainable Development Goals. Nature Sustainability, 5(10), pp. 822-829. United Nations Department of Economic and Social Affairs, 2018. 68% of the world population projected to live in urban areas by 2050, says UN. Department of Economic and Social Affairs. [Online] Available at: https://www.un.org/development/desa/en/news/population/2018-revision-of-worldurbanization-prospect.html United Nations Environment Programme, n.d. Cities and Climate Change. [Online] Available at: https://www.unep.org/explore-topics/resource-efficiency/what-we-do/cities-and-climatechange.html Language, Individual & Society ISSN 1314-7250, Volume 18, 2024 Journal of International Scientific Publications www.scientific-publications.net Page 115 United Nations, 2015. Paris Agreement. [Online] Available at: https://unfccc.int/sites/default/files/english_paris_agreement.pdf UP2030 Project, 2023. Deliverable 3.1: Transformative pathways roadmaps: strategic integration of solutions and interoperability, s.l.: s.n. UP2030 Project, 2024. Deliverable 3.4: Bundle of Digital Twin tools for net-zero decision-making, s.l.: s.n. UP2030 Project, n.d. Pilot cities: Istanbul city section. [Online] Available at: https://up2030-he.eu/unique-city/istanbul/ Xia, L. S., D, T. & Rezaei, R., 2023. thorough examination of smart city applications: Exploring challenges and solutions throughout the life cycle with emphasis on safeguarding citizen privacy. Sustainable Cities and Society, Issue 98. Yigitcanlar, T. et al., 2018. Understanding ‘Smart Cities’: Intertwining Development Drivers with Desired Outcomes in a Multidimensional Framework. Cities, Issue 81, pp. 145-160. Yigitcanlar, T., Kamruzzaman, M. & Buys, L., 2018. Understanding ‘Smart Cities’: Intertwining Development Drivers with Desired Outcomes in a Multidimensional Framework. Cities, Issue 81, pp. 145-160. Zou, J. & Schiebinger, L., 2018. Design AI So That It's Fair. Nature, Issue 559, pp. 324-326.