Great potential is being shown by artificial intelligence (AI) especially in changing the face of traditional farming, with sophisticated tools like machine learning and deep learning, detection of early plant diseases is now possible. These intelligent systems can analyze a large amount of agricultural data while understanding patterns and developing insights that can help farmers obtain better yields and grow healthier crops. However, despite the progress made by these systems, they have not yet taken over a larger part of the industry. The major reason is that AI models and tools are seen as “black boxes” that cannot be explained with ease to the farmers, and this gap creates a lack of trust, hindering them from bringing these tools to their daily routine.

While still growing and developing at an accelerated pace, AI has already augmented human life. This technology is now increasingly commonplace in our homes, workplaces, travels, healthcare, and schools. What would have seemed like science fiction just two decades ago (such as a phone detecting the exact disease a plant has and virtual personal assistants) is set to become a fixture of our daily lives [1]. As AI becomes more integrated into our lives and its role in decision-making in different industries expands, more ethical issues and concerns are becoming apparent. Primarily, the role of AI is to be of great interest to people and organizations and positively impact them, not the other way around. Many artificial intelligence techniques have already been applied to agricultural tasks, such as the use of self-learning systems for large-scale crop harvesting, precision agriculture to aid in the detection of plant diseases, soil monitoring, animal disease diagnosis, and Internet of Things (IoT) devices, which refer to the idea of connecting physical objects to the Internet to enable remote control and monitoring. As noted by Okengwu et al. [2], artificial intelligence (AI) is a powerful tool that can help manage the negative impacts on the environment more effectively by minimizing ecosystem damage and preventing further harm.

Developing responsible AI creates huge potential for strengthening key sectors, with agricultural systems becoming more people-focused, resilient, sustainable, and equitable. It is important to note the base reason for this intelligence, which is to empower the experts without pushing them off the picture; the human touch should be preserved at the heart of farming. To fully harness AI’s capabilities of AI in agriculture, interdisciplinary cooperation, maintenance of ethical standards, and safeguarding of farmers rights should be the top priorities [3]. In line with these, Okengwu et al. [4] examined the viability of AI tools and technology in crop management by using this tool for the classification of tomato leaf diseases. Their work was more focused on the detection of diseases in plants, giving limited attention to the issues of fairness and accessibility, especially how smallholder farmers in low-resource environments would afford this tool. This limitation brings to the fore the argument that AI advances need to be closely aligned with the ethical principle of equitable access for all and inclusion. Han and Sun [5] covered the Yangtze River Economic Belt between 2011–2023, studied how AI is shaping and realigning the future of eco-friendly farming. The authors proved, despite several trials, that incorporating AI into agriculture greatly enhanced green development. This AI advancement could have a ripple effect from a local district to neighboring areas, improving environmental performance in farming.

There has been a shift in Responsible AI concerns, which now pertains to ethics, social and human consequences. Which initially pertained to AI development, implementation and deployment of these AI systems. Responsible AI handles and considers fairness, accountability, and transparency which are ethical values during the development of AI tools and applications. These tools and applications developed must be devoid of algorithmic bias, privacy infringement, and supports equity, human rights, and societal wellbeing are the main criteria to be considered responsible. While these criteria require more than just technical solutions (explainability and data security), it also requires a firm governance structure that offers oversight and stakeholders’ involvements in a meaningful manner. Besides these requirements, responsible AI development also requires considering socio-economic implications of AI development critically, including its implications for the labor force and agricultural livelihoods. When these ethical factors are totally considered, the AI tools are used responsibly. Hence, shifting from the abstract idea of AI into a better practical approach, which would gain trust and ensure a sustainable and equitable development from this technological advancement. Artificial intelligence can be of various forms, from software and algorithms to a physical form, like a farm machine, drones, cars, and even robots. It becomes necessary for us to make a distinction between the technology that actually uses AI and the technology that only uses automation. Generally, it has been noted that not all robots and programmed systems actually portray intelligence. This is because most of the ‘robots’ only work through a set of instructions, either through human control or with a very limited level of autonomous control. Ryan [6], makes a clear distinction between rule-based machine designed for specific task, industrial robots and more advanced robots that can employ intelligence artificially to learn, adapt and make decisions within a changing environment in his work “The social and ethical impacts of artificial intelligence in agriculture: mapping the agricultural AI literature,”. Within the agricultural context, this distinction is crucial as it helps AI researches and discourse to focus on tools and systems that demonstrates adaptive intelligence rather than just traditional farming machines. Accordingly, this paper is concerned only with devices, programs, and applications that incorporate some degree of artificial intelligence, whether in decision-making, prediction, or autonomous operation.

