In this live, unedited conversation, Elon Musk -- the head of Tesla, SpaceX, Neuralink and The Boring Company -- digs into the recent news around his bid to purchase Twitter and gets honest about the biggest regret of his career, how his brain works, the future he envisions for the world and a lot more. (This conversation with head of TED Chris Anderson was recorded live at TED2022 on April 14, 2022. If you want to hear even more, head over to "The TED Interview" podcast, where Musk and Anderson sat down at the new Tesla Gigafactory in Texas to discuss some of the radical innovations he's working on -- including Tesla's intelligent humanoid robot, SpaceX's Starship and Neuralink's brain-machine interfaces.)

I want you to come with me to Tesla's huge gigafactory in Austin, Texas.

So the day before it opened last week, the evening before, I was allowed to walk around it, no 1 else there.

And what I saw there was honestly pretty mind-blowing.

This is Elon Musk's famous machine that builds the machine, and his view, the secret to a sustainable future, is not just making an electric car, it's making a system that churns out huge numbers of electric cars with a margin so that they can fund further growth.

When I was there, none of us knew whether Elon would actually be able to make it here today, So I took the chance to sit down with him and record an epic interview.

And I just want to show you a 9, an 8 minute excerpt of that interview.

Want us to switch now to think a bit about artificial intelligence.

I'm curious about your timelines and how you predict and how come some things are so amazingly on the money and some aren't.

So when it comes to predicting sales of Tesla vehicles, for example, I mean, you've kind of been amazing.

I think in 2014, when Tesla had sold that year 60, 000 cars, you said, 2020, I think we will do half a million a year.

Yeah, we did almost exactly half a million.

5 years ago, last time you came to TED, I asked you about full self-driving and you said yep this very year Well, I'm confident that we will have a car going from LA to New York without any intervention

Yeah, I don't want to blow your mind, but I'm not always right

So talk to talk what's the difference between those 2 why why Why has full self-driving in particular been so hard to predict?

I mean the thing that really got me, and I think it's going to get a lot of other people, is that There are just so many false dawns with self-driving, where you think you've got the problem, have a handle on the problem, and then nope, turns out you just hit a ceiling.

If you were to plot the progress, the progress looks like a log curve.

So most people don't know what a log curve is, I suppose.

But it goes up sort of a fairly straight way, and then it starts tailing off, and you start getting diminishing returns.

In order to solve full self-driving properly, you actually just have to solve real-world AI.

Because you say, what are the road networks designed to work with?

They're designed to work with a biological neural net, our brains, and with vision, our eyes.

And so in order to make it work with computers, you basically need to solve real world AI and vision.

Because we need cameras and silicon neural nets in order to have self-driving work for a system that was designed for eyes and biological neural nets.

I guess when you put it that way, it's sort of quite obvious that the only way to solve full self-driving is to solve real-world AI and sophisticated vision.

What do you feel about the current architecture?

Do you think you have an architecture now where there is a chance for the logarithmic curve not to tail off any time soon?

Well I mean, these may be an infamous last words but I actually am confident that we will solve it this year, that we will exceed.

You said the probability of an accident, what points do you exceed that of the average person?

We could be here talking again in a year and it's like, well, yet another year went by and it didn't happen.

Is there an element that you actually deliberately make aggressive prediction timelines to drive people to be ambitious.

So it feels like at some point in the last year, seeing the progress on understanding that the Tesla AI understanding the world around it led to a kind of an aha moment.

Because you really surprised people recently when you said probably the most important product development going on at Tesla this year is this robot Optimus.

Is it something that happened in the development of full self-driving that gave you the confidence to say, you know what, we could do something special here?

So, you know, it took me a while to sort of realize that in order to solve self-driving you really needed to solve real-world AI.

And at the point at which you solve real-world AI for a car, which is really a robot on 4 wheels, you can then generalize that to a robot on legs as well.

The things that are currently missing are enough intelligence for the robot to navigate the real world and do useful things without being explicitly instructed.

So the missing things are basically real world intelligence and scaling up manufacturing.

Those are 2 things that Tesla is very good at and so then we basically just need to design the specialized actuators and sensors that are needed for a humanoid robot.

But so talk about, I mean, I think the first applications you've mentioned are probably going to be manufacturing.

But eventually, the vision is to have these available for people at home.

If you had a robot that really understood the 3D architecture of your house and knew where every object in that house was or was supposed to be and could recognize all those objects, I mean, that's kind of amazing, isn't it?

Like, the kind of thing that you could ask a robot to do would be what?

of tea to grandma and show her family pictures.

Take care of my grandmother and make sure...

Obviously we need to be careful that this doesn't become a dystopian situation.

I think 1 of the things that's going to be important is to have a localized ROM chip on the robot that cannot be updated over the air, where if you, for example, were to say stop stop stop, that would, if anyone said that, then the robot would stop, you know, type of thing.

I think it's gonna be important to have safety features like that.

And I do think there should be a regulatory agency for AI.

I don't love being regulated, but I think this is an important thing for public safety.

Do you think there will be, basically, say in 2050 or whatever, a robot in most homes is what there will be and people

Yeah, you'll have your sort of buddy robot.

