Y Combinator Podcast #4: Sam Altman Interviews Elon Musk (2016)

https://www.ycombinator.com/blog/elon-musk-on-how-to-build-the-future/

by r3trohack3r • 3 years ago

42 53 3 years ago

gkoberger3• 3 years ago

[2016]

From 2016, so not as bad as I'd initially guessed, but still mostly a mutual pat-on-the-back fest, with very little actual substance.

elil175• 3 years ago

Another reminder that Elon Musk routinely lies about timelines. If Mars colonization had followed the timeline laid out here, we'd be three years from having lots of people on Mars.

On a side note, I loved Altman's interview style - he focused on the interviewee instead of soapboxing, which is unfortunately very rare in interviewers.

bryanlarsen• 3 years ago

A future projection is not a fact, so it's not a lie when it's wrong.

FrankoDelMar• 3 years ago

I will develop self driving basketball shoes by tomorrow, I pinky promise.

JohnFen4• 3 years ago

If I make a prediction without actually believing that my prediction is likely to happen, then I've lied.

signatoremo4• 3 years ago

From Wikipedia:

https://en.m.wikipedia.org/wiki/Technological_singularity

In 1965 I. J. Good predicted an ultraintelligent machine by 2000, Vinge (1993) predicted greater-than-human intelligence between 2005 and 2030, Yudkowsky (1996) predicted a singularity by 2021, and Kurzweil (2005) predicted human-level artificial intelligence by 2030. Moravec (1988) predicted human-level artificial intelligence in supercomputers by 2010 by extrapolating past trend using a chart, while Moravec (1998/1999) predicted human-level artificial intelligence by 2040, and intelligence far beyond human by 2050. In a 2017 interview, Kurzweil predicted human-level intelligence by 2029 and billion fold intelligence and singularity by 2045

Who do you think lied among them?

elil175• 3 years ago

I do not hold people who make unjustifiably confident predictions in high regard. Yud and Kurzweil are the ones from your quote who I am more familiar with and they are constantly wrong and constantly overconfident. They've spent decades making wrong predictions, and they keep doing it without updating their methods. To me, that's not making mistakes, that's lying.

abenga2• 3 years ago

Did any of them sell you kits implementing their predictions for thousands of dollars?

rsynnott• 3 years ago

There is a difference between saying “whatever will happen in 30 years” and saying “_my company_ will do a thing in 10 years”.

That said, these largely don’t sound _particularly_ serious?

Moravec (1988) predicted human-level artificial intelligence in supercomputers by 2010 by extrapolating past trend using a chart

I mean, come on.

autonomousErwin12• 3 years ago

I mean the guy said "the rockets will land themselves in 5 years". He might be off a couple of years but he delivered. Same with electric cars. Stands to reason he'll deliver on the Mars promise (though I think someone will beat him to the self-driving punch).

elil175• 3 years ago

You can cherry-pick the promises that have come true. But for every one of those there's a cyber truck, a "self driving mode," etc.

FireBeyond3• 3 years ago

Same with electric cars.

I mean, electric cars were a thing. Not a big thing. But nonetheless. And FSD is to me a lot less impressive than implied. And on FSD, he has repeatedly well, failed to deliver - "this year, coast to coast, no interventions" is something he's said each and every year since 2016. And it still runs in to the back of stationary emergency lighted vehicles. Until last year it could barely handle roundabouts. It saw level crossings as a weird intersection with a light that was red, and then not, and then red again, with a collection of trucks crossing it.

0xDEF• 3 years ago

TIL Sam Altman became the CEO of OpenAI because he is good at verbally patting billionaires on the back.

gkoberger3• 3 years ago

No, Sam Altman is the CEO of OpenAI because he built it. He's incredibly smart and driven, the only wealthy person I've worked with who I think is, at his core, a good person who cares.

I don't know the backstory, but they've gone their separate ways since this podcast. Musk left OpenAI after a failed takeover attempt:

Musk told fellow cofounder Sam Altman in early 2018 that he thought OpenAI, which has since created ChatGPT, was lagging behind Google, people familiar with the matter told Semafor. Musk offered to take charge of OpenAI to lead it himself, but when Altman and other co-founders said no, Musk stepped down from the board and backed out of a huge donation, per Semafor.

https://www.businessinsider.com/elon-musk-reportedly-tried-l...

bookmark1231• 3 years ago

Have you considered that maybe his skill is making other people, like you, believe he is a good person who cares? He is, after all, most known for his ability to influence others (particularly billionaires) into doing what he wants. And the whole doomsday prepper thing is pretty weird.

shapefrog• 3 years ago

TIL Elon bought Sam Altman a horse (or I assume he followed through with his promise to in exchange ...)

autonomousErwin12• 3 years ago

I think a lot of people wish this were true as it would justify any envy people have of him.

Unfortunately, I don't think you can be as successful at something like Artificial Intelligence by just sucking up to people with money.

I don't know him, but when a lot of smart independent voices are saying "this guy's wicked smart" I really think it would be a lot harder to fake it than to actually be wicked smart (and deeply altruistic - fella seems pretty genuine about progressing humanity instead of MSFT's share price).

JohnFen4• 3 years ago

it would justify any envy people have of him.

What envy?

drcongo• 3 years ago

I mostly agree with you, but then again a lot of people say "this guy's wicked smart" about Musk and he's clearly a buffoon.

akudha• 3 years ago

Deeply altruistic? Isn’t he the guy behind worldcoin, creating a privacy nightmare? How is collecting biometric data from millions of people ethical, especially when most of those people likely don’t even understand what it is that they’re giving away, for little money?

hollerith2• 3 years ago

It seems to me that the really wicked smart people understood long ago that AI research is very dangerous and that founding a new company along the lines of OpenAI would increase the danger: yes, it would provide significant benefits to humanity, but then it would wipe out the human species, either directly or by helping another AI lab to do so.

I'm praying that the reckless actions of people like Altman do not kill me and everyone I care about.

Everyone knows someone who is wicked smart and very ambitious, but has a blind spot such that they cannot apply their intelligence to perceive correctly a situation in which they must lose status (i.e., admit that founding OpenAI was a mistake and shut it down) for the sake of the greater good. Maybe Altman is one of those.

Also, when in human history have the people who've been most successful in scrambling to the top of some sector of the economy been the ones who successfully recognized when that sector has become destructive? When the commercial excesses of America during the Gilded Age needed to be reined in, was it some industrial magnate like Carnegie or Rockefeller who sounded the alarm and led society's corrective action against the excesses? Was it the CEOs and leading scientists of the fossil-fuel companies that led the charge against global warming?

ZephyrBlu• 3 years ago

Even if this is true it's odd to disparage someone based on this. Influencing people is a valuable and important skill.

Y Combinator Podcast #4: Sam Altman Interviews Elon Musk (2016)

Related Stories