Complete AI Mastery Course

{"contentFormat":"slides.v1","completion":{"requireAllSlides":true,"requireQuiz":true},"slides":[{"kind":"title","eyebrow":"Module 1: AI Foundations and Mindset","title":"The History and Evolution of AI: From Turing to Transformers","body":"Trace eight decades of ideas, failures, and breakthroughs that led to today's generative AI. Understand why the Transformer (2017) changed everything — and how this context sharpens your judgment about AI claims.","outcomes":["Identify the major eras of AI development","Explain the shift from symbolic AI to machine learning","Describe how the Transformer architecture enabled modern LLMs","Recognize global contributions to AI research","Apply verification habits when using AI tools"],"narration":"Welcome. Before you shape the future with AI, it helps to understand the road that brought us here. The tools you use today — ChatGPT, Claude, Gemini — are products of roughly eight decades of research across the world. This lesson gives you a mental map for why AI behaves the way it does."},{"kind":"content","heading":"The Founding Era (1940s–1950s): The Question Is Posed","body":"Alan Turing, a British mathematician, published 'Computing Machinery and Intelligence' in 1950, asking 'Can machines think?' He proposed the Turing Test: if a human judge cannot reliably tell a machine's replies from a person's, the machine shows intelligent behaviour.\n\nThe term 'artificial intelligence' was coined for a 1956 workshop at Dartmouth College, widely treated as the field's official birth. Around the same time, early neural network ideas emerged from researchers like Warren McCulloch and Walter Pitts (US) and Norbert Wiener (cybernetics).","callout":{"variant":"insight","title":"Global Roots","text":"While early AI was Western-dominated, parallel work in cybernetics and logic flourished in the Soviet Union (e.g., Alexey Lyapunov) and Japan. Today, AI research is truly global: China's Baidu and India's AI initiatives (e.g., Bhashini language platform) are major contributors."},"narration":"Alan Turing posed the foundational question in 1950. The term 'artificial intelligence' was coined at a 1956 workshop. Early work also happened in the Soviet Union and Japan, foreshadowing today's global research landscape."},{"kind":"content","heading":"Symbolic AI, Expert Systems, and the First AI Winter","body":"Early researchers built symbolic AI (also called good old-fashioned AI): systems that reason with hand-written rules and logic, like IF fever AND rash THEN suspect measles.\n\nExpert systems in the 1980s encoded specialist knowledge as thousands of such rules and were sold commercially. However, rules are brittle — you cannot hand-write a rule for every real-world situation. When funding and results disappointed, investment dried up in periods now called AI winters.","table":{"headers":["Era","Approach","Limitation"],"rows":[["1960s–1970s","Symbolic AI / Logic","Brittle, doesn't scale"],["1980s","Expert Systems","Knowledge acquisition bottleneck"],["1990s","Machine Learning","Needs large data, compute"]]},"narration":"Symbolic AI used hand-written rules, but these systems were brittle. Expert systems in the 1980s failed to scale, leading to AI winters. The shift to machine learning in the 1990s let systems learn patterns from data instead."},{"kind":"content","heading":"The Deep-Learning Breakthrough (2012 Onward)","body":"Neural networks are systems loosely inspired by brain cells: layers of simple units that pass signals forward, with weights tuned during training. Stacking many layers is called deep learning.\n\nIn 2012, a neural network called AlexNet dramatically won the ImageNet image-recognition contest, proving deep learning could beat older methods when given enough data and powerful GPUs. This moment ignited the modern AI boom.\n\nKey milestone: IBM's Deep Blue beat world chess champion Garry Kasparov in 1997, but relied on brute-force search. Deep learning was more flexible.","callout":{"variant":"note","title":"Global Contributions","text":"Deep learning advances came from researchers worldwide: Yann LeCun (France/US), Geoffrey Hinton (Canada/UK), Yoshua Bengio (Canada), and Fei-Fei Li (China/US) who created ImageNet. China's Baidu deployed deep learning for speech recognition early, and India's AI research hubs grew rapidly."},"narration":"Deep learning, powered by neural networks and GPUs, took off after AlexNet's 2012 win. Researchers from many countries contributed, including China and India. This era set the stage for generative AI."