Lesson 1

The History and Evolution of Artificial Intelligence

Humans have imagined creating intelligent machines for centuries, a concept

reflected in history, folklore, and early computing devices. This ambition took a
significant step forward with the introduction of OpenAI’s Generative Pretrained
Transformer (GPT), which demonstrated the potential of AI to process and generate
human-like text.
One of the most notable advancements was GPT-3, a model with 175 billion
parameters, allowing it to perform a wide range of natural language tasks with minimal
fine-tuning. Its ability to learn from just a few examples made it highly versatile, leading
to practical applications in chatbots, virtual assistants, and other AI-driven tools.
Today, AI is deeply integrated into various aspects of daily life, from social media
algorithms to workplace automation. As future professionals, understanding the
evolution of AI will help us navigate its growing role in society. Let’s take a look at some
key milestones in AI development.

Pre-1936: Early Foundations of AI

17th–19th Century: Philosophical and Mechanical Precursors
 René Descartes (1637): Proposed that human thought could be broken down
into mechanical processes, laying the groundwork for computational theories and
artificial intelligence.
 Gottfried Wilhelm Leibniz (1700s): Developed binary logic, a foundational
principle of modern computing and AI decision-making.
 Charles Babbage (1830s): Designed the Analytical Engine, a theoretical
mechanical computer capable of programmable calculations, foreshadowing
modern computing.
 Ada Lovelace (1843): Wrote the first algorithm for Babbage’s Analytical Engine
and theorized that machines could manipulate symbols beyond numerical
calculations, predicting AI’s potential.
 George Boole (1854): Introduced Boolean algebra, a critical foundation for logic-
based computing and AI decision processes.
Early 20th Century: Mechanical and Logical Developments
 Bertrand Russell & Alfred North Whitehead (1910-1913): Published Principia
Mathematica, formalizing logic and influencing AI’s computational theory.
 Leonardo Torres y Quevedo (1913): Built a chess-playing automaton capable
of completing simple endgames, demonstrating early machine intelligence.
 Norbert Wiener (1920s-1940s): Developed cybernetics, emphasizing feedback
loops in machines, foundational to robotics and AI control systems.
1936: Alan Turing and the Concept of Computation
 Alan Turing: Published "On Computable Numbers, with an Application to the
Entscheidungsproblem," introducing the concept of the Turing machine, an
abstract mathematical model defining computability and influencing the
development of AI.
 1939: John Vincent Atanasoff and Clifford Berry develop the first electronic digital
computer prototype, laying groundwork for machine intelligence.
1950: The Turing Test and Machine Intelligence
 Turing proposes the Imitation Game: Later known as the Turing Test, it
assesses whether a machine can exhibit intelligence indistinguishable from a
human.
 Claude Shannon (1950): Published "Programming a Computer for Playing
Chess," pioneering machine strategy.
1950–2000: The Birth and Early Development of AI
1956: The Dartmouth Conference - The Birth of AI
 John McCarthy, Marvin Minsky, Nathaniel Rochester, and Claude Shannon:
Organize the Dartmouth Summer Research Project on Artificial Intelligence,
officially launching AI as a field and coining the term "artificial intelligence."
1966: ELIZA - The First Chatbot
 Joseph Weizenbaum: Develops ELIZA, a natural language processing program
capable of mimicking human conversation, pioneering early human-computer
interaction.
1979: Shakey the Robot - Early AI Robotics
 SRI International: Develops Shakey, one of the first robots integrating computer
vision, natural language understanding, and decision-making capabilities.
1980s: The Rise of Expert Systems
 MYCIN and XCON: AI sees commercial adoption with rule-based expert systems
for medical diagnosis and computer configuration, demonstrating AI’s industrial
applications.
1997: IBM Deep Blue Defeats Chess Champion
 IBM’s Deep Blue: Becomes the first AI to defeat world chess champion Garry
Kasparov, marking a milestone in AI’s decision-making and strategic reasoning.
2000–2020: Advances in Machine Learning and Deep Learning
2000: Kismet - Social Robotics
 Cynthia Breazeal at MIT: Develops Kismet, a robot designed to interact with
humans using emotional and social cues, pioneering AI-driven social interactions.
2006: Geoffrey Hinton and Deep Learning Revolution
 Geoffrey Hinton publishes "Learning Multiple Layers of Representation":
Establishes the groundwork for deep learning and multilayer neural networks,
leading to modern AI breakthroughs.
2007: ImageNet - A Revolution in Computer Vision
 Fei-Fei Li and her team at Princeton: Create ImageNet, a massive database of
labeled images that fuels advancements in deep learning for object recognition.
2009: GPUs Revolutionize Deep Learning
 Rajat Raina, Anand Madhavan, and Andrew Ng: Demonstrate that Graphics
Processing Units (GPUs) dramatically enhance deep learning model training,
making modern AI applications feasible.
2011: IBM Watson and Siri - AI Enters Daily Life
 IBM Watson: Defeats Jeopardy! champions Ken Jennings and Brad Rutter,
showcasing AI’s ability to understand and process natural language.
 Apple introduces Siri: The first AI-powered virtual assistant integrated into
smartphones, enhancing user accessibility.
2012: AlexNet and the Deep Learning Breakthrough
  Geoffrey Hinton’s team at the University of Toronto: Designs AlexNet,
proving deep learning’s superiority in computer vision with a 16% error rate on
ImageNet.

2016: AlphaGo’s Historic Win

 Google DeepMind: AlphaGo defeats world Go champion Lee Sedol,
demonstrating AI’s ability to master complex strategic reasoning beyond
traditional computing.
2017: Facebook’s Chatbot Experiment
 Facebook AI researchers: Train chatbots that unexpectedly develop their own
shorthand communication, highlighting AI’s autonomous learning potential.
2020: GPT-3 and AlphaFold 2
 OpenAI releases GPT-3: A 175-billion parameter language model demonstrating
human-like text generation and natural language processing capabilities.
 DeepMind’s AlphaFold 2: Solves the protein folding problem, revolutionizing
biology and accelerating medical research.
2021–Present: The Age of Generative AI
2021: MUM and Tesla’s Full Self-Driving Beta
 Google unveils MUM: A multimodal AI that processes and generates text,
images, and videos, enhancing search capabilities.
 Tesla launches Full Self-Driving (FSD) Beta: Advancing AI-driven autonomous
vehicle navigation.
2021–2023: AI-Generated Creativity
 OpenAI releases DALL-E and its successors: AI models capable of generating
highly detailed images from text prompts, revolutionizing digital art and media.
2024: Advances in AI Models and Applications
 February: Google launches Gemini 1.5, OpenAI introduces Sora (text-to-video
generation), and StabilityAI releases Stable Diffusion 3.
 May: Google DeepMind extends AlphaFold for genetic disease identification, and
IBM introduces Granite AI models for scalable business applications.
  June: Apple integrates AI into Siri with ChatGPT, enhancing voice assistant
capabilities.
 September: NotebookLM launches DeepDive, a multimodal AI for podcast-style
content creation.

Future of AI
The evolution of AI continues toward:
 Smaller, more efficient models with high performance at lower computational
costs.
 Agentic AI, where multiple AI systems collaborate to complete tasks efficiently.
 Multimodal AI, capable of handling text, images, audio, and video within a single
platform.
 Autonomous vehicles, making self-driving transportation a reality.
 AI assistants that seamlessly integrate into personal and professional life,
enhancing productivity and decision-making.

References:
Mucci, Tim, History of AI, https://www.ibm.com/think/topics/history-of-artificial-
intelligence