I've been telling my friends for years, you only live in San Francisco for the stories. You want the crazy stories—'I came home last night and there were homeless people having sex in my front yard.' Usually you get people to kind of admit that they kind of like that up to a point. But then it's like, I took my three-year-old out yesterday, went down to the park ... A junkie tried to stab my three-year-old with a used syringe. The jokes start to get less funny over time.

Every kid is going to grow up now with a friend, right, that's a bot. And that bot is going to be with them their whole lives ... It's going to know everything about them. It's going to be able to answer any question ... As close as a machine can get to loving you, it's gonna love you. That's going to be a thing, and every kid is going to have that.

We have top-end universities, yes, but with the capacity to teach only a microscopic percentage of the 4 million new 18-year-olds in the U.S. each year, or the 120 million new 18-year-olds in the world each year.

The 100-inch TV set that covers your wall is going to cost you 100 bucks. The four-year college degree is going to cost $1 million ... The four-year college degree is basically—just think about it technologically—it's an unchanged experience from 100 years ago. Getting a degree from a college university today, it's the same set of activities, it's the same format, it's teachers in the classroom, it's written exams. There's been no technological change whatsoever, and prices have exploded.

You want to have a great deal of conviction of where you're trying to head, but you want to have a lot of flexibility inherent in how you're going to get there and what the tactics are. Then, you want to be able to run a lot of experiments against that, and you can diversify your risk against any one theory by doing so.

If you're the CEO of a Chinese company... your relationship with the Chinese Communist party is not optional; it's required ... When the party says 'we get full access to all user data in America,' you say yes. When the party says 'you change the algorithm to optimize to a certain social result,' you say yes.

The reality, usually, with the big successes is that the founder has been chewing on the problem for 5 or 10 years before they start the company, and they often worked on it in school, or they even experimented on it when they were a kid, and they've been kind of training up over that period of time to be able to do the thing. So, they're like a true domain expert.

The truth is all the successful founders are working 80-hour weeks ... They form very strong social bonds with the people they work with. They tend to lose a lot of friends on the outside or put those friendships on ice.

Human beings, from here on out, are going to have an augmentation; and the augmentation is going to be in the long tradition of augmentations, like everything from eyeglasses to shoes to word processors to search engines, but now the augmentation is intelligence...

Steven mentioned this thing of, like, view every edit, basically, box as now an AI prompt. It's going to get very obvious the software that you're using doesn't have this capability, and you're going to kind of wonder whether the developers are asleep, because it's just going to be so clear what the benefits are. And so, I think it's going to hit hard and fast.

Globally, carbon emissions are rising and resource utilization is rising. In the U.S., carbon emissions and resource utilization are actually falling... We have figured out how to grow our economy while reducing our use of natural resources, which is a completely unexpected twist.

[AI] is us ... The reason this thing works. The big breakthrough was we loaded us into it. We loaded the sum total of like human knowledge and expression into this thing ... It's like a mirror. It's like the world's biggest finest detailed mirror.

People always underestimate the degree to which new technology actually creates new jobs.

If there is a lever for human progress across many thousands of domain simultaneously, it is intelligence.

It's actually a lot of the creative jobs that are most directly hit [by AI], I think, the fastest; and then it's the white-collar jobs that are actually less creative that are hit next ... A lot of the blue-collar jobs are actually really hard to hit.

Just think conceptually for a second, how did Lyft and Uber come into being? The thing that literally happened was a programmer sat down, typed in a bunch of computer code on a computer, and pressed enter, and the code started to run. And then, the next day, cars and their drivers started to drive to different places and pick up passengers who they normally never would have even known about. And those passengers, all of a sudden, had a completely new way to get a ride.

Software is the lever of the world. Somebody types in code, that code runs, the real-world changes. And the real-world changes not just arbitrarily but in a way that's more efficient, that's more organized, that's more productive, that provides more opportunity to people.

To start a tech company or to start an artistic movement... there's just incredible levels of complexity, branching future paths ... There's nothing predictable ... The really good innovators, they've got a drive to basically be able to cope with that and deal with that, and they basically do that in two steps. So, one is they try to pre-plan as much as they possibly can...

It doesn't make any sense anymore to have a database of information... about your customers ... Instead, what you're going to do is you're going to have a sales AI, and the sales AI is going to get trained on all customer communication ... And then you're just going to ask it questions.

Our biggest problem for the next decade is not going to be AI having too much effect; it's going to be AI having too little effect, because there's just these huge swaths of the economy that are just off limits.

If you're the CEO of a Chinese tech company, there is a political officer assigned to you who has an office down the hall; and at any given time, he can come down the hall, he can grab you out of your staff meeting or board meeting, and he can... make you sit for hours and study Marxism and Xi Jinping thought and quiz you on it and test you on it—and you better pass the test.

Basically, with science and technology, we're getting better and better at doing more and more with less and less until, ultimately, one day, we'll be able to do everything with nothing.

Entrepreneurship, creativity, innovation is what economists call decision-making under uncertainty.

Any new technology that works is a reordering of status and power.

Science fiction authors always talk about what's called the singularity. The singularity is what basically happens when the machines get so smart that all of a sudden, everything goes into exponential mode, and the entire world changes. My reading of history leads me to believe we're actually in the singularity already and that it actually started 300 years ago.

Software is our modern philosopher's stone. And in economic terms, it transmutes labor into capital.