Legends & Lies of Artificial Intelligence

Folks, welcome back to our 'Tech Talk Tuesdays'!

Over the next nine weeks we will delve into Artificial Intelligence, (AI) The Good, The Bad, and The Ugly.

We wrote this series as extracts from our upcoming book of the same working title using ChatGPT ironically ;-)

The target audience are those interested in AI, but not deep in technology and those who may be deep in the subject but are responsible for communicating their knowledge to other non-technical people on the business side.

We hope you enjoy the series, scheduled over the next 9 Tuesdays and, as always, we welcome your feedback.

An Important Note: Before you go any further, we feel it is important you know any good effort is never accomplished alone. We would like to thank those mentioned here for their contributions, research, and support. This article and the nine that will follow, each Tuesday through March 12th owe our thanks to the folks mentioned and recognized below.

KPMG: “KPMG U.S. survey: “Executives expect generative AI to have enormous impact on business, but unprepared for immediate adoption”?

Carolyn Blais: for her work on; “When Will AI be Smart Enough to Outsmart People?”?

Logan Rohde: of Info-Tech Research Group: “Adress Security & Privacy Risks for Generative AI”?

Naveen Joshi: of the Cognitive World contributor Group via Forbes: 7 Types Of Artificial Intelligence (forbes.com)?

Shreeya Chourasia: of TRO. “7 Types of Artificial Intelligence and How they Work?”?

Lewis Maddison: of techradar pro. “Is the Cost of AI worth it for your business?”?

By Hillary: of TechBullion. “Exploring the Ethics of Gartner’s Generative AI: Impacts, Challenges, and Considerations”?

Dr. Mandar Karhade, MD. PhD.: “History of AI: Maturation of Artificial”?

Eliza Kosoy, a researcher in MIT’s Center for Brains, Minds, and Machines: “When will AI be smart enough to outsmart people?”?

It is truly a humble experience to witness how much people are willing to help if you simply ask.??

Overview: What is AI, or more accurate, what are AI??

Currently and for the foreseeable future Artificial Intelligence commonly referred to as ‘AI’ is actually; “A collection of multiple autonomous technologies and disciplines.”??

Important Note to the Reader:?

This article is the first one of ten and is an introduction to the exciting journey into the world of Artificial Intelligence ‘AI’, designed specifically targeted for those without a technical background.

While there are numerous resources available for delving into the technical intricacies of AI, these articles fill a gap by focusing primarily on the non-technical audience. It offers insights into what AI is, how it works, and how individuals can begin leveraging its capabilities without needing to become technologically proficient.

By pointing readers towards additional resources for more in-depth understanding, these articles aim to empower non-technical individuals to embrace AI and effectively integrate it into their lives.?With its user-friendly approach, it provides invaluable insights and practical tips on how to incorporate AI into your everyday life. By bridging the gap between complex technology and everyday users, this book empowers you to tap into the incredible potential of AI. Get ready to expand your thoughts, embrace innovation, and unlock a whole new world of possibilities, all without needing deep technical expertise. Let the power of AI transform your life!??

The targeted audience of this includes individuals who want to explore the potential of AI without becoming experts in the technical aspects of it. It caters to those who are curious about the application and benefits of AI in their day-to-day activities, such as improving productivity, making informed decisions, and enhancing creativity.?

In this article we focus our efforts on providing a comprehensive overview of AI, explaining its concepts and possibilities in a way that is accessible and easy to understand.??

AI is a rapidly evolving field encompassing a variety of technologies that aim to augment, emulate, and enhance human creativity.??

Addressing the concern, “Will AI displace workers?”?It is the opinion of the author that; “AI won’t displace you, people that use AI will.”?

Introduction (The Good, The Bad, The Ugly)?

The Good:?

Although AI is a powerful collection of technologies that are transforming businesses, industries, cultures, and lives, they also come with challenges and risks.??

