Hacker Newsnew | past | comments | ask | show | jobs | submit | proc0's commentslogin

It better include more than the cost of power to run them. I don't get why big tech is thinking idle hardware is so bad. Wear and tear exists, the more they're used at 100%, the shorter the lifespan. It's crazy Sam Altman wanted to shame people for owning something and taking care of it instead of being 100% "efficient".

Of course even with software the government can't help itself but produce a degenerate amount of documentation that you have to parse through before doing anything. I really hope AI helps out in this regard, i.e. they should train their own models or fine tune one to produce correct answers almost always. Then it's a matter of following the AI instructions. Of course that's probably too good and efficient for the US government.

It has begun.

Looking forward to the postmortem on this one and how much AI contributed to the problem and troubleshooting.

I've been saying AI hasn't been tested properly. All the showcases I see aren't really pushing it far enough and they're always things you could in theory do by searching answers online. The real value is in all the problems that cannot be solved by online search.


I do know that after introduction of AI assisted dev work in the last 2 years, SLA violations in our cloud product became a frequent occurrence - this after we had gone years without facing one. The hype train has pushed QA to the sewer.

I was thinking the same.

Btw, this is why no one should think AI-ops is a thing yet. All the promises of powerful, hands-off AI operations, preventive fixes, etc. are just BS.


Can't do AI-ops if you're the AI-ops provider that's got the outage :)

Because he cheated on his wife?? Or what am I missing? If that's the official reason, I don't buy it. I don't think those who can influence this decision give two shits about this type of morality. It's silly on its face. Obviously it's cover for some other agenda. I'm just surprised people buy this silly excuse instead of getting to the bottom of what is really happening.

That he conspired to smuggle NVIDIA GPUs into China to bypass a US embargo working with the Chinese government via a spy (either willingly or via blackmail). Impropriety adds insult to injury - the board have rightly seen that the market will view it more negatively if he stays than if he leaves.

> Because he cheated on his wife??

And a lot of other shady things around nvidia, half of it we probably will never know about. https://www.youtube.com/watch?v=1H3xQaf7BFI

https://www.businessinsider.com/trump-administration-jail-le...


Cheated on his wife with a spy who was using her connection to Jensen to move GPUs into China through Singapore in breach of the export controls.

The cheating is just a bit of humiliation, the damning part is the knowingly circumvention of export controls.


Oh if he knew all of that, then that's what I'm missing, thanks.

It's only been like a year since the last CEO resigned for viral infidelity.

And that was only with a subordinate, not a subordinate AND with a foreign spy (alleged). Reports say China got billions in chips imported thanks to this operation.

Politicians also drop like flies. The only person impervious to consequences has been Trump. None of his supporters care that he's the devil.


Nvidia bulls are still processing

No, it's because he repeatedly lied. 20% of the company's profits were/are coming from selling sanctioned chips to China and he said 0% 0% 0% (in official filings)

The news made it sound like she was a sophisticated person using him as a simpleton, but if anything he was using her to move his products against sanctions.


the impending investigation regarding the sanctions might have more to do with it, to be honest.

...with an actual spy.

And a subordinate executive. And god knows who else.

But I imagine the billions of dollars in illicit sales is considered a bigger issue.


The board should've: "Fire you? We’ve just spent billions of our market cap educating you to be careful with spies. Why would we fire you?"

allegedly?

> The zero-bug stance means that we address bugs immediately

No word no preventing them, which is funny. If you don't want bugs don't make them in the first place, but the myth in the industry is that is not possible. Is it?


Ah yes, I recall all those times as an engineer where my managers kept insisting I communicate like Socrates to improve my work. Software is solved.

I don't think the trend of replacing contributors with AI will last long. So far I'm betting it's companies trying different things and thinking all you need is prompting skills for 99% of things. I guess we'll see, but AI has this illusion of productivity, not sure how to explain it. There have been studies on it but you can feel it only after trying something large and wrestling with it everyday. It's not going to scale evenly across all domains.

It's really lack of creativity on the part of poltiicians and also the public. While the article is talking about a mathematical shape of voting, it also has the shape of a sports tournament. We certainly don't have problems picking winners for the NBA or World Cup. Voting is just too low resolution and there's so much more on the line. Why aren't there elections every year that have multiple levels of "primaries", showing brackets for candidates as they each go head to head with each other?