Like, how many applications do you thought is there?

You know, can you have a romantic partner, a sex partner?

I mean, I did promise the internet that I'd make cat girls.

So, yeah, I guess it'll be whatever people want, really.

What sort of timeline should we be thinking about of the first models that are actually made and sold?

Well, the first units that we intend to make are for jobs that are dangerous, boring, repetitive, and things that people don't want to do.

And, you know, I think we'll have like an interesting prototype sometime this year.

We might have something useful next year, but I think quite likely within at least 2 years and then we'll see rapid growth year-over-year of the usefulness of the humanoid robots and decrease in cost and scaling up production.

Help me on the economics of this so what Do you picture the cost of 1 of these being?

Well, I think the cost is actually not going to be crazy high.

If you can replace a $30, 000, $40, 000 a year worker, which you have to pay every year, with a one-time payment of $25, 000 for a robot that can work longer hours, doesn't go on vacation.

And that could be a pretty rapid replacement of certain types of jobs.

How worried should the world be about that?

I wouldn't worry about the sort of putting people out of a job thing.

I think we're actually going to have and already do have a massive shortage of labor.

So I think we will have not people out of work, but actually still a shortage of labor even in the future.

But this really will be a world of abundance.

Any goods and services will be available to anyone who wants them.

It'll be so cheap to have goods and services, it'll be ridiculous.

So, And now, arguably, the biggest visionary of them all, Elon Musk.

So Elon, a few hours ago you made an offer to buy Twitter.

Little bird tweeted in my ear or something, I don't know.

By the way, have you seen the movie Ted about the bear?

very important for there to be an inclusive arena for free speech where all, so, yeah.

Twitter has become kind of the de facto town square.

So it's just really important that people have both the reality and the perception that they're able to speak freely within the bounds of the law.

And, you know, so 1 of the things that I believe Twitter should do is open source the algorithm and make any changes to people's tweets, you know, if they're emphasized or de-emphasized, that action should be made apparent so anyone can see that action has been taken.

So there's no sort of behind-the-scenes manipulation, either algorithmically or manually.

But last week when we spoke, Elon, I asked you whether you were thinking of taking up, you said no way.

No, I think everyone will still blame me for everything.

If something, if I acquire a tool and something goes wrong, it's my fault

I think there will be quite a few errors, yes.

It will be miserable, but you still want to do it.

I hope it's not too miserable, but I just think it's important to the function of democracy, It's important to the function of democracy.

It's important to the function of the United States as a free country and on many other countries and to help actually to help freedom in the world more broadly than the US.

And so I think it's a it's a you know I think this the risk civilizational risk is decreased if Twitter, the more we can increase the trust of Twitter as a public platform.

And so I do think this will be somewhat painful, and I'm not sure that I will actually be able to acquire it.

And I should also say the intent is to retain as many shareholders as is allowed by the law in a private company which I think is around 2000 or so.

So well it's not like it's definitely not not from the standpoint of let me figure out how to monopolize or maximize my ownership of Twitter.

But we'll try to bring along as many shoulders as we as we're allowed to.

You don't necessarily want to pay out 40 or whatever it is, billion dollars in cash.

But what I'm saying is this is not a way to sort of make money.

I think this is, it's just that I think this is, my strong intuitive sense is that having a public platform that is maximally trusted and broadly inclusive is extremely important to the future of civilization.

It's for the moral good that you think it will achieve.

You've described yourself, Elon, as a free speech absolutist.

But does that mean that there's literally nothing that people can't say and it's okay?

Well, I think Obviously Twitter or any forum is bound by the laws of the country that it operates in So obviously there are some limitations on free speech in the US and of course, Twitter would have to abide by those rules.

So you can't incite people to violence, like a direct incitement to violence.

You know, you can't do the equivalent of crying fire in a movie theater, for example.

But here's the challenge, is that it's such a nuanced difference between different things.

There's hate speech, which some forms of hate speech are fine, you know, I hate spinach.

I mean, if it's sautéed in a, you know, cream sauce, it would be quite nice.

But the problem is, so let's say someone says, okay, here's 1 tweet, I hate politician X.

Next tweet is, I wish politician X wasn't alive, as some of us have said about Putin right now, for example.

Another tweet is, I wish politician X wasn't alive with a picture of their head with a gun sight over it.

I mean, at some point, someone has to make a decision as to which of those is not okay.

Or surely you need human judgment at some point.

No, I think, like I said, in my view, Twitter should match the laws of the country, and really there's an obligation to do that.

But going beyond that and having it be unclear who's making what changes to where, having tweets sort of mysteriously be promoted and demoted with no insight into what's going on, having a black box algorithm promote some things and not other things, I think this can be quite dangerous.

So the idea of opening the algorithm is a huge deal, and I think many people would welcome that, of understanding exactly how it's making the decision.

Like, what I mean is, like, I think the code should be on GitHub, you know?

So then, so people can look through it and say, like, I see a problem here, I don't agree with this, they can highlight issues, suggest changes, in the same way that you update Linux or Signal or something like that.

