> He told the interviewers that the "superpower of an entrepreneur" is facing enormous tasks and asking themselves, "How hard can it be?"
I get how it could be possible to be stuck can be on the hamster wheel asking yourself this question over and over for a lifetime chasing fulfillment.
However, the guy is worth $35b+...
> If the above sounds somewhat negative, Huang's tone in this part of the interview isn't really regretful but more like an impassive observer. He wouldn't willingly follow the whole journey again, saying, "I think it's too much. It is just too much." He references feelings like pain, suffering, shame, and embarrassment.
I don't doubt there was fear of failure along the way/close calls, lots of risk and potential/actual cases for shame, etc.
I just don't get why he isn't speaking from a position of fulfillment... it's ironic obviously. Why discuss shame when you have so much success to point to?
Then again, if he would've said "There were times I felt shame along the way but overall it's been amazing and I'd totally do it again", we wouldn't be here reading the article/headline/comments about it...
Maybe there is still more of his East Asian cultural roots in him than people think? It can basically just be part of being politely humble in those sensitivities. Glorifying oneself or one’s in-group is generally not well accepted across East Asia afaik.
> It can basically just be part of being politely humble in those sensitivities.
Saying "if I could take it all back and undo all of the success I've achieved and lives I've touched, I wouldn't do it because it wasn't worth it" seems a bit more than politely humble to me.
Indeed, to achieve that kind of success sacrifices have to be made along the way; whether in health, friendships or your morals. Some of those choices you may not have been willing to make looking back knowing the full consequences of your actions.
Reading between the lines, he’s saying his personality isn’t the type to work super hard to make super money. In other words, maybe he could have just been a high level engineer making $300k somewhere and been super happy minus the stress of creating Nvidia.
It’s not a general statement. He’s not giving advice.
That's a lot of words to express "cognitive dissonance." I'm pretty sure you know exactly why "he isn't speaking from a position of fulfillment" -- you know very well that he is a human being much like yourself, and despite "the guy is worth $35b+" he puts his pants on one leg at a time much like yourself.
Your problem is you cannot conceive that once upon a time he had no more than you. But now he is in a different place. Can you not conceive how and why you would get there yourself? Because if you did, I'm pretty sure you'd phrase this comment in a different manner.
> Who are these people who don't know how to put on pants while sitting on a floor/bed/chair/etc.?
That is a great point about the neologism. If you're sitting down you can put your pants on two legs at a time. However, there's something quite "everyman" about putting your pants on standing up, because you can do it anywhere.
Inside a bathroom? One pants leg at a time. Getting kicked out of someone else's home barely clothed because "someone else" got home? One leg at a time, albeit at a quicker pace. True, putting on pants two legs at a time from a sitting position is ideal, but it is a luxury.
Your problem is you cannot conceive that once upon a time he had no more than you.
Generally speaking people who end up super rich do have more than most people when they start. It's much easier to go from rich to super rich than it is to go from poor to super rich.
Why are you saying the word "generally" when you know, or should know, that the specific person is Jensen Huang, whose biography you could easily look up [1]?
It makes no sense to say "It's much easier to go from rich to super rich than it is to go from poor to super rich" when you're talking about someone who obviously grew up poor -- except to make a pet point that is completely inapplicable to the subject at hand.
It's fine to want to make it but I think you'd get more mileage from that point if you placed it in a discussion about a more applicable subject personality.
On the other hand, you will always have to worry about your money.
Either way, essentially what he seems to be saying is that it wasn't really worth it. After all, $35b isn't really what makes life worth living for most of us.
So you think if you had a billion dollars you'd worry about money less?
You'd just worry about different aspects of money. How not to lose what you have, how you will be constantly approached by grifters, how to keep yourself and your family safe.
Different issues, but still worry and still about money. IMO once you're past a certain level it's a lot more about mindset than about what you actually have, right up until you truly stand out from the crowd.
Do you think you can "just" live a normal life with a billion dollars?
Being poor sucks and being rich is nice, but being ultra-rich like that is a bit like being a popstar, where things like going out to a shopping centre or having a normal social life become impossible.