},{"kind":"content","heading":"The Transformer and the Generative Era (2017–Today)","body":"In 2017, Google researchers published 'Attention Is All You Need', introducing the Transformer architecture. Its key idea, attention, lets a model weigh how much each word relates to every other word in a sentence, all at once — like reading a whole page instead of through a keyhole.\n\nTransformers scale beautifully with more data and computing power. This gave us Large Language Models (LLMs) — models trained to predict the next word, which produces remarkably capable writing, reasoning, and coding.\n\nPublic milestones: ChatGPT (late 2022), GPT-4o and o1 (2024–2026), Claude, Gemini, and open-weight models like Meta's Llama. China's Baidu launched ERNIE Bot, and India's CoRover.ai built BharatGPT.","cards":[{"title":"Attention Mechanism","text":"Weighs relationships between all words simultaneously. Example: in 'The dog chased its tail,' attention links 'its' to 'dog'."},{"title":"Scaling Law","text":"More data + more compute = better performance. This drove the rapid improvement of LLMs."},{"title":"Global Impact","text":"Transformer-based translation now serves languages like Vietnamese and Finnish. China's Baidu and India's AI initiatives are key players."}],"narration":"The 2017 Transformer paper introduced attention, enabling models to understand context like never before. This architecture powers all major LLMs today, from GPT-4o to Claude, with significant contributions from Chinese and Indian AI research."},{"kind":"content","heading":"Real-World Examples Across the Globe","body":"AI history is not just academic — it shapes practical tools used worldwide:\n\n- Healthcare, Africa: Deep-learning image models screen chest X-rays for tuberculosis in Kenya and Nigeria, extending scarce specialist expertise.\n- Translation, Asia and Europe: Transformer-based translation handles languages from Vietnamese to Finnish, powering cross-border e-commerce for small sellers.\n- Agriculture, Latin America: Farmers in Brazil use vision models on phone photos to spot crop disease early — a direct application of the machine-learning shift.\n- China: Baidu's ERNIE Bot and Alibaba's Tongyi Qianwen serve millions in Chinese.\n- India: The Bhashini platform uses AI to translate between dozens of Indian languages, supporting digital inclusion.","callout":{"variant":"exercise","title":"Try It Yourself","text":"Open any free chatbot and paste: 'Explain the difference between symbolic AI and machine learning to a beginner, using one analogy. Then explain what the 2017 Transformer paper changed. Keep it under 200 words.' Then verify the year of the Transformer paper using a second source. This habit is invaluable."},"narration":"AI is deployed globally: healthcare in Africa, translation in Asia, agriculture in Latin America, and large-scale language models in China and India. Practicing verification with AI tools is a key skill."},{"kind":"quiz","heading":"Check Your Understanding","questions":[{"question":"What was the key innovation of the 2017 Transformer paper?","options":["It introduced convolutional neural networks for image recognition.","It proposed the attention mechanism, allowing models to weigh relationships between all words simultaneously.","It demonstrated the first expert system for medical diagnosis.","It created the Turing Test."],"questionId":"cmrf73k2l0003pd27az21lukt"},{"question":"Which of the following is a real-world AI application mentioned in this lesson?","options":["Using symbolic AI to predict stock prices.","Deep-learning models screening chest X-rays for tuberculosis in Kenya.","Expert systems powering modern chatbots.","Turing Test used for hiring decisions."],"questionId":"cmrf73k2l0004pd27xq72hasu"},{"question":"Which country's AI initiatives include the Bhashini language platform?","options":["China","India","Brazil","Nigeria"],"questionId":"cmrf73k2l0005pd278zblxu89"}],"quizId":"qz_cmk5q70z10006g4eo4ns0ze2h"},{"kind":"summary","heading":"Key Takeaways","takeaways":["AI evolved from symbolic rules (brittle) to machine learning (data-driven) to deep learning (neural networks).","The 2017 Transformer architecture, using attention, enabled modern LLMs like GPT-4o and Claude.","AI research is global: contributions from China (Baidu, ERNIE Bot) and India (Bhashini, BharatGPT) are significant.","Always verify AI-generated facts using independent sources — a critical skill for responsible AI use.","Understanding AI history helps you evaluate new claims and see beyond the hype."],"narration":"To recap: AI has moved from symbolic rules to machine learning to deep learning. The Transformer revolutionized language AI. Global contributions from China and India are shaping the field. Always verify AI outputs. This context will make every later lesson easier to absorb."}]}