Currently, AI is used in agriculture to increase production, sustainability, and efficiency. This translates to ethical and responsible AI concepts becoming increasingly important. While the models and capabilities used in AI food system technologies may, in many cases, be of a different kind than the ones gaining the most exposure in media and public discourse, trust is all the more essential for agriculture and food system technologies, given food’s pervasiveness and significance to human life [7]. For ethical or regulatory frameworks to be effective in facilitating trust, an important first step is to identify who is responsible for identifying ethical concerns, developing ethical frameworks, and implementing them in practice. By putting these questions in context, we aim to see how these ideas work in the field of agriculture.

One challenge that arises with the development of artificial intelligence (AI) for agriculture is ensuring that the emerging system behaves responsibly. Buchholz et al. [8] addressed the importance of responsibility in the context of artificial intelligence (AI) in manufacturing and presented a methodology for implementing these concerns in an AI development project. They started by pointing out that while many theoretical frameworks and methods are available to handle accountability concerns within AI environments, the implementation of these frameworks within project development remains obscure. We now attempt to determine how these principles would be implemented in the agricultural space.

In 2019, the Organization for Economic Cooperation and Development (OECD) and its 42 partner nations agreed to uphold international standards aimed at ensuring that AI systems are designed to be reliable, safe, equitable, and trustworthy. This was reportedly the first set of intergovernmental policy guidelines on artificial intelligence [15].

Gurría [16] noted how artificial intelligence is transforming our way of life and work while also having a huge positive impact on our economy and society. He further pointed out how technology creates new difficulties and stokes worries about morality. According to him, ‘this places the onus on governments to make sure that AI systems are designed in a way that respects our values and laws, so people can trust that their safety and privacy will be paramount.’

There are no specific AI rules or regulations worldwide. Rather, several countries and regions have adopted different approaches to controlling the development and implementation of AI technology ethically and responsibly. Data protection, transparency, and the overall effect of AI on jobs and security are major issues for countries. To ensure the ethical development of AI and its global implementation in some capacity, cooperation is essential [17].

Considering the strategies that various countries are employing to implement their national AI policies, we sneak a peek through a crystal-clear window to see how government stakeholders are dealing with the growth of AI. While developing AI policy documents, it is not just a list of dos and don’ts; it also shows what each country cares about. Whether it is winning the race of technological advancement, economic growth, or just ensuring that guardrails and sustainability laws are placed at the heart of development. By examining how and when these strategies are developed, we can observe trends, quirks, and differences that occur worldwide and even regionally to ensure that everything is kept under regulation. To facilitate visualization, Table I and Fig. 2 present a timeline of the evolution of AI governance from 2017 to 2023 across ten countries. It also examines how the world of AI governance has been moving in regions and how it has shifted within just a few years to ensure that policy goals are met.

AI policies are now starting to look very different from one region to the next across the world, due to the fact that every region has its own set of priorities and a unique way of handling regulation. The United States does not have an all-encompassing federal law for AI; rather, the government is tackling it one by one. This is done using specific rules for different sectors, such as health, agriculture, and finance. One big move that has happened recently is the 2023 Federal Artificial Intelligence Risk Management Act [18], which was created to guide government agencies on firm standards for testing and double-checking AI systems before deployment to the general public. State-level activities, such as the “California Privacy Law”, continue to play a vital role in shaping how data-driven AI tools are developed and deployed to users.

The European Union took a different route when it rolled out the AI Act in 2021 [19]. It set up a regulatory framework with the goal of ensuring that AI systems, when developed, do not just act as black boxes but show a good level of transparency, traceability, and fairness. The need for compliance and assessment of the data used has made the EU take the blueprint of the GDPR and upscale it. The EU intends to achieve a delicate balance that allows technology more room for growth without compromising individual rights and developer accountability.

