Artificial Intelligence and Sustainability

The sustainability of ecosystems and human well-being are both directly impacted by air quality, which is a crucial component of environmental health. Due to its negative effects on societal advancement, the environment, and public health, the deteriorating air quality around the world has sparked serious worries. It is essential to have accurate and fast air quality forecasts in order to address this urgent problem since it can offer helpful information for making wise decisions, carrying out mitigation strategies successfully, and protecting sensitive communities. Using AI and the linear regression technique, we will investigate many facets of air quality forecasting in this study. To guarantee the dependability and quality of the dataset, we will examine data gathering, preprocessing, and feature engineering strategies. We will also go over selection criteria and compare AI algorithms, emphasizing the benefits of Linear Regression over alternative approaches. We will also give a thorough explanation of the Linear Regression algorithm, complete with a pseudocode that fits the Vinnytsia air quality dataset’s particular context.

Although the world is eager to use these new AI systems and trends to take advantage of their potential benefits, unclear implications remain evident now and in the long run. Using the dimensions of sustainability in structuring this analysis, the scope of the research revolved around exploring the strategies that may be used in developing and evaluating AI products (Z. Zhang, D. Lyu, P. Arcaini, L. Ma, I. Hasuo and J. Zhao, IEEE Trans Softw Eng 1–17, 2022). According to the research findings, the fundamental goal of AI’s design and ethical development is increasing acceptance and trust of emerging technologies. This, therefore, establishes the need to create sustainable, reliable, secure AI products which remains critical as AI continues to evolve globally. The implications of this study will go a long way in ensuring that product developers focus on the sustainability of AI products and software, achievable by observing the prospects of sustainability in the product development lifecycle.

In the contemporary realm of agricultural research, the integration of artificial intelligence (AI) with traditional practices is progressively emerging as a focal theme. Nonetheless, the scholarly landscape has demonstrated a notable dearth in examining the intricate relationship between AI and the tripartite principles of sustainable agriculture: economic viability, environmental stewardship, and social responsibility. Utilizing the PRISMA methodology, this systematic review endeavors to bridge this gap, offering an exhaustive examination of the aforementioned relationship. The analysis revealed that while many studies have explored individualized AI applications, few have situated these advancements within a holistic sustainability framework. The findings underscore AI’s potential, when aptly channeled, to address the multifaceted challenges inherent in sustainable agriculture. In essence, this review highlights AI’s transformative capacity to redefine agricultural frameworks, emphasizing its central role in guiding agriculture towards a future deeply rooted in sustainability principles.

With unabated global warming and climate change, the concept of Green Artificial Intelligence has emerged in which companies in the information and communication sector are increasingly embracing sustainable methods for designing, developing, and applying AI software. This paper examines current research on Green AI coding practices, design principles, use cases, and applications, as well as policy, ethical, and regulatory issues. The overarching premise is that Green Artificial Intelligence is a promising paradigm for mitigating climate change through collaboration. However, it requires a streamlined regulatory framework combining AI policy, industry standards and best practices, and legal frameworks for addressing emerging issues as AI technology evolves.

Amidst escalating concerns about marine environmental pollution, this systematic review delves into the role and potential of artificial intelligence (AI) in monitoring and managing marine ecosystems. Through the meticulous adoption of the PRISMA methodology, various databases were scrutinized, resulting in a curated list of seminal studies that encompass AI’s diverse techniques, such as machine learning and deep learning, in marine monitoring. The findings elucidate that AI techniques offer transformative advancements in real-time data analysis, predictive modeling, and anomaly detection, making them indispensable for contemporary marine conservation efforts. However, despite their potential, AI systems also bear limitations that need thoughtful consideration. This review not only delineates AI’s current applications in marine environmental monitoring but also extrapolates its future trajectories, providing critical insights for researchers, policymakers, and AI developers. By bridging the gap between AI’s technological prowess and marine ecosystem conservation needs, the research underlines the urgency of integrating advanced computational methodologies to safeguard oceans for future generations.

Marine debris poses a significant threat to the marine ecosystem, and its detection and removal are crucial for environmental sustainability. This study presents a comprehensive study on the application of computer vision techniques for marine debris detection, with a specific focus on the task of trash detection. The study utilizes a diverse and carefully curated dataset named “YOLOv5 Marine Debris” (or any suitable name for the YOLOv5 dataset), obtained from the Ultralytics open-source research repository for YOLOv5 models. Through the utilization of computer vision techniques, this study strives to contribute to the development of sustainable solutions for marine debris detection. The YOLOv5 model’s state-of-the-art capabilities, complemented by the diverse and challenging “YOLOv5 Marine Debris” dataset, enable accurate and reliable trash detection across various underwater environments. By pushing the boundaries of object detection, image processing, and machine learning, this research envisions a future where autonomous robot platforms efficiently monitor and remove marine debris, safeguarding the marine ecosystem.

In the digital innovation landscape, the integration of sustainability and mobile app development emerges as a critical convergence, exemplifying the blend of technological advancements with environmental awareness. Guided by the PRISMA framework, this systematic review meticulously navigated a spectrum of academic sources to identify prevailing insights, emerging trends, and discernible gaps in the realm of green mobile app development. Employing methodological rigor, diverse databases were consulted, and salient studies were selected. Subsequent data extraction and analysis processes revealed the transformative potential of digitalization with an emphasis on Sustainable Development Goals. Notably, emerging paradigms such as the Green Internet of Things (G-IoT), 6G networks, and inclusive citizen science highlighted the expanded role of mobile apps as instrumental vectors in promoting environmentally conscious behaviors and sustainable strategies. Nonetheless, challenges abound, with developers navigating the terrain of balancing innovation with ethical and environmental obligations. In essence, the review accentuates the significance of interweaving sustainability in mobile app development, championing a vision for a future marked by technological responsibility and eco-centric innovation.

The increasing levels of CO₂ emissions have become a significant concern worldwide. Accurate prediction of CO₂ emission levels plays a crucial role in implementing sustainable practices and driving policy decisions. This study aims to develop a machine learning model for predicting CO₂ emission footprints using various socio-economic and environmental factors. The model will facilitate effective planning and decision-making in reducing global carbon emissions. The main objective of this study is to develop a machine learning model that accurately predicts CO₂ emissions from fossil fuels and identifies the most important factors that contribute to these emissions. The results of this study could provide insights into the most effective strategies for reducing CO₂ emissions and mitigating the impacts of climate change.

The increased energy consumption around the globe has consequently led to a high amount of energy waste. While people need energy in various forms, such as electricity, fossil fuels, and natural gas, energy wastage and abnormal consumption are alarming for their day-to-day activities. Building and cloud computing are among the leading energy consumption and wastage fields following the increased number of residential and commercial buildings and the recent growth in cloud computing. Energy waste and abnormal consumption lead to increased gas emissions, which threaten the sustainability of the global climate. Using artificial intelligence as a sustainability tool, this study’s author researched anomaly detection of energy consumption in cloud computing and buildings. Using qualitative research methodologies, the researcher established that artificial intelligence methods and techniques, such as machine learning, are more effective and efficient in detecting abnormalities in data consumption. Though various researchers have established frameworks to solve and militate abnormalities in energy consumption, they face serious challenges such as increased cost, lack of efficiency for large data, and lack of skilled detectors. Intending to overcome these challenges, the researcher developed a new framework that employs machine and deep learning technologies to determine anomalies in cloud and building energy consumption.