7 min read
AI in 2041: The Vision of Dr. Kai-Fu Lee
Sep 12, 2021
If you had to predict what AI will look like in 20 years, what would you envision?
Terminator-style killer bots or conversational agents that are your friends and mentors on the internet?
Does 20 years of AI progress scare you, or excite you?
I can’t think of a better person to answer these questions than my dear friend and visionary investor Dr. Kai-Fu Lee.
Kai-Fu is Chairman and CEO of Sinovation Ventures, a $3 billion venture fund with a focus on exponential technologies and AI, and a world-renowned AI expert with over 30 years in the field, including senior executive roles at Apple, Google, and Microsoft.
In his new book AI 2041, Kai-Fu and his co-author, science fiction writer Chen Qiufan, lay out 10 stories—10 visions—of where 20 years of AI progress may lead us.
I recently sat down with Kai-Fu to discuss AI 2041 and his vision for AI. This Tuesday (September 14th) will be the book’s “coming out party” available on Amazon and on Audible. (I can’t think of a more fun read!)
In today’s blog, I’m going to give you an overview of our discussion of AI 2041, what you should and shouldn’t fear, and what this means for managing our AI-driven future.
Understanding “Your AI Future” is one of the key focus areas during my year-round Abundance360 Mastermind and Executive program.
Let’s dive in…
THE LATEST ADVANCES IN AI: HOW WE GOT TO NOW
What was the inflection point when AI really started to take off? Was it merely a matter of advancing computation? What drove the rapid change?
These are questions that Kai-Fu is uniquely qualified to answer.
He’s been at the heart of AI for over three decades: as world-renowned AI expert, as an investor in hundreds of companies focused on AI and other exponential tech, and as a senior executive at some of the world’s top tech companies, including Apple, Google, and Microsoft.
As Kai-Fu puts it, the driving forces behind the recent advances in AI have certainly been increasing computation—but even more importantly, an abundance of available data.
The fundamentals of many of today’s algorithms have been around for over 30 years. But it’s really only now, in the last 5 to 8 years, that we have had enough data and enough computation to make those technologies work.
For Kai-Fu, there have been two pivotal developments in AI over the last 5 to 7 years: one in computer vision and the other in natural language processing.
2014 was a watershed moment where computer vision began beating out human vision for recognition tasks. As a result, all sorts of applications were enabled in the ensuing years. Just think of autonomous vehicles, deep fakes, facial recognition, and automated retail.
And Kai-Fu believes that natural language processing reached a similar inflection point in 2020, beating out human comprehension for the first time.
For those of you who have seen OpenAI’s GPT-3 natural language model at work, it is not difficult to imagine an immense technological transformation unlocked by AI across all industries.
For example, OpenAI’s codex can now generate functional code merely by speaking to it with simple, natural language commands.
AI will turn everyone into a coder.
This will have implications for every industry.
Kai-Fu predicts deep disruption in search and advertising in particular. An NLP-driven search engine could be the “ultimate search.” Advertising and search will become more personalized and language-driven—ultimately guided by our own personal AI.
What happens to advertisers when you allow your AI to purchase products and services on your behalf?
SHOULD WE FEAR AI?
One of the stories in AI 2041, “Isle of Happiness,” explores an essential question: can we use AI to make ourselves happier?
Now, it’s not useful to think of an AI as a living being with emotions and desires—at least not yet, barring several breakthroughs.
Instead, today’s AI is great at objective function optimization: give the AI some metric to optimize, and in many cases it can do it better than any human.
Isle of Happiness is an exploration of how to choose the right things to optimize.
In the story, a Middle Eastern monarch creates an island where people live and share all their data in hopes of optimizing endorphin-based happiness. But it ultimately turns out poorly for the residents because simply reaching certain chemical levels doesn’t help people find a purpose for living and spending time with people you love.
Here’s the lesson: be very careful about what and how we optimize with AI.
That’s the real concern and potential danger of AI.
As Kai-Fu points out, “AI is brainwashing us because of the misalignment of interest in the objective functions: there are issues with transparency, explainability, and issues with personal data protection, security, and fairness and bias.”
Nevertheless, techniques are emerging to help us solve these challenges. I agree with Kai-Fu that 90% of the problems we’ll face with AI will ultimately be solved by technology.
For example, techniques such as federated learning, which we could use to train AI models while keeping your data private, allow us to “have our cake and eat it too.”
And when AI is optimized well, we can reach new possibilities and create global abundance: societies with less poverty, cheaper goods and services, more intelligence, and more opportunity.
How will people live in a society where things are abundant, easily accessible, and AI-driven?
in an age of abundance, purpose IS EVEN MORE IMPORTANT
When people no longer have to work to meet their basic needs, how should they spend their time? If robots take over most people’s jobs, how will they find significance in their lives?
Two stories in particular from AI 2041 explore the topic of having purpose in one’s life.
In “Dreaming of Plenitude,” advances in AI and other exponential tech drive down the costs of goods and services, and for the first time in human history, developed countries like Australia have eradicated poverty.
At the same time, continued improvements in solar and battery technology dramatically reduce the cost of energy.
People no longer need to pursue the accumulation of wealth.
And in another story from the book, “The Job Savior,” AI is doing most routine tasks and many people are put out of work.
Set in San Francisco, a new industry arises: job re-allocation firms that retrain and reassign displaced workers.
But what happens if the new jobs don’t satisfy many people’s need to feel productive and useful?
I’ve always believed that finding your purpose—your Massive Transformative Purpose (MTP)—is critical. Not only for entrepreneurs, but for everyone.
It will be even more important as rapidly advancing technologies move us closer to a world of true abundance.
These are just some of the stories and themes that Kai-Fu and Chen explore in their book AI 2041.
I highly recommend you pick up a copy. It’s an excellent, informed, and realistic look at what 20 more years of AI progress may actually bring.
With technologies as powerful as AI, it's important to understand the potential pitfalls (the “Black Mirror” perspective), but nevertheless forge ahead and build empowering and useful tools (the “White Mirror” perspective).
AI 2041 offers a valuable exploration of both perspectives.
If you’re a practitioner in the field of AI, this is a critical time to gain additional insight into how the technology is evolving, and where an AI-driven, abundant future may lie.
And if you’re an ambitious entrepreneur interested in AI, there’s never been a better time to get involved.
JOIN MY ABUNDANCE360 COMMUNITY
If you want to become an exponential entrepreneur and reinforce the mindsets that will inspire and guide you to create a hopeful, compelling and abundant future for yourself and humanity, then consider joining my year-round Abundance360 Mastermind and Executive program and come to our in-person A360 Summit February 2-4, 2022.
Every year, my team and I select a group of 360 entrepreneurs and CEOs to participate in the year-long program. You can join at any time for a full year’s worth of training on Exponential Tech, Longevity, Moonshots, and Abundance thinking and a mix of in-person and virtual content.
My mission is to help A360 members obtain mastery in four specific mindsets: an Abundance Mindset; an Exponential Mindset; a Longevity Mindset; and a Moonshot Mindset. Together we will actively select and reinforce your preferred Mindsets.