Further reading: From problems to purpose
Over the past summer I continued my reading streak, devouring several more books on the subject of AI. I was interested in gaining a deeper understanding of the technology and its impacts, and I needed more perspectives to round things out.
The next book I picked up after finishing Ethan Mollick’s Co-Intelligence was The Coming Wave by Mustafa Suleyman, a co-founder of Google DeepMind, and co-writer Michael Bhaskar. This book provides a poignant and profound look at the trajectory of AI from the perspective of someone who was on the team that shocked the world by developing AlphaGo, a computer program that in 2016 defeated the world champion of Go, a game that was until that point considered too difficult for computers to master. About halfway through, I described the book to my colleagues as the most terrifying thing I had ever read. Not only does it lay out a frightening potential future should AI be employed by nefarious forces, it also traces the parallel developments in synthetic biology and the risks of bio-terrorism. This book is not for the faint of heart. Of some consolation, the authors lay out a clear ten-point framework they claim can save us from self-destruction. Time will tell if their warnings are heeded and their advice followed.
As soon as I got through The Coming Wave I needed an antidote to my anxiety. I picked out not one, but two books: Mastering AI by the acclaimed journalist and Forbes AI editor Jeremy Khan, who has been covering AI since 2016, and The Worlds I See by Dr Fei-Fei Li, the renowned professor and founder of the Stanford Human-Centered AI Institute, who is often referred to as "the godmother of AI".
Mastering AI provides an unbiased and non-alarmist overview of the history and current state of AI. It does a great job of stringing together the events and developments that spanned the past 6 decades in a matter-of-fact way, weaving in the concerns and impacts on society without inducing panic. If you want to learn about AI, how we got here, and where things are heading, this is a good choice. It’s a fairly easy read too. The author asks good questions and aims to provide clear answers, as you’d expect from a seasoned journalist. I finished this book pretty quickly.
By this point I had read four books on AI all written by men. I was missing an important perspective. Fei-Fei Li’s book, The Worlds I See is as much an autobiography as a book about AI, and it filled the gap in my learning journey. It’s the story of a young girl from China whose parents sought to give her a better life away from authoritarian rule and sacrificed everything to give their only child, a girl with endless curiosity, the opportunity to become a scientist who would eventually spark the biggest wave of AI research and discovery ever through her passion project called ImageNet, a vast data set of categorized images used to train early neural networks. Her story, in other words, describes the history of modern AI.
What I learned from The Worlds I See was much deeper, and more profound in some ways than all the other books I had read. As a narrative that intertwines the author‘s personal challenges and insights with their academic and professional achievements, the story revolves around a central theme of loving support and care. Through all the struggles and tireless efforts to advance the science, Li reminds us that in the end none of it matters as much as the human connections we have with our partners, our family, our friends, and our community. Advancing science is important, but the goals should be to help people above all else. This eventually became the author’s purpose in life, and explains why she wrote this book.
This literary journey through AI has been both enlightening and transformative. Starting with the encouraging optimism of How to Speak Machine and the insightful guidance of Co-Intelligence, followed by the alarming predictions in The Coming Wave, the balanced journalism of Mastering AI, and finally the deeply personal and humanizing narrative of The Worlds I See, I've gained a multifaceted understanding of AI's past, present, and potential future. The big takeaways for me were John Maeda's prompts to be a positive influence on AI's direction, and Fei-Fei Li's emphasis on the importance of human connections and using technology to help people. Both authors urge those of us in positions of influence to use our privilege to steer AI's development and application with a human-centered perspective. Mustafa Suleyman warns us what could happen if we don't. Our survival as a species might just depend on it.