Introduction

Artificial Intelligence (AI) is a hot topic these days, and for a good reason! AI, in simpler terms, refers to the capability of machines that undertake those tasks that requires the intelligence of a human being like problem-solving, learning and reasoning. From healthcare to transportation to entertainment, various aspects of our lives hold the promise of transformation thanks to this technology. This blog will explore the definition of AI, its benefits and limitations, and its role in automation. The blog will also discuss the algorithmic and datafication aspects of AI, examining how data is used to train AI systems and the potential implications of this process. By the end of this blog, readers will have a better understanding of what AI is, how it works, and its impact on our society and daily lives.

Definition of Artificial Intelligence (AI)

Stuart Russell and Peter Norvig, two well-known computer scientists, defined AI in their book “Artificial Intelligence: A Modern Approach” as the study of agents that receive precepts from the environment and perform actions. According to the authors, an agent is a computational system that operates autonomously in some environment, such as a robot or a software program who precepts are the inputs or information that the agent receives from the environment, while actions are the outputs or effects that the agent has on the environment (Russell & Norvig, 2021). Microsoft’s definition of AI highlights that AI technology has the potential to undertake tasks that generally requires the intelligence of human beings (Microsoft, 2023).

Kate Crawford’s in her publication “Atlas of AI,” suggests that AI is not just a technology or an industry, but a complex socio-technical system that has far-reaching implications for society and the planet (Crawford, 2021). Crawford argues that AI is an idea that shapes our understanding of intelligence, cognition, and the relationship between humans and machines. AI is also an infrastructure that requires vast amounts of energy and mineral resources to operate. This means that the creation of contemporary AI systems is dependent on the exploitation of natural resources, particularly in the Global South, where many of these resources are found (Crawford, 2021). AI systems are trained on massive amounts of data, and this data shapes the way that they perceive and interpret the world. This means that AI can reinforce existing biases and stereotypes, particularly if the data used to train these systems is not diverse or representative of different groups. Also, Frank Pasquale, in his book “The Black Box Society” contends that the use of AI systems can perpetuate existing biases and discrimination and that their impact on society and democracy must be carefully scrutinized and regulated (Pasquale, 2015).

Benefits and Limitations of AI

Artificial Intelligence (AI) possess immense potential to transform individual’s lives by eradicating tasks that are complex in nature like data analytics, advanced reasoning, and critical understanding. Key benefits of AI are as follows:

Figure 1: Benefits of AI

Made by author

1. Efficiency and Productivity: Individuals can use AI to automate tasks that are repetative and time-consuming in nature. This will allow them to focus on more advance tasks and creative work. This can lead to increased productivity and efficiency in various domains, such as manufacturing, logistics, and customer service (Challen et al., 2019).
2. Improved Healthcare: AI can help diagnose diseases, design treatment plans, and monitor patient progress, leading to improved healthcare outcomes and reduced healthcare costs.
3. Enhanced Education: AI can personalize learning experiences for students, allowing them to learn at their own pace and style. This can lead to improved educational outcomes and increased access to education for people in remote or underprivileged areas (Dwivedi et al., 2019).
4. Increased Safety: AI can be used to detect and prevent accidents and disasters, such as traffic accidents, natural disasters, and industrial accidents.
5. Improved Environmental Sustainability: AI can help optimize resource use and reduce waste, leading to improved environmental sustainability and reduced carbon footprint.
6. Enhanced Entertainment: AI can personalize entertainment experiences for users, such as recommending movies or music based on their preferences.

The benefits of AI are vast and wide-ranging, and have the potential to transform various aspects of human life.

Figure 2: 7 challenges of AI
(Hall, 2018)

In a 2018 article, the World Economic Forum identified seven challenges for the development and implementation of creative AI in the field of journalism (Hall, 2018). However, these challenges are universal and are applicable in every field associated with AI. These challenges are further divided into technical and governance challenges. Technical challenges in AI refer to the difficulties faced in developing, designing, and implementing AI systems. These challenges can include issues related to data quality, model complexity, interpretability, scalability, and optimization (Hall, 2018). Governance challenges in AI refer to the ethical, legal, and societal issues that arise with the use and deployment of AI systems. These challenges can include concerns around privacy, bias and discrimination, transparency and accountability, safety and security, and the impact of AI on employment and society.

Figure 3: Governance and Technical Limitations of AI

Made by author

Technical Limitations

1. Availability of data: The availability of data is one of the key challenges in the development and implementation of artificial intelligence (AI) systems. As mentioned in the previous response, machine learning algorithms rely heavily on data to recognize patterns and learn from them (Hall, 2018). Without enough data, the accuracy and effectiveness of AI systems can be limited.
2. Understanding unstructured data: As mentioned in the previous response, AI systems rely heavily on data to recognize patterns and learn from them. However, much of the data available today is unstructured, such as images, videos, and text, which can be difficult for AI to process and synthesize.
3. Lack of self-awareness: Unlike human beings, AI is unable to explain its output and the reasoning behind it. This can be a major issue in the creative economy, where consumers expect transparency and accountability from content providers (Allam & Dhunny, 2019).
4. Verifying authenticity: It is a critical challenge in the age of fake news and disinformation. AI systems cannot distinguish between accurate and inaccurate input, which can lead to the false or misleading output.

