Sam Altman's Whale AMA

The best thing from I learned from program is how to start a startup. I think it's one of the biggest superpowers in the world today. Before I met Paul, I thought it was impossible. Now I think it's really hard, but still possible. I hope to teach other people how to do it.

Johnny Chin is definitely my favorite founder with a man bun, and probably everybody else's favorite founder with a man bun too. Looks like I need that answer to be longer than it already was. Maybe this is long enough.

The social problem that I would most like to see startups tackle in 2017 is the problem of equality. I think we are moving towards a more and more unequal world. We need especially education, but other social systems to change to fix that. I'm also interested in solutions like basic income.

This is a tough one. I would treat school as your day job, and then with the rest of your time, I would start working on projects that you're excited about. There's no reason a high school student can't start a startup. Hamlet's a very good choice. My favorite play is probably Macbeth. I think the life "To be or not to be" speech is amazing. I also really like Waiting for Godot.

Um, although truth be told, I am more into books than plays. The top three emotions I have felt this week are anxiety, sadness, and then happiness. I'm super, super nervous about the Trump presidency and how it's going so far. I'm scared that it's going to get a lot worse.

I've not been sleeping well at night, and I've been pretty upset most of the time. That's not consistent with most weeks. Most weeks, I'm pretty optimistic and pretty happy. I was happy sometimes this week when I've seen people come together to talk about how we build the future that we want. The spirit of that makes me tremendously happy.

I don't actively cultivate the emotions I feel. I don't think I'm particularly good at emotion control, but that's who I am, and so be it. What is the 10-year goal for Y Combinator?

Well, if we do really well, if we execute, if we're able to keep growing and finding the best young entrepreneurs in the world and making them successful and everything goes really well, then maybe in 10 years we can get acquired by a whale. I do you know I still find it faster for me to use this than to have it in a Word document.

I'm trying periodically to switch to a Word document, and there's something that just works for my brain and my workflow about pen and paper. I don't think that's the way that you build a hugely successful company. I think the most impactful, most important companies tend to have a pretty new idea, or if it's an old idea, they at least do it in a really new way.

I've tried a bunch of the nootropics, and I didn't think they worked, or I had headaches. Modafinil definitely does work, but it has side effects I didn't like. I do have one tip, though. I'm a vegetarian, and I found that taking omega-3 and B12 has significantly boosted my performance.

We're giving them money, and we're going to provide resources to help them however they'd like. Obviously, they run the ACLU. I don't think we can give them too much advice there, but if they have specific things they need help with, we will try to help them just like we do for any other nonprofit or for-profit we fund for that matter.

The one that I would start with is called Neural Networks and Deep Learning. It's by Michael Nielsen; it's available for free online, and I think it's a wonderful introduction. There are a lot of other books beyond that that you can find, but that's a good one to start with.

Our MOOC is going to start in early April; it'll run through early June. It's going to be awesome. I'm going to teach it, and we are going to try to replicate some of the Y Combinator experience in community for people taking the MOOC. I hope you'll check it out; I hope you like it.

The question I find myself struggling with the most is what will happen to the economy and to jobs as automation becomes more and more of a powerful force. I don't think I have a great answer here yet. I have a lot of ideas, but I think it's a really hard problem and critically important.

The two best things I've learned about enterprise sales, or sales of any sort, are: one, personal relationships are super important, and two, you have to be very persistent. I think if you keep those two things in mind, you can do pretty well and keep the sales cycle length as short as possible.

I think this is a really great idea. YC is not set up for this. I think you need a very different financial instrument, a different way to make returns. But I think someone else should do this. In fact, I've been encouraging a friend of mine to start this, and if she doesn't, I hope someone else will. I think it would have a really big impact on the economy.

Stripe has a product called Stripe Atlas. It makes it really easy to set up a U.S. company, get a U.S. bank account, everything else you need to do business. We recommend that to companies that want to form a U.S. entity that don't live here.

I think the most number of times someone has been rejected before getting into YC is six; that's the most I know of. There could have been one with more, and many of our most successful startups have been rejected. Some, like Dropbox, were rejected once with a previous idea.

There are far worse things that will happen to you as a startup than getting rejected by YC, so if that's a problem for you, you probably shouldn't be doing a startup anyway. But we love it when people reapply. My favorite Paul Graham essay is definitely "How to Start a Startup." That changed my life. I think it's changed thousands or tens of thousands of other people's lives.

I think it's an answer to what to do with your career that really appeals to some people. I believe in the 10,000 hours model for a lot of things, and certainly, I think for learning how to become a great engineer, it takes at least 10,000 hours.

So I think the way to become a great engineer is just to start building a lot of stuff, work with really smart people, and don't expect it to happen overnight. I haven't been able to in the last week, but I want to talk to him, and my plan is to explain why I think specific policies are bad.

I think that it's important that we keep talking to people that are in power and that hold different views. I think the left, we've done a very good job of convincing all of ourselves that we're on a path that we're nervous about and think is really bad. But we need to convince people on the other side.

And I think to them is the best way I can think of to do that, although I'm also happy to see things like the protests that happened this weekend seem to get noticed.