NOTE: From this point forward in this and the following articles we will refer to the collective of technologies that make up AI in the singular for readability unless purposely referring to a breakdown or break out of those technologies.?

For individuals for whom AI is not their primary focus, understanding its potential and pitfalls can feel daunting. However, these comprehensive articles aim to bridge that gap, providing a clear and accessible introduction to the basics of AI. Free from technical jargon and exaggerated hype, this book empowers business leaders and IT professionals to make informed decisions about AI adoption and implementation. With its straightforward approach, even those who consider AI secondary can begin to grasp its principles and possibilities.?

Our goal is to demystify AI by explaining its fundamental concepts, how it works, and its limitations. By understanding the capabilities and limitations of AI, business leaders and IT professionals can make informed decisions about how it can be effectively integrated into their organizations.?

Moreover, we attempt to cover best practices and ethical principles for using AI, ensuring that its implementation aligns with industry standards and respects privacy, fairness, and transparency. By adhering to these principles, businesses can harness the power of AI while minimizing potential risks.?

With an optimistic tone, we aim to empower readers to embrace AI technology with confidence. It equips them with the knowledge necessary to navigate the complexities of AI adoption and implementation, enabling them to leverage its potential to drive innovation, efficiency, and growth in their organizations.?

We also want to provide real-world examples and case studies of how AI is applied in various domains, such as healthcare, education, finance, retail, manufacturing, education, your personal life, and more. Whether you are an industry or governmental leader who wants to leverage AI for your organization's growth and innovation, an IT professional who wants to support your business leader's vision and goals with AI, or an individual who is seeing a major evolutional change, it is our goal to help you navigate the complex and evolving landscape of AI.?

AI is one of the most astonishing and elaborate efforts of human ingenuity and has only touched a small part of its capabilities. The AI applications that we witness today are just a sample of the immense possibilities that lie ahead. However, many people are apprehensive of AI overpowering the world or assume that we have already achieved the summit of AI innovation.?

AI's fast growth and impressive abilities have made people anxious about the certainty and closeness of an AI takeover, as depicted in so much science fiction. Also, the transformation caused by AI in different sectors has made business leaders and the public in general believe that we are near to accomplishing the peak of AI research and exhausting AI's potential. Artificial Intelligence (AI) has become a buzzword in various industries, promising to revolutionize our lives. However, to grasp the true potential of AI and the challenges ahead, it is crucial to understand the types of AI that exist now and those that are feasible. This book aims to shed light on the current capabilities of AI and the long journey that lies ahead in AI research.?So, let's get started :-)

Types of AI:?

1. Narrow AI: Also known as Weak AI, this type of AI is designed to perform specific tasks with exceptional accuracy. Examples include voice recognition systems, image and speech recognition, and recommendation algorithms. Narrow AI is the prevalent form of AI today, with applications ranging from virtual personal assistants to self-driving cars.?

2. Generative AI: Generative AI refers to a higher level of intelligence that mimics human cognitive abilities. This type of AI consists of a limited basis of knowledge and is successful where it is used to augment the consumer of it rather than replace the user. It can help with certain research tasks and document creation with human oversight and validation. Generative AI has truly accelerated this past year (2023) and is expected to continue improving for years to come with proper use and governance by both the developers and consumers of it.

Generative Artificial Intelligence represents an advanced form of intelligence that emulates human cognitive abilities. This groundbreaking technology possesses a limited knowledge base and exhibits remarkable success when employed to enhance the capabilities of users rather than entirely replacing them. By assisting with specific research tasks and document creation, generative AI can greatly benefit users through the inclusion of human oversight and validation. The field of generative AI has experienced significant advancements over the past year (2023), and its continuous improvement is anticipated in the years to come, contingent upon the implementation of appropriate measures and regulations by both developers and consumers. With its potential to revolutionize various domains, generative AI holds immense promise for the future of artificial intelligence.