Governance Limitations

• Redefining copyright and fair use: The issue of copyright and fair use is a complex one in the context of artificial intelligence. As AI systems learn from human-created content, there is the potential for these systems to generate their own content that could be subject to copyright protection (Dash et al., 2019). This raises questions about who owns the copyright to the output generated by AI systems, and whether the use of copyrighted material by AI systems constitutes fair use.
• Ensuring corporate accountability: The rise of AI in the media industry has given rise to concerns over accountability. As one cannot consider the machine as legally liable for any infringement thereby, human accountability is considered to be necessary throughout the content value chain (Russell & Norvig, 2021). This includes content distributors like Facebook, Google, and Twitter, who hold unprecedented power to shape public opinion through their AI algorithms.
• Exacerbating asymmetrical power: The development and implementation of AI require significant resources, and smaller companies may not have the means to invest in such technologies. This could lead to an even greater concentration of power and influence in the hands of a few large companies (Breadmore et al., 2019). Additionally, the use of proprietary AI systems could create barriers to entry for new players in the industry, further exacerbating the concentration of power.

AI in the field of automation

AI has been playing an increasingly important role in the field of automation, from manufacturing to logistics to customer service.

Figure 4: AI in automation
Made by author

1. Robotics: AI has been implemented in the field of robotics where users are using AI-powered robots to undertake various tasks like managing inventory, welding, and assembly in the manufacturing industry. For example, Tesla uses robots to assemble cars in their factories (Endsley, 2017).
2. Autonomous Vehicles: Another example of AI in automation is in autonomous vehicles. Self-driving cars now come with advanced sensors, camera based on machine learning algorithms that facilitate in navigation on the road without human intervention. Companies like Waymo, Tesla, and Uber are all working on autonomous vehicle technology (Perkins & Murmann, 2018).
3. Chatbots: Chatbots are AI-powered virtual assistants that can handle customer service inquiries and support tasks. They can be programmed to answer common questions, troubleshoot issues, and provide information. For instance, e-commerce giant Amazon recently revealed that the employees in the company has been using OpenAI’s ChatGPT for software coding (Kim, 2023).
4. Supply Chain Optimization: Modern supply chain management is leveraging AI to optimize supply chains by anticipating demand, managing levels of inventory, and synchronising shipping routes. For example, FedEx has been using an AI-enabled Alexa app to manage inventory, shipments and routes (Woyke, 2017).

Algorithms & Datafication

Algorithms are sets of instructions or rules that a computer follows to complete a specific task. They can be used for various purposes, including data processing, problem-solving, and decision-making (Abualigah & Diabat, 2021). Datafication refers to the process of converting different aspects of our daily lives and activities into digital data. With the increasing use of technology and the internet, almost everything people do leaves a digital footprint that can be collected, analysed, and utilized. Datafication enables the collection and analysis of this data, which is then used to train AI models to perform tasks such as image or speech recognition, natural language processing, and decision-making (Çaliş & Bulkan, 2015).

In her book “Algorithms of Oppression,” Safiya Umoja Noble argues that algorithms are not neutral tools but are shaped by the values and biases of their creators. She explained how search engines like Google use complex algorithms to determine which results are displayed when a user types in a search term (Noble, 2018). However, these algorithms are not objective, but rather reflect the values and beliefs of their creators. This means that search results can be influenced by factors such as user location, browsing history, and the popularity of certain websites. As a result, search engine results can often reinforce negative stereotypes and discriminatory practices. Noble also discusses how algorithms are used in the context of neoliberalism, where they are often used to automate decision-making processes and reduce costs, but can also lead to the erosion of privacy and the concentration of power in the hands of a few large tech companies (Noble, 2018). OpenAI ChatGPT is a recent example of an AI that generates a reponse to user input by using natural language processing. It is designed to learn from a large dataset of text and generate human-like responses to user queries, making it useful for a variety of applications such as customer service, chatbots, and personal assistants (Hill-Yardin et al., 2023).  However, as with any AI algorithm, there are potential risks associated with its use.

OpenAI’s ChatGPT has been accused of breaching privacy enactments by using people’s personal information without their consent. OpenAI has developed ChatGPT in a way that it systematically gleaned around 300 billion words all over the internet that, includes personal information without informed consent. The company on the other hand, does not provide individuals with any kind of procedure or options that allow the company to hold their data or request its deletion which is a guaranteed right in accordance with the European General Data Protection Regulation (GDPR) (Gal, 2023; Pollina & Mukherjee, 2023). Overall, while AI algorithms like ChatGPT have the potential to be transformative in their applications, it is important to be aware of the potential threats related with their use, including the concentration of power and erosion of privacy. It is essential that people consider the ethical and social implications of AI algorithms and work to ensure that they are developed and used in a way that is fair, transparent, and equitable.