This was a reply to another comment on how uncanny it is that gaming hardware allowed the current machine learning advancements. It was deleted before I finished writing, but I'm not letting that stop me from sending this anyhow.
Creative immersive experiences, across the history of computing, are computationally expensive. The better computing gets, the more (visually at least) immersive the experience. The search for photo realism is an asymptotic search to compute a convincing simulacrum of reality. So much human nature drove the tip of that spear forward.
Immersion itself implies some kind of tricking of the consciousness, blending the familiar with the novel. Instructions and ideas piped into a machine that generates images. GPUs are specialized in parallel computation because that's precisely the architecture that best performs the multi-modal magic trick.
It's surprising to me how incredulous people seem about the possibility of creating a sentient machine that is fed off a large enough diet of human written material. Language is crystalized consciousness. It's the primary medium we use to communicate our inner experience to the outside. What other fuel would you use?
I've often wondered this. Would be interesting to see how useful that 3dfx IP ended up being? When I was a teenager I tried to get my parents to open a brokerage account so I could buy some 3dfx stock as I was so excited by what they were doing.
My language skills are damaged by a childhood brain injury. I think in pictures; I remember information with emotions, music, movement, and basically anywhere my brain could stuff information. Words are ephemeral, I often forget the beginning of a sentence by the time I reach the end.
The fact that we can take ideas and serialize them into language doesn't mean the language and the ideas are the same thing. They become ideas again when we deserialize them into our minds.
Language models can't do that (something else more complex could, but then it would not be just a language model).
The next generation of AI must be more than a language model. I believe AI will reach a stagnation point otherwise.
Also slowing down AI is that:
1) We don't presently know that much about the human brain or psychology. A lot, yes, but relatively not that much.
2) The scientific method is systematic, not systemic. Both perspectives are needed to interpret a reality made of interconnected systems.
3) There are many specialists and not many generalists. Generalists are needed to tie together the multiple domains required to emulate the human brain.
Can I message you about your brain injury experience? Your description of what your life/experience matches my own, and I had several concussions as a child. I never considered the possibility my "unique" way of remembering things and speech patterns was linked to brain trauma.
That doesn't really track. There's no obvious link between video games and machine learning, it's just that video games graphics were the one sector demanding low latency sky-high performance at the time the hardware became possible, so that's where the money went.
If video games didn't exist and instead we had found some extremely valuable ore that could only be mined by performing lots of hash calculations (so like bitcoin, but actually valuable), machine learning would have developed on miner asics, and then people would be telling stories about how of course sentient machines could only come from hashing numbers together because hashing is what our brain does or whatever.
As someone else said, not really. The link is linear algebra which is perhaps the best computational tool we have. In many areas of applied math, as soon as things get nonlinear (or non-quadratic if we’re taking derivatives since those derivatives become linear), it is impossible to find closed-form solutions (I.e., solutions you can write in a simple form). A big reason for this is that solving Ax = b might require a matrix (pseudo)inverse, but once you have that inverse, you have the solution for any pair (x, b), whereas if you had to solve f(x) = b, a new pair requires a fresh computation. Put in other terms, for linear problems, local properties are often global properties when studying linear problems.
A lot of the numerical methods we use to solve hard problems are based on solutions to linear problems (or quadratic problems if we’re taking derivatives) that predate computer graphics because those problems are simple to solve. We can just solve them a lot faster now so approximating nonlinear things with linear things iteratively becomes a lot more computationally tractable.
That's a pretty big claim. I disagree with it, but I agree about the idea of language models being a fundamental part of an artificial mind. Language is like the TCP/IP of ideas, we put data into it, and then extract data from it.
But the language and the data we extract from it are two different things, IMO.
Some of the commenters to the article state "if it's so bad, why did you not retire 3 years ago and live well"
The commentary is missing the context of the sunken cost problem.
This CEO, like many others, already paid his dues. He's missed many bdays, wrecked his health, and lost relationships. And now he gets to enjoy the benefits of the hard work, with the company is firing in all cylinders.
Walking away now would mean all the sacrifice was for nothing.
If you spent 30 years struggling, and now everything's running well, do you really step away in the hopes that the next person doesn't immediately bring it crashing down? Because the fastest way to destroy a company is the wrong CEO.