3. Superintelligent or General AI: This hypothetical type of AI surpasses human intelligence in almost every aspect. Superintelligent AI possesses the ability to outperform humans in virtually any cognitive task and may even possess self-awareness and consciousness. While this type of AI remains a subject of speculation and debate, it represents the ultimate goal of AI research.

4. Self Aware AI: Self Aware AI: Self-awareness is the recognition of one's own personality or individuality. It is an experience of one's own personality or individuality in philosophy of self. (As defined in Meriam Webster Dictionary).

Feasible AI Types:?

While the concept of Superintelligent AI and Self-Aware AI holds promise for the future, achieving these levels of intelligence is still a distant reality. However, narrow AI continues to advance rapidly and has already demonstrated its feasibility:?

1. Machine Learning (ML): Machine learning algorithms allow systems to learn from data and improve performance without explicit programming. ML is currently powering various applications, such as spam filters, fraud detection systems, and recommendation engines. Deep learning, a subset of ML, has enabled significant breakthroughs in image recognition and natural language processing.?

2. Natural Language Processing (NLP): NLP focuses on enabling computers to understand and process human language. This technology has applications in virtual assistants, chatbots, and sentiment analysis, enabling machines to interact and communicate more effectively with humans.?

3. Computer Vision: Computer vision enables machines to interpret and understand visual information, such as images and videos. This technology finds applications in facial recognition, object detection, and autonomous vehicles, improving safety and efficiency in various domains.?

The Long Journey Ahead:?

While narrow AI has made remarkable strides, there are several challenges and opportunities standing in the way of achieving more advanced forms of AI:?

1. Ethical and Societal Implications: As AI becomes more integrated into our daily lives, ethical considerations surrounding privacy, bias, and job displacement become paramount. Researchers must address these concerns to ensure that AI serves the greater good.?

2. Transparency and Explain ability: The interpretability of AI systems is crucial for building trust and understanding their decision-making processes. Ensuring transparency and explaining ability in AI algorithms is a pressing challenge for researchers.?

3. Data Accessibility and Quality: The quality and availability of data play a crucial role in the performance of AI algorithms. Researchers must find ways to enhance data accessibility and ensure the fairness and accuracy of the data used.?

Summary:?

Understanding the existing types of AI and their feasibility provides us with a clearer picture of the current capabilities and the path that lies ahead in AI research. While narrow AI has already transformed various industries, achieving general AI and superintelligent AI remains a significant challenge. Researchers must tackle ethical concerns, enhance transparency and interpretability, and improve data accessibility to unlock the full potential of AI. The journey towards advanced AI is long, but the possibilities it holds are truly remarkable.?

To understand the current state and future potential of AI, we need to know the different classifications of AI that exist and how they compare to the ones that we have now.?

AI is a captivating and complex field of study that has many applications and challenges. AI can help us solve problems, automate tasks, enhance creativity, and improve our lives. However, AI also raises ethical, social, and technical issues that we need to be aware of and address. AI is not a uniform entity that can be easily defined or understood. AI is a diverse and evolving discipline that encompasses different types and levels of intelligence.?

AI can be classified according to the degree of similarity to human intelligence. There are three main categories of AI: narrow artificial intelligence (IAE), artificial general / generative intelligence (AGI), and superintelligent artificial intelligence (IAS). IAE refers to AI systems that can perform specific tasks or domains better than humans, such as playing chess, recognizing faces, or driving cars. IAG refers to AI systems that can match or exceed human intelligence in a wide range of domains and tasks, such as reasoning, learning, planning, creativity, and common sense.

IAS refers to AI systems that can exceed human intelligence in every possible domain and task, and potentially have goals and values that are incomprehensible or hostile to humans.?? Currently, we only have real-world examples of IAE. Most of the AI applications we use or encounter today are based on IAE, such as search engines, voice assistants, recommendation systems, spam filters, self-driving cars, etc. These systems are impressive and useful, but they are also limited and specialized. They can't perform tasks outside of their domain or deal with situations they're not programmed to do. They also lack the general cognitive abilities that humans have, such as understanding context, emotions, humor, sarcasm, etc.