Conclusion

In summary, AI now plays crucial plays in transforming the way people live, work and interact with the world. AI not only helps in enhancing human’s efficiency and accuracy, improves decision-making processes, and the ability to perform tasks but also facilitates routine tasks. This blog has defined AI and its role in the contemporary world. The blog has also discussed the benefits and limitations of AI, and its role in automation has also been discussed. AI can also be used to develop algorithms that allow machines to learn from data, known as machine learning. Eventually, it can be said that AI continues to evolve, it is crucial that we prioritize ethical considerations, diversity and inclusion, and responsible use to ensure that AI benefits all members of society.

References

Abualigah, L., & Diabat, A. (2021). Advances in sine cosine algorithm: A comprehensive survey. Artificial Intelligence Review, 54(4), 2567–2608.

Allam, Z., & Dhunny, Z. A. (2019). On big data, artificial intelligence and smart cities. Cities, 89, 80–91.

Breadmore, R., Marshall, K., & Kimber, R. (2019, July 16). Artificial intelligence (AI) and the great privacy challenge—KPMG Australia. KPMG. https://home.kpmg/au/en/home/insights/2019/02/artificial-intelligence-great-privacy-challenge.html

Çaliş, B., & Bulkan, S. (2015). A research survey: Review of AI solution strategies of job shop scheduling problem. Journal of Intelligent Manufacturing, 26, 961–973.

Challen, R., Denny, J., Pitt, M., Gompels, L., Edwards, T., & Tsaneva-Atanasova, K. (2019). Artificial intelligence, bias and clinical safety. BMJ Quality & Safety, 28(3), 231–237. https://doi.org/10.1136/bmjqs-2018-008370

Crawford, K. (2021). The Atlas of AI. Yale University Press.

Dash, R., McMurtrey, M., Rebman, C., & Kar, U. K. (2019). Application of artificial intelligence in automation of supply chain management. Journal of Strategic Innovation and Sustainability, 14(3), 43–53.

Dwivedi, Y. K., Hughes, L., Ismagilova, E., Aarts, G., Coombs, C., Crick, T., Duan, Y., Dwivedi, R., Edwards, J., & Eirug, A. (2019). Artificial Intelligence (AI): Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and policy. International Journal of Information Management, 101994.

Endsley, M. R. (2017). Autonomous driving systems: A preliminary naturalistic study of the Tesla Model S. Journal of Cognitive Engineering and Decision Making, 11(3), 225–238.

Gal, U. (2023, February 8). ChatGPT is a data privacy nightmare. If you’ve ever posted online, you ought to be concerned. The Conversation. http://theconversation.com/chatgpt-is-a-data-privacy-nightmare-if-youve-ever-posted-online-you-ought-to-be-concerned-199283

Hall, S. B. (2018, January 15). Can you tell if this was written by a robot? 7 challenges for AI in the media. World Economic Forum. https://www.weforum.org/agenda/2018/01/can-you-tell-if-this-article-was-written-by-a-robot-7-challenges-for-ai-in-journalism/

Hill-Yardin, E. L., Hutchinson, M. R., Laycock, R., & Spencer, S. J. (2023). A Chat(GPT) about the future of scientific publishing. Brain, Behavior, and Immunity, 110, 152–154. https://doi.org/10.1016/j.bbi.2023.02.022

Kim, E. (2023, January 27). Amazon employees are already using ChatGPT for software coding. They also found the AI chatbot can answer tricky AWS customer questions and write cloud training materials. Business Insider. https://www.businessinsider.com/tech/news/amazon-employees-are-already-using-chatgpt-for-software-coding-they-also-found-the-ai-chatbot-can-answer-tricky-aws-customer-questions-and-write-cloud-training-materials-/articleshow/97360169.cms

Microsoft. (2023). Artificial Intelligence vs. Machine Learning | Microsoft Azure. https://azure.microsoft.com/en-us/resources/cloud-computing-dictionary/artificial-intelligence-vs-machine-learning/

Noble, S. U. (2018). Algorithms of Oppression: How Search Engines Reinforce Racism. In Algorithms of Oppression. New York University Press. https://doi.org/10.18574/nyu/9781479833641.001.0001

Pasquale, F. (2015). The Black Box Society: The Secret Algorithms That Control Money and Information. Harvard University Press.

Perkins, G., & Murmann, J. P. (2018). What Does the Success of Tesla Mean for the Future Dynamics in the Global Automobile Sector? Management and Organization Review, 14(3), 471–480. https://doi.org/10.1017/mor.2018.31

Pollina, E., & Mukherjee, S. (2023, March 31). Italy curbs ChatGPT, starts probe over privacy concerns. Reuters. https://www.reuters.com/technology/italy-data-protection-agency-opens-chatgpt-probe-privacy-concerns-2023-03-31/

Russell, S., & Norvig, P. (2021). Artificial Intelligence: A Modern Approach (4th edition). Pearson.

Woyke, E. (2017, February 3). FedEx Bets on Automation as It Prepares to Fend Off Uber and Amazon. MIT Technology Review. https://www.technologyreview.com/2017/02/03/154207/fedex-bets-on-automation-as-it-prepares-to-fend-off-uber-and-amazon/