But as I understand it, at some point right now what the algorithm would do is it would look at, for example, how many people have flagged a tweet as obnoxious.

And then at some point, a human has to look at it and make a decision as to does this cross the line or not.

But the algorithm itself can't, I don't think yet, tell the difference between legal and okay and definitely obnoxious.

And so the question is, which humans, you know, make that call?

Right now, Twitter and Facebook and others, you know, they've hired thousands of people to try to help make wise decisions.

And the trouble is that no 1 can agree on what is wise.

Well, I think we would want to err on this.

If in doubt, let the speech, let it exist.

It would have, you know, if it's a, you know, a gray area, I would say let the tweet exist.

But obviously, in a case where there's perhaps a lot of controversy, that you would not want to necessarily promote that tweet.

I'm not saying that I have all the answers here, but I do think that we want to be just very reluctant to delete things and just be very cautious with permanent bans.

Timeouts I think are better than sort of permanent bans.

But just in general, like I said, it won't be perfect, but I think we wanted to really have, like I said, the perception and reality that speech is as free as reasonably possible.

And a good sign as to whether there is free speech is someone you don't like allowed to say something you don't like and If that is the case then we have free speech and it's it's damn annoying when someone you don't like says something you don't like That is a sign of a healthy functioning free speech situation

So I think many people would agree with that and looking at the reaction online many people are excited by you coming in and the changes you're proposing.

We agree that that Twitter is an incredibly important town square.

It is where the world exchanges opinion about life and death matters.

How on earth could it be owned by the world's richest person?

So How do you, I mean, what's the response there?

Is there any way that you can distance yourself from the actual decision making that matters on content in some very clear way that is convincing to people?

Well, like I said, I think it's very important that the algorithm be open sourced and that any manual adjustments be identified.

So if somebody did something to a tweet, there's information attached to it that that action was taken.

And I won't personally be in there editing tweets, but you'll know if something was done to promote, demote, or otherwise affect a tweet.

As for media ownership, I mean you've got Mark Zuckerberg owning Facebook and Instagram and WhatsApp, and with a share ownership structure that will have Mark Zuckerberg the 14th still controlling those entities.

If you commit to opening up the algorithm, that definitely gives some level of confidence.

Talk about some of the other changes that you've proposed.

So the edit button, that's definitely coming if you have your way.

I mean, I think, I mean, 1, frankly, a top priority I have, I would have is eliminating the spam and scam bots and the bot armies that are on Twitter.

You know, I think these fun influence, They make the product much worse.

If I had a Dogecoin for every crypto scam I saw, we'd have more than 100 billion Dogecoins.

sparking a sort of storm of excitement over Doge and, you know, where it's gone or?

I mean, I think Doge is fun and, you know, I've always said don't bet the farm on Dogecoin, FYI.

But I think it's, I like dogs and I like memes, and it's got both of those.

But just on the edit button, how do you get around the problem of, so someone tweets Elon rocks, and it's tweeted by 2 million people.

And then all those retweets, they're all embarrassed.

And how do you avoid that type of changing of meaning so that retweeters are exploited?

Well, I think you'd only have the edit capability for a short period of time, and probably the thing to do upon the edit would be to 0 out all retweets and favorites.

So in 1 way, the algorithm works kind of well for you right now.

So this is a typical tweet of mine, kind of lame and wordy and whatever, and look at the amazing response it gets, is this.

And then I tried another 1, And 29, 000 likes.

So the algorithm, at least seems to be at the moment, if Elon Musk expand to the world immediately, not bad, right?

So help us understand how it is you've built this incredible following on Twitter yourself when...

I mean, some of the people who love you the most look at some of what you tweet and they think it's somewhere between embarrassing and crazy.

But is that actually why it's worked or why is it worked?

I mean, I'm, you know, tweeting more or less stream of consciousness, you know.

It's not like let me think about some grand plan about my Twitter or whatever, you know.

I'm like literally on the toilet or something.

I'm like, oh, this is funny and then tweet that out, you know.

But you are obsessed with getting the most out of every minute of your day, and so why not?

So, I don't know, I just like try to tweet out like things that are interesting or funny or, you know, and then people seem to like it.

So if you are unsuccessful, actually before I ask that, let me ask this.

I have sufficient assets to complete the...

It's not a forward-looking statement, blah, blah, blah.

But I have to, I mean, I can do it if possible.

So, and I should say actually, even in the, originally, with Tesla back in the day, funding was actually secured.

In fact, this may be a good opportunity to clarify that.

If funding was indeed secured, and I should say, why do I not have respect for the SEC in that situation and I don't mean to Blame everyone at the SEC, but certainly the San Francisco office It's because The SEC knew that funding was secured, but they pursued an active public investigation nonetheless.

At the time, Tesla was in a precarious financial situation, and I was told by the banks that if I did not agree to settle with the SEC, that the banks would cease providing working capital and Tesla would go bankrupt immediately.

So that's like having a gun to your child's head.

See all Elon Musk transcripts on Tedtalk

Elon Musk talks Twitter, Tesla and how his brain works — live at TED2022