The Pocket Guide to Artificial Intelligence
summary
Maps AI into three concentric goals: what’s possible in research, what’s practical for startups, and what’s ubiquitous at scale. Gives concrete data points: email spam filters affect ~4 billion users, recommender systems drive >50% of Netflix views, ImageNet error falls from ~30% to ~2% (human top‑5 ~5%). Notes compute‑heavy game wins (AlphaGo used millions in hardware; Dota 2 needed massive self‑play simulation), level‑2 autonomy is commercial, level‑3 imminent and level‑5 far off, and current systems lack cause‑and‑effect reasoning.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 0:53Artificial Intelligence Opinions
- 2:14The Circle of Human Knowledge
- 4:37The Circle of Artificial Intelligence
- 5:50The Inner Circle
- 8:00Use Cases
- 11:40Speech Recognition
- 12:53Games
- 18:40Machine Translation
- 21:45Image Generation
- 23:15Autonomy
- 25:25General Intelligence
- 27:40Cause and Effect
- 33:05Questions
related talks
42mDawn of the planet of artificial intelligence
14mTwo Keys to AI's Future
01h 01mIntroduction to Practical AI
13mArtificial General Intelligence: Why aren't we there yet?
29mScaling AI - Creating value with the application of AI
47mThere is no such thing as Artificial Intelligence
52mThe Journey of AI in Games: A Technical and Philosophical overview
01h 51mAI and Social Entrepreneurship