Do you build companies for money, fame, or change? You have to be honest with yourself. Not everybody builds companies for fame or change, otherwise acquisitions won't be a thing.
> This CEO, like many others, already paid his dues. He's missed many bdays, wrecked his health, and lost relationships.
Many of us have made similar sacrifices without being CEOs. A few lucky ones have even saved enough to be able to escape that lifestyle.
> And now he gets to enjoy the benefits of the hard work, with the company is firing in all cylinders.
What are those benefits?
He is already wealthy beyond imagination. He could stop any time and enjoy a life of luxuries that we can hardly imagine.
Beyond wealth, he has a legacy. In the industry he will be remembered as a highly influential figure and pioneer.
> Walking away now would mean all the sacrifice was for nothing.
How so? What were his goals when he started? Has he not achieved everything that he wanted and more?
Here's what I think happens to people like these who don't quit when they are already wealthy: they like what they have and they fear losing it.
They like feeling in power, respected by their peers and subordinates, they like feeling important, accomplished, even admired.
And, I suspect, they fear quitting because they realize that once they quit, they are just another bored rich dude with a estranged spouse and children. A has-been with too much free time who will forever try to talk about his past accomplishments to people who don't care.
Unless they reinvent themselves like Gates, but many seem to get stuck in the founder/entrepreneur mindset and start yet another company, never having learned their lesson.
Most of the CEOs I personally interacted with spend their day bullshitting around in their offices, making bad decisions, going home at 3pm and then writing an @company email from bed telling everyone that it's on them to make the place successful.
Not everyone is Steve Jobs, or maybe this guy. Most people probably aren't, and just get lucky/have dedicated staff that can work around bad choices from leadership.
They are at best managers, not entrepreneur or founders.
>Not everyone is Steve Jobs, or maybe this guy.
His name is Jensen Huang, mostly know as Jensen. I know HN isn't exactly a place for hardware. To the point it is worst than most hardware forums. But not even mentioning or knowing Nvidia's CEO's name is just on another level.
I knew the CEO’s name but I don’t see why there needs to be any reverence for him. “This guy” is plenty clear enough, we’re all talking about the same person.
Most of my accomplishments were things that I threw myself into without thinking and then figured out how to dig myself out. It's hard advice to give to someone else but it has worked for me.
Without thinking may be a stretch but for sure any big accomplishments require 'go for it' attitude, and normally there is no way to properly prepare yourself for unknown unknowns and assess risks correctly without previously walking similar/same path.
Bullcrap. This is just a demoralization play from an ubermensch who doesn't want young bucks trying hard like he did and perhaps giving him and his friends some competition/grief.
No one knows what they're doing when they are just starting out. This is just dumb.
I went to a talk by PayPal founders - Peter Thiel and Max Levchin - more than a decade ago (before I had any opinions about the former), and they essentially said the same thing.
And from what I see in the comments here, a lot of people are misunderstanding the point.
If Jensen, Thiel, or Levchin fully understood the challenges they'd face to get their companies off the ground, they would never have pursued it. That's not to say they regret the results.
For better or worse, they all persevered, and so it's easy for us to say, well yeah, obviously that was the right move. Because PayPal and Nvidia are obvious slam dunks in retrospect. But none of us have been there, as pioneers of the unproven, and it's the height of arrogance to contradict that.
The PATRIOT act (and others) was passed by congress, who wasn't then voted out of office by the electorate. So I don't think Palantir is really any more complicit than "the American people".
I don’t remember having any say at all one way or another.
I think most people feel completely apathetic and helpless when it comes to our government. They do what they want and there isn’t really anything the average person can do about it without an absurd amount of money backing you - so might as well do something different with your time.
Almost nothing done by congress that involves the military-industrial complex is “the will of the people” unless you redefine “the people” to be the thousand people who call the shots regarding the real allocation of tax money to industry.
There is a reason we spend two billion dollars a day on the US military whilst not at war and it has nothing to do with “the people” who are mostly struggling to make rent and afford groceries whilst working constantly without healthcare.