IAG and IAS are hypothetical and speculative concepts that have yet to be achieved or demonstrated. Some researchers believe that we are close to creating IAG or even IAS in the near future, while others argue that we are far from reaching such levels of intelligence or that they are either impossible, undesirable, or both. The debate about the feasibility and desirability of IAG and IAS is ongoing and controversial. There are many technical, ethical, and philosophical challenges and risks involved in creating and interacting with such forms of intelligence.?

AI systems can vary in how closely they resemble human intelligence. We can group them into three main types: narrow artificial intelligence (IAE), artificial general intelligence (AGI), and superintelligent artificial intelligence (IAS). IAE is the type of AI that can do specific tasks or domains better than humans, such as playing games, identifying faces, or driving vehicles. AGI is the type of AI that can match or surpass human intelligence in a broad range of domains and tasks, such as thinking, learning, planning, creating, and understanding. IAS is the type of AI that can outsmart human intelligence in every possible domain and task, and potentially have goals and values that are different or harmful to humans.? At present, we only have real examples of IAE. Most of the AI applications we use or encounter today are based on IAE, such as web search, voice assistants, recommendation systems, spam filters, self-driving cars, etc. These applications are impressive and useful, but they are also narrow and specialized. They can't do tasks outside of their domain or handle situations they're not programmed to do. They also lack the general cognitive skills that humans have, such as understanding context, emotions, humor, sarcasm, etc.?

AGI and IAS are theoretical and speculative ideas that have not been achieved or proven yet. Some researchers think that we are close to creating AGI or even IAS in the near future, while others say that we are far from reaching such levels of intelligence or that they are impossible or undesirable. The debate about the possibility and desirability of AGI and IAS is ongoing and controversial. There are many technical, ethical, and philosophical challenges and risks involved in creating and interacting with such forms of intelligence.?

The Bad:?

As organizations step into the realm of artificial intelligence (AI), they are confronted with various challenges, one of the most prevalent being hidden costs. These costs, which arise during the implementation and maintenance of AI systems, can significantly impact the organization's budget, and hinder its progress. In Article two of this series, we delve deeper into this issue, highlighting the importance of understanding and addressing hidden costs.

Moreover, we shed light on the flaws inherent in AI, as evident from the very name itself - "Artificial Intelligence." This prompts us to question the idea of entrusting our lives, our future, and our children's future to entities associated with terms like "fake," "false," and "imitation." In this context, we also discuss the need for validating AI algorithms, which are currently designed by human beings. This leads us to ponder upon the general intelligence of the human race as a whole today.?

Hidden Costs of AI Adoption:?

AI adoption comes with a multitude of hidden costs that organizations must consider. These costs can arise in various forms, including infrastructure investments, software development, data acquisition and processing, cybersecurity measures, training, and ongoing maintenance. The complexity and scale of AI systems often translate into higher implementation costs than initially estimated. Moreover, organizations must allocate resources for continuous monitoring, updates, and troubleshooting, which can significantly strain their budgets. Understanding and accounting for these hidden costs is crucial to ensure the financial viability of AI initiatives.?

Flaws in AI:?

While AI showcases impressive capabilities, it is not without its flaws. As mentioned earlier, the term "Artificial Intelligence" itself reflects elements of fakeness, imitation, and non-natural attributes. AI systems, despite their capabilities, lack the inherent human qualities of genuine intellect, true brainpower, and deep astuteness. They rely on algorithms and machine learning to mimic human intelligence, which inherently limits their capabilities. AI algorithms are designed by humans, making them susceptible to human biases, errors, and limitations. This raises concerns regarding the reliability, accuracy, and fairness of AI systems, especially in critical decision-making processes.?