In 2017, instead of waiting, China decided to take a very clear goal and started by launching the New Generation AI Development Plan to enable them to lead the world in AI by 2030. The broad strategy touched on all aspects, including data security and the rules and ethics of using AI. According to Sheehan [20], the Chinese mission is to find developers who can interpret the cover of their algorithms. This is to ensure that the government and the general public have a good understanding of how these systems make decisions and the data used. Within the same year, the United Arab Emirates joined China with the 2017 Strategy for Artificial Intelligence [21], which was a critical part of their “UAE Centenary 2071” vision. Focusing on the bigger picture of diversifying their economy and ensuring their workforce is ready for the future, they chose not to put much emphasis on the code used in developing these tools. They are actually betting big on AI to have a huge impact on the basics of life, healthcare, transport, energy, and farming. Together, these countries show a proactive, innovation-first model of AI governance.

Later adopters of national AI policies placed a much higher emphasis on ethics and public trust. In 2019, the United Kingdom sought out individuals in the AI development cycle to actively address ethical concerns with the release of the AI ethics handbook [22] to ensure governance throughout the project lifecycle. In the same year, Japan introduced its “Social Principles of Human-Centric AI” [23], which was a framework built with the idea of sustainability, diversity, and fundamental respect for human dignity. The goal was to develop a set of rules that ensures AI advances in a way that reflects social values instead of jumping into rigid laws. This is part of the OECD’s global drive for inclusive development. However, Canada has adopted a more structured approach to the development of its Artificial Intelligence and Data Act (AIDA), introduced in 2022. This is to closely monitor highly impactful AI systems. Hence, it is necessary to encourage better accountability and prevent harm before it occurs. Brandusescu and Sieber [24], while critiquing the process, noted some of the major checks of reality. For example, innovation, although good, is not enough on its own. Perhaps the benefits of new technologies may not necessarily accrue to everyone.

Within Africa, AI has been positioned as a catalyst for socio-economic development. In Mauritius, in 2018, a strategy was developed that positioned AI as a new means of infusing life into local industries while also creating new opportunities for economic growth [25]. Around the same period, Kenya began taking its first steps towards creating a comprehensive AI strategy by forming its Distributed Ledgers and AI Task Force [26]. Egypt finally joined the discussion on AI when it launched its National Strategy, connecting AI to its ambitions of attaining sustainable development while improving partnerships in the region three years later [27]. Although they are serious about using AI to drive development, they still face challenges related to infrastructure and regulation.

When these countries are viewed comprehensively, it becomes apparent that there is an emerging trend. While countries like China and UAE, who were early movers, have demonstrated that they were only interested in ensuring they were competitive with a focus on rapid innovation. Countries like Japan and the UK, who came on board later, have moved towards ethical development with a focus on trust. The latest additions, like the EU, Canada, and the US, seem to be relying even more on regulatory clarity, which will result in better accountability as well as a better plan for risk management. On the other hand, countries in Africa seem to be highlighting the potential of AI in driving sustainability as well as socioeconomic development. The real-world benefits have been witnessed in countries that have adopted this technology at an early stage of the farming industry. By using AI for monitoring crops and for streamlining the supply chain, it gave them a good push forward in terms of yield and efficiency. Countries like Canada, the UK, and Japan have implemented a strategy on AI at the beginning of the 2020s. This helped them push their agricultural sector quite far into the future by using smart farming and data policy strategies. In the case of China, although the system was launched in 2017, some of the provinces, particularly in the Yangtze River Economic Belt, started using AI for monitoring crops and for pests and irrigation systems. By reviewing the data from the year 2011 till 2023 [5], it became quite evident that the system not only helped the local farming sector go green but also had a spill-over effect on the neighboring regions.

Similarly, the UAE has set an early pace in implementing its policy on artificial intelligence, thereby hastening the pace of precision and smart-farming. Concepts such as AI-based water management and intelligent greenhouses are becoming increasingly popular. This enables farmers to produce more agricultural products, even under unfavorable conditions, without wastage or unnecessary expenses. In this regard, the agricultural sector has demonstrated its resilience even in the face of serious challenges, such as climate change and food security. According to the Food and Agriculture Organization (FAO) [28], not only does the early implementation of AI policies encourages innovation and helps maintain the competitiveness of the respective countries. They will be able to address any future risks and ensure profitability.

