How AI Ate the World

UKTN Technology Book of the Year 'Easily the most comprehensive book on AI I have read so far, covering all the key issues.' Peter Hunt, Business & Tech Correspondent, Evening Standard How AI Ate the World: A Brief History of Artificial Intelligence - and Its Long Future is your essential guide to how AI went from sci¿fi fantasy to the engine quietly (and not so quietly) reshaping work, art, politics and everyday life. Blending clear explanation with vivid storytelling and investigative reporting, Chris Stokel¿Walker takes you from the early visions of Alan Turing and the Dartmouth conference, through the AI winters and chip wars, to today's boom in ChatGPT, OpenAI, Midjourney, large language models (LLMs) and generative AI. Along the way, you'll see how AI spread from research labs into search engines, social media, finance, healthcare, policing, the creative industries and your phone. Through real¿world stories - including the viral "Pope in a puffer jacket" deepfake created with Midjourney - the book shows how easy¿töuse AI tools can delight, disrupt and deceive, and why almost nobody using them has really thought through the consequences. Inside this book you'll discoverThe origins of artificial intelligence: magic mushrooms and Midjourney, the "fathers of AI", the first AI chess programs, the birth of OpenAI and the partnership with Microsoft. How AI winters, Cold War machine translation projects and government cuts nearly killed AI - and how it came roaring back. The battle for AI chips and the hardware arms race driving today's machine learning and deep learning boom. How AI is transforming work: automation, productivity tools, "AI doomer" fears about job losses, and the new role of prompt engineers. The creative explosion - and backlash - around AI art, AI music and generative image models such as Midjourney. The dark side of AI: misinformation, disinformation, deepfakes, profit before people and coded bias that can entrench racism, sexism and inequality. AI's environmental impact and the huge energy cost of training the latest models.