Artificial Intelligence – Comprehensive Overview
Part One
Artificial Intelligence Patterns and Their Role in Digital Transformation
Scientists typically classify artificial intelligence technology into three main patterns based on their distinguishing capabilities and level of simulation of human intelligence. These patterns are: Narrow AI, which has narrow capabilities, General AI that equals human capabilities in some functions, and Super AI, which is expected to surpass human capability in the future. These three patterns are categorized based on their ability to simulate human characteristics, the technology used for this simulation, and their practical applications.
Narrow AI (ANI) is the only pattern of artificial intelligence that has been successfully achieved so far. This type of artificial intelligence is goal-directed, designed to perform individual tasks with high intelligence in executing specific programmed tasks. While these machines may appear intelligent, they operate within a narrow set of constraints and conditions, which is why this pattern is referred to as Weak AI. It does not replicate human intelligence absolutely but mimics human behavior based on a narrow range of criteria such as Face recognition, Voice recognition used in voice assistants and chatbots, Natural Language processing, and applied in Autonomous Cars. Narrow AI systems are currently used in medicine for the precise diagnosis of cancers and other diseases. Currently, humans have only achieved narrow artificial intelligence, with ongoing efforts to develop machine learning capabilities and continuous research to achieve General AI.
The second type is General AI (AGI), also known as Strong AI or Deep AI, mimics human intelligence and behaviors, with the ability to learn and apply intelligence to solve any problem the system may face. General AI can think, understand, and act in a way indistinguishable from humans in any given situation. However, General AI is still an emerging field because the human brain is the model for creating general intelligence, and it does not seem likely to be achieved relatively soon due to the lack of comprehensive knowledge of the functions of the human brain.
One of the main approaches to achieve General AI is called “Whole Brain Emulation,” where the brain’s memory and mental state are transferred to a computer device. The computer’s structure resembles the brain’s structure as it operates through a system of neural cells called neural networks, allowing technology to learn and form intelligent neural pathways. Quantum Computing, which relies on quantum mechanics to process data faster and with higher capacities than conventional computers, is also being used to push the boundaries of General AI.
Various predictions exist regarding the expected time to achieve General AI. In 2017, opinions from over 350 experts in machine learning and neuroscience were surveyed, with about 50% of them believing it would happen before 2060. Louis Rosenberg, CEO of Unanimous AI, predicts it could occur around 2030, while the former director of the AI lab at the Massachusetts Institute of Technology foresees advancements in this field.
The third pattern is super artificial intelligence (ASI), which is the virtual artificial intelligence that presents the path to a future that not only mimics or understands human intelligence and behavior, but is intelligence in which machines become self-aware and surpass human intelligence and ability. Furthermore, artificial intelligence will evolve to be very close to human emotions, experiences, and capacities, so it not only understands them but also evokes its own emotions, needs, beliefs, and desires. Additionally, super intelligence will be able to perform exceptionally and theoretically excel in everything we do in fields such as mathematics, science, sports, art, medicine, hobbies, and emotional relationships. It will have a larger memory, faster data processing and analysis capabilities, thus the decision-making and problem-solving abilities of superintelligent beings will far exceed those of humans.
Given the diversity of fields and types of artificial intelligence and its potential applications in various areas, it constitutes a fundamental pillar of digital transformation and serves as a real cornerstone for other digital transformations such as the Internet of Things, cloud computing, big data, and cyber security, through the use of machine learning in analyzing and processing large datasets to extract important and impactful information for decision-making.
It is also utilized in the Internet of Things to analyze data from sensors and make appropriate decisions, as mentioned by Gartner in its report in November 2018 that ” artificial intelligence will be applied to a wide range of Internet of Things data, including video, still images, speech, network traffic movement, and sensor data.”
Artificial intelligence is also closely linked to Cyber Security in analyzing electronic piracy attacks and threats and combating malware that electronic systems may face, acting as a first line of defense by intervening to protect security, privacy, and detect harmful viruses and malicious software.
Furthermore, artificial intelligence is closely linked to cloud computing, where artificial intelligence tools are integrated into the IT infrastructure, enabling both private and public clouds to utilize these artificial intelligence tools for monitoring, management, and even self-healing in case of a problem. According to Statista, the global market value of artificial intelligence will exceed $89 billion annually by 2025. A significant portion of this value will be realized when artificial intelligence supports cloud computing; thus, cloud computing acts as an engine to enhance the scope and impact that artificial intelligence can have on the market.
By Engineer: Belal Khalid Al-Hafnawi
Specializing in the telecommunications, information technology, and digital transformation sector