Most Americans don’t vote (but are forced to pay taxes to fund these things) so assuming their will without data is spurious at best and actively misleading at worst.
Choosing not to vote also factors in to the "will of the people". If you don't vote, you have no one to blame but yourself. If you vote for candidates that enable/abide the military-industrial complex, the same applies.
If you vote, but not for those people, and those people win anyway, well... that's how democracy works.
> If you don't vote, you have no one to blame but yourself.
This is the central lie/blame assignment of democracy, and it is plainly false on its face. How can someone totally uninvolved in a process be in any way whatsoever responsible for the outcome of that process?
You also go on to refute it yourself in the next paragraph.
On that, we are in agreement: there is no fundamental problem with the state that can be fixed by voting. Voting legitimizes the fraudulent and coercive system; abstaining is a vote for its invalidity. More people consider the system invalid than participate in it.
I think political posts should be empty if more people abstain from
voting than vote for any single candidate.
Aren't they just padding their own sholders? If the point was: "If i understood at the time what it would cost me, i would have refrained." OK. But it sounds more like "If I fully understood the challenges at the time i would have been scared." The latter is more or less drinking your own kool aid.
It's exactly this. It's a way to validate their money comes from facing lots of adversity and being very courageous. Tapping their own backs seems like an appropriate description.
> fully understood the challenges they'd face to get their companies off the ground, they would never have pursued it
What kind of challenges is that? I'm genuinely curious. I don't particularly see the job of an entrepreneur as difficult. Not in the sense that it's easy to be successful, but in the sense that it's laborious, painful, sufferable, miserable, relentless, etc.
For example, I'd consider medical surgeon a pretty difficult career. Years of study and practice. 24h+ shifts. Dealing with life and death on a consistent basis.
So I'm interested what are the kinds of hardships faced by entrepreneurs that get them to say that?
He explains in the interview. It's mostly the fear that if you fail you just let down and disrupted the lives of all your employees in a very public way. The fear of shame.
83 comments
[ 6.6 ms ] story [ 164 ms ] threadI get how it could be possible to be stuck can be on the hamster wheel asking yourself this question over and over for a lifetime chasing fulfillment.
However, the guy is worth $35b+...
> If the above sounds somewhat negative, Huang's tone in this part of the interview isn't really regretful but more like an impassive observer. He wouldn't willingly follow the whole journey again, saying, "I think it's too much. It is just too much." He references feelings like pain, suffering, shame, and embarrassment.
I don't doubt there was fear of failure along the way/close calls, lots of risk and potential/actual cases for shame, etc.
I just don't get why he isn't speaking from a position of fulfillment... it's ironic obviously. Why discuss shame when you have so much success to point to?
Then again, if he would've said "There were times I felt shame along the way but overall it's been amazing and I'd totally do it again", we wouldn't be here reading the article/headline/comments about it...
Saying "if I could take it all back and undo all of the success I've achieved and lives I've touched, I wouldn't do it because it wasn't worth it" seems a bit more than politely humble to me.
Any new announcement features a video of him making the announcement. They’re more CEO focused than Jobs lead Apple was.
And if the author is particularly respected and well spoken, even more silly to not squeeze that asset for all its juice
It’s not a general statement. He’s not giving advice.
Your problem is you cannot conceive that once upon a time he had no more than you. But now he is in a different place. Can you not conceive how and why you would get there yourself? Because if you did, I'm pretty sure you'd phrase this comment in a different manner.
Who are these people who don't know how to put on pants while sitting on a floor/bed/chair/etc.?
That is a great point about the neologism. If you're sitting down you can put your pants on two legs at a time. However, there's something quite "everyman" about putting your pants on standing up, because you can do it anywhere.
Inside a bathroom? One pants leg at a time. Getting kicked out of someone else's home barely clothed because "someone else" got home? One leg at a time, albeit at a quicker pace. True, putting on pants two legs at a time from a sitting position is ideal, but it is a luxury.
I imagine one might be in a sitting position there as well.
Generally speaking people who end up super rich do have more than most people when they start. It's much easier to go from rich to super rich than it is to go from poor to super rich.
Why are you saying the word "generally" when you know, or should know, that the specific person is Jensen Huang, whose biography you could easily look up [1]?
