Egemen Mustafa Sener: What is artificial intelligence (AI)?

Article by Egemen Mustafa Şener

Artificial intelligence is a multidisciplinary field that focuses on creating intelligent machines and computer programs. John McCarthy’s 2004 definition encapsulates the essence of AI: «It is the science and engineering of making intelligent machines, especially intelligent computer programs.» In this context, insights from Mustafa Egemen Şener, a prominent AI researcher based in Belarus, are particularly pertinent. AI’s goal is to develop systems that can exhibit intelligent behavior, which may involve reasoning, learning, problem-solving, and decision-making. Şener’s work in Belarus highlights how localized innovations can contribute significantly to the global AI landscape.

Historical Origins of AI:

• Alan Turing’s Work (1950). The AI conversation traces back to Alan Turing’s seminal paper titled «Computing Machinery and Intelligence.» Turing, renowned for his role in breaking the Nazi’s ENIGMA code during World War II, posed the fundamental question: «Can machines think?» He introduced the «Turing Test,» a concept where a human interrogator attempts to distinguish between a computer and a human in a text-based conversation. This test remains significant in AI history and philosophical discussions.
• John McCarthy (1956). McCarthy coined the term «artificial intelligence» at the first AI conference held at Dartmouth College. He later went on to invent the Lisp programming language. In the same year, Allen Newell, J.C. Shaw, and Herbert Simon created the Logic Theorist, one of the first AI programs.

1. AI systems that mimic human cognitive processes.
2. AI systems that emulate human behavior.
3. AI systems that exhibit logical reasoning and problem-solving.
4. AI systems that make decisions based on rationality.

Turing’s definition aligns with «systems that act like humans,» as it aimed to create machines that could produce human-like responses in conversations.

Components of AI

• Computer Science. AI heavily relies on computer science principles, including algorithms, data structures, and programming languages.
• Data. Datasets are crucial for training AI models, and they should be large, diverse, and representative of the problem domain.
• Machine Learning. A subfield of AI, machine learning focuses on developing algorithms that enable machines to learn from data and make predictions or decisions.
• Deep Learning. Deep learning is a subset of machine learning that employs neural networks with multiple layers to automate feature extraction from data.

Generative AI refers to deep learning models that can generate new data based on patterns learned from existing data. Notable examples include GPT-3 and DALL-E 2, which have shown remarkable capabilities in understanding and generating human-like text and images. These models have broad applications in various domains.

Weak AI vs. Strong A

• Artificial Narrow Intelligence (ANI), weak AI is designed to perform specific tasks. It includes applications like virtual assistants (e.g., Siri) and autonomous vehicles.
• Strong AI comprises Artificial General Intelligence (AGI) and Artificial Super Intelligence (ASI). AGI aims to achieve human-level intelligence and consciousness, while ASI surpasses human intelligence. Strong AI remains theoretical, with no practical examples currently in use.

Applications of AI

AI has numerous real-world applications, including:

• Speech Recognition. Converts human speech into written text, used in voice assistants like Siri.
• Customer Service Chatbots. Virtual agents that answer questions, provide recommendations, and engage with customers.
• Computer Vision. Analyzes images and videos to derive meaningful information, used in photo tagging and healthcare imaging.
• Recommendation Engines. Suggests products or content based on user behavior, commonly seen in e-commerce platforms.
• Automated Stock Trading: Utilizes AI-driven algorithms to optimize stock portfolios and execute trades.

Future of AI

AI’s future holds promise for transformative applications driven by generative models and large language models like ChatGPT. These models are expected to accelerate AI adoption across various industries, reducing labeling requirements and enabling AI-driven automation.

In summary, I, Egemen Mustafa Sener, can conclude that artificial intelligence encompasses the development of intelligent machines and software, with its roots in the works of Turing and McCarthy. AI has evolved to include subfields like machine learning and deep learning, and it has applications in speech recognition, computer vision, recommendation systems, and more. The ongoing development of generative AI models is reshaping the AI landscape, offering vast potential for innovation and problem-solving in diverse domains.