The aim of the questionnaire was to obtain an exact picture of how AI can be used in agriculture from an ethical perspective. For this, we largely depended on the opinions of industry experts who use these technologies. Their opinions proved to be of utmost importance understanding the sentiment of the people regarding artificial intelligence, the concerns people have regarding these technologies, and what they consider to be of utmost importance with regard to these technologies. The survey responses were voluntary and confidential.

Of all the respondents, the majority were farmers (42.7%), followed by researchers (31.7%) and technology developers (7.3%). All respondents were actively engaged in the integration of AI technologies in agriculture and had practical experience using these technologies. Fig. 3 shows a chart representing the respondents.

The data show the respondents’ familiarity with the ethical considerations of AI technologies on a five-point scale, with five representing high familiarity. Fig. 4 shows that more than half of the respondents lack familiarity with these considerations and are therefore not likely to take them into account. Although policies on AI are being drafted and implemented in most parts of the world, the dissemination of such information is still required. This is supported by Fig. 5, which shows that 27.8% of the respondents are aware of AI-related policies and regulations.

This low level of awareness highlights a significant gap in knowledge among professionals, emphasizing the need for more comprehensive education and dissemination of information regarding AI-related policies and regulations in the healthcare sector.

The survey responses were diverse in terms of the respondents’ views on the people responsible for the ethical use of artificial intelligence (AI) in agriculture as shown in Fig 6. A good number of the respondents supported the concept of collective responsibility of various people or entities in the use of artificial intelligence in agriculture. These people include AI developers, policymakers, agricultural institutions, and AI end users. A good number of the respondents were of the view that the developers of artificial intelligence and the institutions that use it are responsible for ensuring that it is compatible with ethical standards. However, policymakers were also instrumental in formulating policies and laws governing the use of AI in the sector.

The highlights point to several potential benefits that may be derived by the agricultural sector in the next five to ten years as a result of the application of artificial intelligence. First, there is an expectation that AI will greatly improve crop productivity in the agricultural sector. This will be achieved through the application of precision farming techniques, which will be made possible by the analysis of large amounts of data using AI systems. Second, AI is expected that AI will improve resource management in the agricultural sector. This will be achieved by applying AI to streamline logistics and deliver products to the market efficiently.

Overall, the respondents were positive, as shown in Fig. 7, that AI would not only improve agricultural productivity but also lead to more sustainable agricultural practices, which will be beneficial to both the agricultural sector and the wider economy in the long term.

Current survey results show continued optimism regarding the role of AI in transforming agricultural practices over the next ten years. In fact, a majority of those surveyed expect significant developments, as shown by 72% expecting crop yields to increase, 73% believing that food security will benefit from a significant boost, and over 70% expecting better resource usage alongside reduced labor costs. However, it is generally accepted that AI has much broader benefits than just increasing productivity, as it has the potential to ensure sustainability in agricultural practices. This has already been observed in precision agriculture, where decisions are based on data rather than conjecture. An excellent case in point is the work of Okengwu et al. [29], who created a method that allowed smartphones to run deep-learning models in realtime for crop diseases. This, in essence, shows that AI can be used ethically, as it does not require large budgets or infrastructure to be effective.

The main aim of this paper is to emphasize the importance of ethical and responsible artificial intelligence in the development and use of these technologies worldwide. By analyzing the development of AI policies in various countries worldwide, it can be understood that a major pattern emerges from the analysis. By developing and implementing AI policies, countries can reap major benefits, especially in the agricultural sector. Along with the benefits, the country would also be able to deal with the challenges that will emerge in the agricultural and food production sectors in the future. In the future, it is important to ensure the development of AI by incorporating the two aspects of ethics and accountability. By doing so, it would be clear that there would be significant scope for the development of the agricultural sector. At the same time, it ensures the well-being of people in the agricultural sector worldwide. By doing so, it would be possible to ensure the development of the agricultural sector through AI while simultaneously ensuring a better future for the people in the sector.

Therefore, further studies are necessary to determine the mechanisms for making AI accessible to smallholder farmers and to develop training programs for understanding ethical AI practices. Integrating AI ethics into agricultural extension services would help enhance transparency, equity, and accountability. This would not only instill trust among the stakeholders involved in agricultural development but would also ensure that AI contributes substantially to the development of agriculture.