It makes no sense to say "It's much easier to go from rich to super rich than it is to go from poor to super rich" when you're talking about someone who obviously grew up poor -- except to make a pet point that is completely inapplicable to the subject at hand.
It's fine to want to make it but I think you'd get more mileage from that point if you placed it in a discussion about a more applicable subject personality.
[1] https://english.cw.com.tw/article/article.action?id=3442
MSEE from Stanford though... so he probably had student loan debt. ;-)
That on its own seems like a very mixed blessing.
On one hand, you never have to worry about money.
On the other hand, you will always have to worry about your money.
Either way, essentially what he seems to be saying is that it wasn't really worth it. After all, $35b isn't really what makes life worth living for most of us.
I have less than a billion dollars, and yet I worry about money a lot.
Curious.
You'd just worry about different aspects of money. How not to lose what you have, how you will be constantly approached by grifters, how to keep yourself and your family safe.
Different issues, but still worry and still about money. IMO once you're past a certain level it's a lot more about mindset than about what you actually have, right up until you truly stand out from the crowd.
Being poor sucks and being rich is nice, but being ultra-rich like that is a bit like being a popstar, where things like going out to a shopping centre or having a normal social life become impossible.
Cash out your XXbn and buy your dream house already!
Creative immersive experiences, across the history of computing, are computationally expensive. The better computing gets, the more (visually at least) immersive the experience. The search for photo realism is an asymptotic search to compute a convincing simulacrum of reality. So much human nature drove the tip of that spear forward.
Immersion itself implies some kind of tricking of the consciousness, blending the familiar with the novel. Instructions and ideas piped into a machine that generates images. GPUs are specialized in parallel computation because that's precisely the architecture that best performs the multi-modal magic trick.
It's surprising to me how incredulous people seem about the possibility of creating a sentient machine that is fed off a large enough diet of human written material. Language is crystalized consciousness. It's the primary medium we use to communicate our inner experience to the outside. What other fuel would you use?
https://en.wikipedia.org/wiki/3dfx
The human brain is so much more than language.
The fact that we can take ideas and serialize them into language doesn't mean the language and the ideas are the same thing. They become ideas again when we deserialize them into our minds.
Language models can't do that (something else more complex could, but then it would not be just a language model).
Also slowing down AI is that:
1) We don't presently know that much about the human brain or psychology. A lot, yes, but relatively not that much.
2) The scientific method is systematic, not systemic. Both perspectives are needed to interpret a reality made of interconnected systems.
3) There are many specialists and not many generalists. Generalists are needed to tie together the multiple domains required to emulate the human brain.
If video games didn't exist and instead we had found some extremely valuable ore that could only be mined by performing lots of hash calculations (so like bitcoin, but actually valuable), machine learning would have developed on miner asics, and then people would be telling stories about how of course sentient machines could only come from hashing numbers together because hashing is what our brain does or whatever.
A lot of the numerical methods we use to solve hard problems are based on solutions to linear problems (or quadratic problems if we’re taking derivatives) that predate computer graphics because those problems are simple to solve. We can just solve them a lot faster now so approximating nonlinear things with linear things iteratively becomes a lot more computationally tractable.
That's a pretty big claim. I disagree with it, but I agree about the idea of language models being a fundamental part of an artificial mind. Language is like the TCP/IP of ideas, we put data into it, and then extract data from it.
But the language and the data we extract from it are two different things, IMO.
The commentary is missing the context of the sunken cost problem.
This CEO, like many others, already paid his dues. He's missed many bdays, wrecked his health, and lost relationships. And now he gets to enjoy the benefits of the hard work, with the company is firing in all cylinders.
Walking away now would mean all the sacrifice was for nothing.
He's got a family friends, and one of the world's largest and most succesful companies.
Many of us have made similar sacrifices without being CEOs. A few lucky ones have even saved enough to be able to escape that lifestyle.
> And now he gets to enjoy the benefits of the hard work, with the company is firing in all cylinders.
What are those benefits?
He is already wealthy beyond imagination. He could stop any time and enjoy a life of luxuries that we can hardly imagine.