This is where I put on my metaphorical tin foil hat and think there is a lot of corruption that takes place to prevent fair elections that could let anybody win. The truth is those in power have every incentive to pull every lever they have to remain in power, and to gatekeep aggressively to make sure only those who think the same way even have a chance.


Isn't that what politicians in municipalities are? You only get to vote for nation-wide politicians after they already competed against other politicians by successfully ruling municipalities and countries.

You mean a political career? As in they get elected at the local level and then make their way up? That would be different, because every time they get elected is with the current limited form of voting for a bunch of candidates in one day. I'm saying voting itself could be spread out in multiple stages. But it's just one example of how it can be changed things instead of thinking we're stuck with one person one vote forever.

But doesn't that voting already happen, just inside a specific party? And your chances to get voted for in the party correlate to getting votes from the public. No party wants to elect people for their list, who then don't get votes from the public.

That's probably not the case. Parties operate like corporations, so there is probably a ton of bias, gatekeeping etc. when it comes to choosing candidates and that is often not in the public's best interests. An easy example is populists, who have public support but go against the party. Usually the party will not treat them as well as establishment candidates.

Sure, I didn't claim party rankings are made up by transparent voting alone.

> An easy example is populists, who have public support but go against the party. Usually the party will not treat them as well as establishment candidates.

That's kinda what I was referring to. Party support is supposed to favor people who actually have already governed or actually represent the ideals of the party over those who just talk big.


Everyone talks big. What are parties even for? Shouldn't holding office be completely separate like an isolated job? The fact they are held to party lines is somewhat of a failure of a democratic system or republic. The people should be choosing every step of the way and then once elected it should be like any other job. I don't know, just thoughts.

Placing power into the hands of a single person is a bad idea, because that reduces the chance errors are caught by multiple people needing to agree and a it's too much work for a single person.

If you are in the parliament, you can't potentially read into every topic and bill in deep. That's why you partner with other member of parliaments, who have the same ideals as you, and split the topics between you. That's a parliamentary group. Wouldn't it be great if you could know who is going to partner before the election, so you could vote based on that? Also if they are gonna partner after the election anyway, they can already do it before, so that they are able to fund an election campaign and talk to people, because a single person could not afford to do that, because of both money and time. Voila, you have a party.

If you only want rich people to become members of parliament, you can get of course rid of parties, because they can afford all of the above on their own.


Funding elections is broken. You're just outlining how you need to be a career politician to even have a shot. That's part of the problem in my opinion. What exactly are politicians/representatives supposed to do? That's a question that hasn't been asked in a long time.

I don't have all the details, and I'm sure it's pretty complicated. The point is that there is no way this is the only way to run a democracy. It shouldn't be a career, because a career is a self-interested, long-term endeavor. Holding public office is a public service and people who do it should only think about the wellbeing of their constituents and the rest of the country as a close second.


I've been saying this for a while. Programming has been unbelievably underestimated, almost as much as art (i.e. everyone thinks they are an artist when having opinions but rarely understand anything about it).

Programming is about engaging with high level mathematical abstractions.

To the extent AI can do this high level thinking, it is going to be able to do absolutely everything else humans can. Programming (when not underestimated) is as hard as the hardest cognitive task humans can do.


I used to think this too, but AI capabilities over the last couple years prove it completely wrong. The skills are more specialized per domain than I thought (though there is still a lot in common between fields)

And further, surely you know some very strong programmers who are very bad at art, or math, or human relationships?


Every new model is only marginally better than the last, since last year, and I'm sure they are throwing all the compute at every new model, which means it's hit diminishing returns.

Even after all of the resources it takes, it still needs human guidance. At the very least the conclusion is that it's not doing what humans are doing. They might know all the programming languages and all kinds of domains but then it needs to be told basic things to not fail at simple prompts.

Part of the problem is that this is very hard to talk about and describe. It requires a long conversation, probably better in real-time, to explain exactly how AIs are not really living up to the hype, and are actually introducing long-term risk by using them.


companies do not care about you doing hard programming, god they care about good programming only laterally

They care to the extent they don't want hard programming because it's more unpredictable and hard to hire for. The flip side is software builds complexity anyway and what I've seen is companies spending tons of money rewriting their codebases, or maintaining a disaster of a legacy system.

You are right, every time i see some arrow function i am sspicious

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: