Translate this page into:
Artificial intelligence for sustainable consumption and production in green chemistry
*Corresponding author: Sowmyalakshmi Venkataraman, Department of Pharmaceutical Chemistry, Sri Ramachandra Faculty of Pharmacy, Sri Ramachandra Institute of Higher Education and Research (DU), Chennai, Tamil Nadu, India. sowmyamahesh30@gmail.com
-
Received: ,
Accepted: ,
How to cite this article: Raagavan Sarkgunan S. Chandrasekar V, Venkataraman S. Artificial intelligence for sustainable consumption and production in green chemistry. Sri Ramachandra J Health Sci. 2025;5:3-10. doi: 10.25259/SRJHS_25_2024
Abstract
Machine intelligence or artificial intelligence (AI) helps in the transformation of green chemistry by improving the sustainable consumption and production. The present study specifically focuses on the green chemistry aspects as how AI technologies are used to enhance the chemical processing, utilization of environment friendly solvents, and even monitor the processing through real time techniques. Enhancements achieved in other areas include yield of the desired products in various reactions, decrease in amounts of waste produced, and improvement in energy consumption. Nevertheless, there are still some unexplored problems which are still relevant now included data quality problems, limitations on model interpretability, and AI interface with the current systems. AI technologies also face other challenges that are attributed to the economic nature, which includes high initial costs and challenges of proving viability of the investments. In the same aspect, standards and ethical issues should be observed when it comes for applying AI in chemical processes. Even more studies also point out about research opportunities to advance such as refining the underpinning data fusion techniques, to design the interpretable AI models, and to expand the potential applications considering the new fields such as biodegradable materials and renewable energy systems. Therefore, overcoming interdisciplinary collaborations and ethical issues will be vital to the development of AI for green chemistry. If these difficulties are to be met and AI’s potential harnessed, the chemical industry can go a long way toward making real improvements to a range of environmental and economic metrics.
Keywords
Artificial intelligence
Chemical optimization
Eco-friendly solvents
Green chemistry
Real-time monitoring
Sustainable consumption
Sustainable production
INTRODUCTION
Over the years, research on the combination of artificial intelligence (AI) in different fields has shown tremendous improvements, especially in sustainable environmental practices and policies. Green chemistry which involves designing products and processes that would reduce the production of hazardous substances is the key fundamental aspect of sustainability.[1] The combination of AI and green chemistry will go a long way in changing the future of efficient sustainable consumption and production (SCP) by reducing the negative impacts on the environment.
AI comprises technologies such as machine intelligence, deep intelligence, and natural intelligence that could be used in handling other elaborate issues in chemical and industrial advancement.[2] These technologies help in data mining of Big Data, refining of chemical processes, and creation of upto-date solutions to the test of sustainability.
Scope and objectives of the study
Therefore, this chapter is proposed to understand the impact of AI in green chemistry specifically in promoting the idea of SCP. The specific objectives are to following objectives:
Provide an insight over the most important technologies in the field of AI and their use in green chemistry
Explain how AI update enhances chemical processes as well as facilitate sustainability
Explore at examples or case studies that has involved AI in green chemistry
Discussed potential future directions to pave a way to solve current issues in adoption of AI with green chemistry
Draw attention toward further research in the emerging field of this analysis and the future possibilities with this.
OVERVIEW OF AI
AI refers to the simulation of human intelligence processes by computer systems. These processes include learning, reasoning, problem-solving, and decision-making.[3] AI systems are designed to perform tasks that typically require human intelligence, such as understanding natural language, recognizing patterns, and making predictions. The fundamental goal of AI is to create machines that can operate autonomously and adapt to new situations based on their experiences.
Types of AI technologies
AI is a combination of several technologies [Figure 1] and methodologies, all of which supports various applications of AI. Of these, the machine learning (ML) is one of the most well-known components. ML is an applied science that deals with the design of the algorithm through which a computer has to learn to operate from the data.[4] Compared with other forms of programming such as a definite set of instructions given to the system which executes some specific task, the ML models improve from the data that they receive or the more data they handle, and the more accurate their models become. Such algorithms are particularly useful in classification, regression, and clustering as they are capable of identifying the complex relationships between the data and then making accurate predictions or clustering.[5]
- Types of artificial intelligence.
In the broader conceptual framework of ML, Deep Learning can be seen as a subset yet more feature-packed solution. Deep Learning uses artificial neural networks, which include many hidden layers; such networks are commonly called the deep neural networks while performing large-scale data modeling for complex and abstract patterns.[6] This capability had brought with it a lot of advancements in areas such as computer vision, where deep learning models can now identify and sort images to a human-like precision, automatic speech recognition and natural language processing (NLP).[7] Due to the depth and complexity of such networks, they are able to teach hierarchies of features from the raw data to come up with very developed models that are capable of performing complex tasks.[8]
Another important branch of AI is called NLP that concerns the capability for the computer to read, comprehend, and even produce natural language in the way that makes sense. NLP technologies can facilitate the corporate’s strategy for processing and analyzing a significant volume of natural language data, which includes language translation, sentiment analysis, and chatbot interaction. NLP is used to make AI systems more natural in their interactions with people, and thus useful when responding to requests, filtering comments or moderating content.
Thus, using the ML algorithm and supported by Deep Learning and NLP, we have the current basis of modern AI that catalyzes developments changing the vast number of industries and gives people new opportunities to solve previously unresolved problems.
Current trends and developments in AI
The field of AI is rapidly evolving, with several trends and developments shaping its future:
Advances in algorithms and models
The new AI algorithms, which include reinforcement learning and the generative adversarial networks make the AI systems more powerful in solving problems and creating realistic fake data.[9]
AI in industry and research
AI is being deployed in many fields today that include the healthcare, financial sector, and manufacturing among others. In the case of healthcare, AI has been applied in diagnostics and for individualized treatment, and in finance, it has been used for algorithm trading and fraud prevention.[10]
Ethical and societal implications
Researchers have raised some concerns over the future of AI and its use such as the following: The problem of bias, privacy, and employment in the future society. Meeting these challenges is important for the effective and safe use of AI solutions.[11]
GREEN CHEMISTRY AND SUSTAINABLE CONSUMPTION
Principles of green chemistry
Green Chemistry, also known as sustainable chemistry, focuses on designing chemical processes and products that reduce or eliminate the use and generation of hazardous substances. The 12 principles of green chemistry [Table 1] in a way help the researchers to develop more sustainable practices.[1]
| S. No | Principle | Description |
|---|---|---|
| 1. | Prevention | It is better to prevent waste than to treat or clean up waste after it has been created. |
| 2. | Atom economy | Design synthetic methods to maximize the incorporation of all materials used into the final product. |
| 3. | Less hazardous chemical syntheses | Design syntheses to use and generate substances that are non-toxic to human health and the environment. |
| 4. | Designing safer chemicals | Design chemical products to be effective while being safe for human health and the environment. |
| 5. | Safer solvents and reaction conditions | Avoid the use of solvents and reaction conditions that are hazardous. |
| 6. | Energy efficiency | Use energy-efficient processes to reduce energy consumption and associated environmental impacts. |
| 7. | Renewable feedstocks | Use renewable raw materials or feedstocks whenever technically and economically feasible. |
| 8. | Reduce derivatives | Minimize the use of auxiliary substances or reaction by-products. |
| 9. | Catalysis | Use catalysts that are selective and reduce the need for stoichiometric reagents. |
| 10. | Design for degradation | Design chemicals to break down into innocuous substances after use. |
| 11. | Real-time analysis for pollution prevention | Develop analytical techniques to monitor and control chemical processes in real-time. |
| 12. | Inherently safer chemistry for accident prevention | Design processes to minimize the potential for chemical accidents. |
SCP
SCP is about using resources in the least amount possible while at the same time not harming the environment or the populace’s health (The United Nations Environment Programme (UNEP), 2012) [Table 2].[12] This is in a bid to minimize the utilization of resources thereby regulates the ecological footprint in a sustainable manner.
| Indicator | Traditional chemistry practices | Green chemistry practices | References |
|---|---|---|---|
| Average energy consumption (MJ/kg) | 30.2 | 20.1 | [4] |
| Waste generation (kg/kg product) | 0.75 | 0.40 | [3] |
| Use of hazardous chemicals (%) | 60% | 25% | [1] |
| Percent of renewable feedstocks | 10% | 45% | [11] |
| Cost of pollution control (USD/ton) | 1200 | 800 | [12] |
USD: The United States dollar
INTEGRATION OF AI IN GREEN CHEMISTRY
AI applications in chemical research and development
Green chemistry is being enhanced through AI in a number of ways including improving chemical processes, decreasing wastes and improving the design of chemicals that are safer and environmental friendly[13] [Table 3]. Employment of ML, deep learning, and data analytics are found to be useful in these fields.
| Application Area | AI technology used | Key achievements | Type of study | References |
|---|---|---|---|---|
| Process optimization | Machine learning | Reduced reaction time and improved yields | Optimization of chemical reaction conditions | [5] |
| Predictive modeling | Deep learning | Accurate prediction of chemical properties | Prediction of polymerization outcomes | [10] |
| Waste reduction | Data analytics | Decreased waste generation | Optimization of waste management processes | [4] |
| Design of safer chemicals | Generative models | Design of less hazardous compounds | Design of new environmentally friendly solvents | [13] |
| Real-time monitoring | AI-powered sensors | Enhanced real-time process control | Real-time monitoring of chemical synthesis | [14] |
AI: Artificial intelligence
AI for sustainable production
AI also plays a significant role in promoting sustainable production practices by enhancing resource efficiency, optimizing energy use, and minimizing environmental impact. Key areas where AI contributes include:
Resource efficiency: AI techniques improve product utilization of inputs and energy hence cutting on resource utilization and spread.
Energy management: The use of AI models for forecasting energy use is useful for minimizing a company’s carbon emissions and increasing overall sustainability.
Environmental impact assessment: ML helps in the assessment of effects of chemicals and their functions, which facilitates in pinpointing environmentally friendly methods.
CASE STUDIES AND EXAMPLES
Case Study 1: AI-optimized chemical reactions
The improvements of AI technologies have helped to enhance the chemical reactions effect, hence improve the efficiency of the entire chemical process [Table 4].[14-16] An example that is worth mentioning is the application of the ML algorithms, for improving the yield and the selectivity of the chemical reactions. Figure 2 effectively illustrates the enhanced efficiency and sustainability achieved through AI-driven techniques. It represents the improvements achieved using AI-optimized methods over traditional methods in chemical processes. It shows four key parameters: Reaction yield, reaction time, waste generation, and energy consumption. The largest segment, represents about 45.5% reduction in waste generation, highlights the significant environmental benefits of AI optimization. Other notable improvements include a 37.5% reduction in reaction time, a 31.3% decrease in energy consumption, and a 21.4% increase in reaction yield.
| Parameter | Traditional Methods | AI-Optimized Methods | Improvement (%) | References |
|---|---|---|---|---|
| Reaction yield (%) | 70 | 85 | +21.4 | [15] |
| Reaction time (hours) | 8 | 5 | −37.5 | [14] |
| Waste generation (kg/kg product) | 0.55 | 0.30 | −45.5 | [16] |
| Energy consumption (MJ/kg) | 32 | 22 | −31.3 | [14] |
AI: Artificial intelligence
- Comparison of traditional versus artificial intelligence-optimized methods.
Case study 2: AI-driven green solvent design
AI has been used to design new green solvents that are environmentally friendly and effective. The application of
AI in this field has led to the development of solvents with improved performance and lower toxicity [Table 5].[17,18]
| Metric | Conventional solvents | AI-Designed solvents | Improvement (%) | References |
|---|---|---|---|---|
| Toxicity (mg/kg) | 150 | 90 | −40 | [15] |
| Boiling point (°C) | 180 | 165 | −8.3 | [17] |
| Solubility in water (%) | 60 | 80 | +33.3 | [15] |
| Cost (USD/L) | 5 | 3 | −40 | [14] |
AI: Artificial intelligence, USD: The United States dollar
Figure 3 compares the properties of conventional solvents and AI-designed solvents across four metrics: They include toxicity, boiling point, and solubility in water and cost. The chart employs two groups of bars one for normal solvents and the other for AI-designed solvents where results can be easily compared. The X-axis denotes the different metrics while the Y-axis shows the various values that corresponds to each of the metrics. The chart shows a notable reduction in toxicity, cost, and boiling point and marked increase in solubility for AI-designed solvents.
- Comparison of conventional versus artificial intelligence-designed solven. AI: Artificial intelligence.