Beyond wealth, he has a legacy. In the industry he will be remembered as a highly influential figure and pioneer.
> Walking away now would mean all the sacrifice was for nothing.
How so? What were his goals when he started? Has he not achieved everything that he wanted and more?
Here's what I think happens to people like these who don't quit when they are already wealthy: they like what they have and they fear losing it.
They like feeling in power, respected by their peers and subordinates, they like feeling important, accomplished, even admired.
And, I suspect, they fear quitting because they realize that once they quit, they are just another bored rich dude with a estranged spouse and children. A has-been with too much free time who will forever try to talk about his past accomplishments to people who don't care.
Unless they reinvent themselves like Gates, but many seem to get stuck in the founder/entrepreneur mindset and start yet another company, never having learned their lesson.
If you watch the video he actually answered that question. Infinite ambitions. They started a journey, a mission, a goal and it is still not reached.
Not everyone is Steve Jobs, or maybe this guy. Most people probably aren't, and just get lucky/have dedicated staff that can work around bad choices from leadership.
They are at best managers, not entrepreneur or founders.
>Not everyone is Steve Jobs, or maybe this guy.
His name is Jensen Huang, mostly know as Jensen. I know HN isn't exactly a place for hardware. To the point it is worst than most hardware forums. But not even mentioning or knowing Nvidia's CEO's name is just on another level.
No one knows what they're doing when they are just starting out. This is just dumb.
https://www.youtube.com/watch?v=5aiFp-XqV_M
YouTube: https://youtu.be/y6NfxiemvHg?si=FpfBRSVNRQk8ebjb
Spotify: https://spotify.link/kP791sg35Db
It is really easy to get disheartened by modern C suite culture and the players.
This interview showed the other side. A great thinker and leader who appears to deserve any accolade bestowed upon him.
Why so binary? Like any person, he's probably great but also in some parts not.
But have you listened to that interview?
It was just refreshing compared to normal CEO speak / mythology / ego.
Is it just the stress of it all? Negotiating the deals, worrying you're going to lose it all, all your money, and be left with nothing?
And from what I see in the comments here, a lot of people are misunderstanding the point.
If Jensen, Thiel, or Levchin fully understood the challenges they'd face to get their companies off the ground, they would never have pursued it. That's not to say they regret the results.
For better or worse, they all persevered, and so it's easy for us to say, well yeah, obviously that was the right move. Because PayPal and Nvidia are obvious slam dunks in retrospect. But none of us have been there, as pioneers of the unproven, and it's the height of arrogance to contradict that.
The PATRIOT act (and others) was passed by congress, who wasn't then voted out of office by the electorate. So I don't think Palantir is really any more complicit than "the American people".
https://avalon.law.yale.edu/20th_century/eisenhower001.asp
There is a reason we spend two billion dollars a day on the US military whilst not at war and it has nothing to do with “the people” who are mostly struggling to make rent and afford groceries whilst working constantly without healthcare.
Most Americans don’t vote (but are forced to pay taxes to fund these things) so assuming their will without data is spurious at best and actively misleading at worst.
If you vote, but not for those people, and those people win anyway, well... that's how democracy works.
That description covers all of the candidates.
This is the central lie/blame assignment of democracy, and it is plainly false on its face. How can someone totally uninvolved in a process be in any way whatsoever responsible for the outcome of that process?
You also go on to refute it yourself in the next paragraph.
On that, we are in agreement: there is no fundamental problem with the state that can be fixed by voting. Voting legitimizes the fraudulent and coercive system; abstaining is a vote for its invalidity. More people consider the system invalid than participate in it.
I think political posts should be empty if more people abstain from voting than vote for any single candidate.
What kind of challenges is that? I'm genuinely curious. I don't particularly see the job of an entrepreneur as difficult. Not in the sense that it's easy to be successful, but in the sense that it's laborious, painful, sufferable, miserable, relentless, etc.
For example, I'd consider medical surgeon a pretty difficult career. Years of study and practice. 24h+ shifts. Dealing with life and death on a consistent basis.
So I'm interested what are the kinds of hardships faced by entrepreneurs that get them to say that?
Someone you don't forget