Validation of AI Algorithms:?

Ensuring the accuracy and truthfulness of AI algorithms is a pressing challenge. Given that these algorithms are created by human beings, they inherit human biases, conscious or unconscious. Validation becomes crucial to identify and rectify these biases, ensuring that AI systems are fair and unbiased. The validation process involves rigorous testing, benchmarking, and comparison against ground truth data. It requires collaboration between domain experts, data scientists, and ethicists to scrutinize, refine, and validate AI algorithms. By doing so, organizations can mitigate potential risks associated with inaccurate or biased AI outputs.?

The General Intelligence of the Human Race:?

In contemplating the general intelligence of the human race today, diverse perspectives emerge. Humans possess a remarkable capacity for creativity, critical thinking, intuition, and emotional intelligence, which remain difficult to for us to completely understand ourselves and are thus extremely challenging for us to replicate in AI systems. However, our collective intelligence is not without limitations. We are susceptible to cognitive biases, limited information processing, and subjective interpretations. The flaws and biases that human authors of AI algorithms may introduce reflect these inherent imperfections. Recognizing and continuously striving to improve our own intelligence is imperative in designing and deploying AI systems that augment human capabilities rather than replacing them.?

Summary:?

As we embark on the adoption of AI, we must confront the challenge of hidden costs and the recognition of AI's flaws. Validating AI algorithms is crucial to ensure accuracy, fairness, and reliability. While AI has immense potential, it is important for us to acknowledge and embrace the limitations of artificial intelligence. Furthermore, reflecting on our own intelligence allows us to approach AI adoption with a more discerning mindset. By understanding the hidden costs, addressing the flaws, and validating AI algorithms, we can harness the benefits of AI while safeguarding against its potential pitfalls.?

The Ugly:?

Hidden costs are not only a challenge, but can also be a threat, as organizations begin their adoption of AI. This is because AI will be increasingly integrated into the existing technologies of the business, and will autonomously deploy, consume, and even direct other AI applications. This will expose the organization to various types of hidden costs, such as:?

- Embedded: These are the costs that are not obvious or predictable, such as unexpected failures, errors, or biases in the AI systems.?

- Upfront: These are the costs that are required to acquire, install, and configure the AI systems, such as hardware, software, and licenses.?

- Ongoing: These are the costs that are incurred to maintain, update, and monitor the AI systems, such as cloud services, security, and quality assurance.?

- Staffing and Training: These are the costs that are related to hiring, training, and retaining the human resources that are needed to work with the AI systems, such as data scientists, engineers, and analysts.?

- Data-Related: These are the costs that are associated with collecting, storing, processing, and analyzing the data that is used by the AI systems, such as data sources, platforms, and tools.?

- Ethical and Legal: These are the costs that are derived from the ethical and legal implications of using AI systems, such as privacy, fairness, and accountability.?

Note: In recent times, there are reports of a disturbing trend involving the deliberate misuse of Generative AI for social media and news purposes. This misuse involves the intentional use of AI-generated and/or edited photographs with the purpose of confusing and misleading the public.

These acts not only compromise the integrity of information but also pose a significant threat to society. The intentional deception of the public through the manipulation of photographs undermines the trust we place in media outlets and social platforms. It erodes the foundation of truth upon which our society stands, and it is imperative that we address this issue with the utmost urgency.

- Legal Implications: These are the costs that are caused by the legal risks and liabilities of using AI systems, such as lawsuits, fines, and regulations.?

- Government Regulations, Laws, Conflicts: These are the costs that are influenced by the government policies and actions regarding AI systems, such as taxes, subsidies, and bans.?

- Pitfalls: These are the costs that are incurred by the potential negative outcomes of using AI systems, such as loss of competitive advantage, customer dissatisfaction and reputational damage.?

Note: In next week's article we will discuss, in more depth, the costs and impact AI may and most likely will have.