Case study 3: AI for real-time monitoring in chemical manufacturing
Advanced sensors and automatic real-time monitoring systems have also been adopted in chemical industries to improve the control and optimization of the manufacturing processes [Table 6].[17-19] The use of AI not only has enhanced the stability of the process but also minimized the occurrence of-dangerous events.
| Metric | Conventional systems | AI-powered systems | Improvement (%) | References |
|---|---|---|---|---|
| Process stability (%) | 85 | 95 | +11.8 | [18] |
| Downtime (hours/year) | 120 | 45 | −62.5 | [14] |
| Incident rate | 12 | 3 | −75 | [19] |
| Maintenance costs (USD/year) | 50,000 | 25,000 | −50 | [17] |
AI: Artificial intelligence, USD: The United States dollar
Figure 4 compares the performance of conventional systems and AI-powered systems across four key metrics: Downtime, incidence rate, maintenance costs, and process stability. The process stability of AI powered systems is around 80%, which is 40 h per year downtime, compared to 120 h for conventional systems. In addition, the incident rate of AI powered systems is reduced to approximately 4 incidents per year, compared to 10 for conventional systems, suggesting improved safety and operational integrity. In addition, AI powered systems require <$20,000 per year in maintenance costs compared to $50,000 per year for conventional systems. In general, the figure shows that AI powered systems are more efficient, safer, and cost effective than traditional methods.
- Comparison of conventional versus artificial intelligence-powered systems.
CHALLENGES AND LIMITATIONS
As it has been noticed that the application of AI in green chemistry is highly promising but has its own limitations and challenges as well. These challenges include technical, economic, and regulatory factors as presented below.
Technical challenges
Data quality and availability
Training good AI models are dependent on the quality of data fed to the artificial systems to mimic. As in other areas of chemical analysis, it is challenging in green chemistry to obtain numerous and precise data sets due to the complexity of chemical reactions and variability in experimental conditions.[20]
Model interpretability
AI techniques can be used well to complement current chemical processes and industrial structures; however, the integration of AI tools into existing chemical processes and structures may be challenging. This is not easy to achieve because introduction of AI techniques with conventional systems and processes is often very difficult and cannot be replaced that easily.[12]
Integration with existing systems
AI tool integration with chemical processes and industrial architecture is not a simple process. Integration challenges come up since numerous existing systems were not developed with AI integration in mind and will require an overhaul for the AI integration to happen seamlessly. Furthermore, implementing AI-based tools require modifications on existing processes, technologies of computers and other digital devices as well as applications. Such changes can interfere with regular business functioning and demand sheer efforts to facilitate proper integration, so the integration becomes rather difficult.[19]
Economic challenges
High initial costs
The use of IT to support AI solutions means that there incurs a high initial costs for the identification and purchase of the technologies, collection of the data, and integration of the technologies into the supporting systems. These costs may become challenging for small firms and research organizations due to the high levels of complexity.
Return on investment (ROI)
It can often be difficult to quantify the ROI made with AI when it comes to green chemistry. For the benefits of AI to emerge, time may sometimes be required and therefore such benefits are not easy to monetize.[16]
Regulatory and ethical issues
Regulatory compliance
As it stands, the application of AI in chemical processes most go through existing regulations and standards. The following are some of the challenges when trying to deal with the above regulatory requirements; they are as follows;
Dealing with these regulations are not only cumbersome but also time-consuming
Tricky when dealing with novel technologies.[16]
Ethical concerns
AI application in green chemical processes is surrounded with ethical issues concerning data security, privacy, and unanticipated outcomes. Ethically, the appropriate use of AI systems is a considerable task.[21]
Data privacy and security
Data privacy
While it is relatively simple to obtain sizeable datasets to train the AI models, the process entails potential avenues for privacy violation. Keeping personal information and maintaining them as anonymous and safe always poses a big challenge.[22]
Cybersecurity risks
Chemical processes are prone to be at risk when AI systems that are responsible for controlling them undergo cyberattacks. It is crucial to note that cybersecurity is critical to ensuring that organizations are shielded from possible threats.[14]
FUTURE DIRECTIONS
AI in green chemistry is relatively in its early stage, and a number of new research perspectives for the future development is identifiable. These objectives attempts to fill the existing gaps, improve the impact of AI solutions, and make additional advancements in the sustainable chemistry concept.
