Co-Intelligence: Living and Working with AI by Ethan Mollick

Early Fascination with Thinking Machines

• 💡 Humanity has long been fascinated by machines that can think, dating back to the 1770s with the mechanical chess computer, the Mechanical Turk.
• 🎭 The Mechanical Turk, which toured for decades and beat notable figures, was ultimately a hoax with a human chess master hidden inside.

Foundations of Artificial Intelligence

• 🧠 In 1950, Claude Shannon developed Thesius, a mechanical mouse that could navigate a maze, demonstrating an early form of machine learning.
• 🔑 Alan Turing introduced the "imitation game," laying theoretical groundwork for machines to mimic human functionality.
• 🗓️ The term "Artificial Intelligence" was coined in 1956 by John McCarthy of MIT.

Cycles of Hype and Disillusionment

• 📈 AI has historically been plagued by hype cycles, with early rapid progress in logic problems and checkers leading to unfulfilled promises.
• ❄️ These unfulfilled expectations often resulted in "AI winters," periods where progress stalled and funding for AI research dried up.

The Latest AI Boom: Machine Learning

• 🚀 The most recent AI boom, beginning in the 2010s, centered on machine learning for data analysis and prediction.
• 📊 This era heavily relied on supervised learning, which necessitates large amounts of labeled data for training AI systems.
• 🏢 Early applications were primarily for large organizations with extensive data, used for tasks like optimizing shipping or content recommendations.

Impact of Predictive AI

• ✅ Predictive AI, such as forecasting hotel demand, allowed businesses to move from guessing games to more accurate predictions based on vast data inputs.
• ⚙️ AI algorithms shifted the focus from being "correct on average" to being "right for each specific instance," revolutionizing back-office functions and supply chains.