Enhanced data integration and quality
Future research should seek to enhance the integration and quality of data that are incorporated in the models. The integration of multiple data types such as experimental data, simulation data, and real-time process data can improve the prediction capabilities of the AI algorithms. For the same reason, attempts to harmonize formats and enhance data exchange practices will also be critical.
Development of explainable AI models
To manage the problem of model accuracy explanation, there is a need to conduct research that will focus on deploying explainable AI models in the future. These models intend to explain how AI systems reach decisions, knowledge that is crucial for confidence and the appropriate use of AI in green chemistry.[23]
Integration with advanced computational techniques
AI should be complemented with other high-level computational methods including quantum computing and high performance computing to solve a number of chemical issues. Applying Quantum AI might help challenge the chemical simulations and enhance the processes compared at the scale which was never observed before.[18]
Expansion of AI applications to emerging areas
Further research studies should focus on finding the potential of AI in more novel and existing themes of green chemistry such as in biodegradable products, renewable energy systems, and carbon capture. AI is expected to be extremely important in coming up with solutions to these complex issues for which no straightforward solution exists.
Addressing ethical and regulatory challenges
Future steps should also involve the efforts to develop an understanding of ethical and regulatory issues of AI in green chemistry.[24] The application of transparent processes and values to determine the AI-generated decisions, strategies, policies, and procedures will be critical for safer and successful deployment of technologies; it will also be crucial to define the frameworks to address appropriate AI use and conform to the regulatory requirements.
CONCLUSION
AI in green chemistry is another progressive way of improving and implementing concepts of SCP. This study has pointed out on how AI applications have given great breakthroughs in areas such as chemical processes enhancement, green solvents design, and improved real time monitors. Although, these considerations have been partly addressed, there are still hurdles that still persist such as data quality, model interpretability, as well as the regulatory issue. The future research direction can be the enhancement of data integration process, the development of software for explainable AI, and AI application in the field of material production and renewable energy. Mitigating ethical and regulatory concerns will be of great importance in order to properly approach the problem of AI adoption. Finally, there is a necessity to further explore the opportunities of integration and application of AI, as well as to focus on interdisciplinary cooperation to achieve more significant results in the context of the sustainable development of chemical practices.
Acknowledgment:
The authors thank the management of Sri Ramachandra Institute of Higher Education and Research, Chennai, India for infrastructural facilities to carry out the research work and the Principal, Sri Ramachandra Faculty of Pharmacy, SRIHER (DU) for the constant motivation and encouragement.
Ethical approval:
Institutional Review Board approval is not required.
Declaration of patient consent:
Patient’s consent not required as there are no patients in this study.
Conflicts of interest:
Dr. Vinodhini Chandrasekar is on the Editorial Board of the Journal.
Use of artificial intelligence (AI)-assisted technology for manuscript preparation:
The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.
Financial support and sponsorship: Nil.
References
- Green chemistry: Theory and practice. Oxford: Oxford University Press; 2000. Available from: https://academic.oup.com/book/53104?login=false [Last accessed on 2024 Oct 28]
- [CrossRef] [Google Scholar]
- The elements of statistical learning: Data mining, inference and prediction. Math Intell. 2005;27:83-5.
- [CrossRef] [Google Scholar]
- A proposal for the Dartmouth summer research project on artificial intelligence, August 31, 1955. AI Mag. 2006;27:12.
- [Google Scholar]
- Generative adversarial nets In: Advances in neural information processing systems. Vol 27. United States: MIT Press; 2014. [Last accessed on 2024 Oct 28]
- [CrossRef] [Google Scholar]
- Human-level control through deep reinforcement learning. Nature. 2015;518:529-33.
- [CrossRef] [PubMed] [Google Scholar]
- Deep medicine: How artificial intelligence can make healthcare human again. Paris: Hachette UK; 2019 Available from: https://books.google.co.in/books?hl=en&lr=&id=_EFlDwAAQBAJ&oi=fnd&pg=PT8&dq=6.%09Topol,+E.+(2019).+Deep+medicine:+how+artificial+intelligence+can+make+healthcare+human+again.+Hachette+UK.&ots=BJ4Bx-2y9Q&sig=auIPhqGajK2pHPLwwn12j06epkw&redir_esc=y#v=onepage&q=6.%09Topol%2C%20E.%20(2019).%20Deep%20medicine%3A%20how%20artificial%20intelligence%20can%20make%20healthcare%20human%20again.%20Hachette%20UK.&f=false [Last accessed on 2024 Oct 28]
- [Google Scholar]
- Regularization for deep learning In: Deep learning. Cambridge: MIT Press; 2016. p. :216-61.
- [Google Scholar]
- Reaction condition optimization for non-oxidative conversion of methane using artificial intelligence. React Chem Eng. 2021;6:235-43.
- [CrossRef] [Google Scholar]
- Opportunities and challenges of artificial intelligence for green manufacturing in the process industry. Engineering. 2019;5:995-1002.
- [CrossRef] [Google Scholar]
- A decade advance in eco-efficiency and cost-benefits of emissions reduction targeting fine chemical manufacturers. J Environ Manage. 2021;298:113476.
- [CrossRef] [PubMed] [Google Scholar]
- Some studies in machine learning using the game of checkers. IBM J Res Dev. 1959;3:210-29.
- [CrossRef] [Google Scholar]
- Speech and language processing. Hoboken: Prentice Hall; 2000 Available from: https://books.google.co.in/books?hl=en&lr=&id=LCnx6xaDZsIC&oi=fnd&pg=PA21&dq=12.%09Jurafsky,+D.+(2000).+Speech+and+language+processing.&ots=h8OGdwYFPy&sig=erQE3SxoAMQlAhCr9ecGUP3FIRk&redir_esc=y#v=onepage&q=12.%09Jurafsky%2C%20D.%20(2000).%20Speech%20and%20language%20processing.&f=false [Last accessed on 2024 Oct 28]
- [Google Scholar]
- “Why should I trust you?” Explaining the predictions of any classifier In: Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. New York: Association for Computing Machinery; 2016. p. :1135-44.
- [CrossRef] [Google Scholar]
- Real-time monitoring and fault detection in Ai-enhanced wastewater treatment systems In: The AI cleanse: Transforming wastewater treatment through artificial intelligence: Harnessing data-driven solutions. Cham: Springer Nature Switzerland; 2024. p. :165-99.
- [CrossRef] [Google Scholar]
- Transparent AI-assisted chemical engineering process: Machine learning modeling and multi-objective optimization for integrating process data and molecular-level reaction mechanisms. J Clean Prod. 2024;448:141412.
- [CrossRef] [Google Scholar]
- Predictive analysis of methane-enhanced conversion of organic waste into sustainable fuel: A machine learning approach. Process Saf Environ Protect. 2024;189:1428-49.
- [CrossRef] [Google Scholar]
- Application of machine learning methods for the prediction of organic solid waste treatment and recycling processes: A review. Bioresour Technol. 2021;319:124114.
- [CrossRef] [PubMed] [Google Scholar]
- A systematic review on intensifications of artificial intelligence assisted green solvent development. Ind Eng Chem Res. 2023;62:20473-91.
- [CrossRef] [Google Scholar]
- The algorithmic society: Technology, power, and knowledge. Milton Park: Routledge; 2020. Available from: https://www.taylorfrancis.com/books/edit/10.4324/9780429261404/algorithmic-society-marcschuilenburg-rik-peeters [Last accessed on 2024 Oct 28].
- [CrossRef] [PubMed] [Google Scholar]
- Artificial intelligence: A modern approach. 2016. London: Pearson; Available from: https://api.pageplace.de/preview/DT0400.9781292401171_A41586057/preview-9781292401171_A41586057.pdf [Last accessed on 2025 May 23]
- [Google Scholar]
- Chemistry and technology innovation to advance green and sustainable chemistry In: Sustainable green catalytic processes. Hoboken: John Wiley and Sons; 2024. p. :273-300.
- [CrossRef] [Google Scholar]
- Chemical data intelligence for sustainable chemistry. Chem Soc Rev. 2021;50:12013-36.
- [CrossRef] [PubMed] [Google Scholar]
- Artificial intelligence applications for industry 4.0 A literature-based study. J Ind Integr Manage. 2022;7:83-111.
- [CrossRef] [Google Scholar]
- Local public services and the ethical deployment of artificial intelligence. Gov Inf Q. 2023;40:101865.
- [CrossRef] [Google Scholar]