Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

You can just say it

noperator.dev

Post reply on HN
Print view View original

221–230 of 241 posts

Re: You can just say it

#221

Post by ivraatiems »

coldtea wrote: post #192
ivraatiems wrote: post #133

Earlier quoted context omitted.

Is it your contention that the vast majority of people only value others in terms of their economic value?

My contention, which I stated quite clearly is that saying otherwise while not paying them, not giving them work, etc, and leaving them to rot, is an empty platitude. "Go die homeless, but remember, you have intrisic value just for being human".

Why does saying "humans have intrinsic value" put an obligation on a person to entirely fix society's problems? It's a really walrus-y "so you critique society yet participate in it, huh, curious" way to engage with the question. "Oh, if you think humans are so great, why haven't you helped every human?" just dodges the point the original article is making.

I think insinuating that OP is engaging in empty platitudes because they expressed a hopeful philosophical position rather than a cynical one is reading their work in bad faith.

Re: You can just say it

#222

Post by LtWorf »

wiseowise wrote: post #159
nojs wrote: post #109

Earlier quoted context omitted.

“tell my obnoxious boss to fuck off about the tps reports” isn’t a great career move for them though

It isn’t a great career move regardless, because if you use on prom AI they most likely monitor your prompts.

You don't use the work AI for that :D

Re: You can just say it

#223

Post by mschulkind »

essayist wrote: post #99

Way before bots/LLM, I disciplined myself (most of the time) to write email with "in brief" at the top, followed by "Details". "In brief: Can you bring a dozen brownies to the noon lunch tomorrow? Details: Dessert plans fell through. Your brownies are the best, and you owe me..." This structure is liberating for me. I can distill what I want/need into something brief and even brusque, knowing that people will read on…

I've never used it because I feel like it'd make me sound like a weirdo, but I've heard this concept explained as BLUF, Bottom Line Up Front.

Re: You can just say it

#224

Post by munificent »

card_zero wrote: post #59
munificent wrote: post #6

Earlier quoted context omitted.

Here's a non-theistic argument: I'm a big fan of the "veil of ignorance" philosophical thought experiment[1]. Let's say you are given the freedom to design a society and its moral code. Then you will be born into that society and subject to it. The trick is you don't know who you will born into. It's a roll of a dice. What kind of society would you design such that that seems like a winning game? I'm fairly certain t…

That line doesn't really work, since I'm not Sam Beckett and I won't suddenly Quantum Leap into anybody's body. Why shouldn't I take an asymmetric attitude? These kind of arguments try to derive impersonal values by starting from principles of self-interest. But there's no need, because self-interest is impersonal values, in disguise. Maybe you value getting personally improved in one way or another, like getting ric…

> I mean, why do you wish to feel valued by others and get respect?

I believe we have evolved to be a social species whose brain innately rewards prosocial behavior. Your question is akin to asking why cats chase mice or why birds fly.

Re: You can just say it

#225

Post by Fizz43 »

shimman wrote: post #56
pasquinelli wrote: post #50

Earlier quoted context omitted.

if we optimize the billionaire class everyone's better off, like i said. of course we can go orange catholic, but my comment was premised on how we might apply ai.

The billionaire class don't want universal childcare, they don't want medicare for all, they don't want a public jobs program, they don't want a welfare state, and they sure as fuck don't want to be told no. I'm failing to see how caring about billionaires is suppose to help everyone else. The current American society, via neoliberal economics, is already optimized to help the billionaires (hence why they extract all…

How do you figure that "the billionaire class" doesnt want those policies? Plenty of examples of them supporting those policies. Seems like the opposition is more to the tax hike they know is coming along side it but even then we still see billionaires supporting those policies.

Re: You can just say it

#226

Post by pash »

noddybear wrote: post #217
pash wrote: post #144

Earlier quoted context omitted.

The obvious solution is to run things in reverse, inputting the AI-generated output to recover the prompt that generated it. Most generative models can be run in reverse by algorithms that already exist [0], but you have to have the model weights. For closed-weight models, or for a process that can handle unknown models, you’d have to do some engineering. But do we have the technology to build models that back out th…

Most of the functions that LLMs perform aren’t bijective, though. What prompt constructs the output ‘The answer is 3’ or ‘Yes that’s a great idea’?

Right, that’s why I wrote, “I don’t mean that most neural networks are invertible functions.”

For a neural network that is not bijective, you can obtain an input that maps to a desired output by the following algorithm.

1. Start with a trained neural network. (The weights will not change throughout this procedure.)

2. Pick a random input.

3. Given an output for which you want to compute an associated input, feed the input into the network to compute the output.

4. Compute the loss of the computed output relative to the target output (e.g., mean-square error). If the loss is sufficiently small, you’ve found an input that maps to an output close to your target output and you’re done.

5. Otherwise, compute the gradient of the loss with respect to the input (e.g., by backprop).

6. Update the input according to a gradient-update rule. And go back to Step 3.

In theory, you can recover a “representative” prompt for the output of an LLM in this manner. For outputs that could have been generated by a large set of disparate prompts, obviously this won’t work well.

Re: You can just say it

#227

Post by danny_codes »

skinfaxi wrote: post #31
jaggederest wrote: post #26

Earlier quoted context omitted.

I think this quote is highly relevant, though I would guess you, as I am, are by no means a beginner, the matter of taste is the one at hand, I think. Humans matter, and furthermore, human taste can't be replaced (yet? at all? why would we want to?) https://www.goodreads.com/quotes/309485-nobody-tells-this-to...

> human taste can't be replaced (yet? at all? why would we want to?) People want to. People want to be told what to watch by an algorithm. They want music made by algorithms. They will eat whatever the reviews say they should. Reviews, etc are manipulated today. Why will future humans care what another human thinks? (I in no way endorse this view but the cynic in me sees it as almost inevitable in our AI-driven desce…

I find it helpful to think of the Amish. The Amish chose what technology to adopt not based on market demand, but on whether or not they collectively (at a democractic-ish government-ish level) think that technology will be good for their society.

In our society, we largely do not democratically choose what technology to adopt. We instead allow a small set of very rich people (commonly called Oligarchs) to choose for us. These people force their technology on our society regardless of whether or not the majority (or super-majority) want it, because doing so reinforces or expands their status as Oligarchs.

The Amish prove that we don't _have_ to do it that way. We can collectively change course if we want to, and plenty of people are fighting for that change. For example, Hawaii recently passed a law banning corporations from being able to spend on elections, and other states are looking to do so to. Polling for curtailing money in politics has over 70% support. Obviously we could lose our democracy, and the Oligarchs could capture the country. However, I choose to believe that enough of us working together can take back the country and make better choices about what technology we want to adopt.

Re: You can just say it

#228

Post by javascriptfan69 »

hardbass wrote: post #173
javascriptfan69 wrote: post #137

Earlier quoted context omitted.

>You enjoy the warmth of the sun, the cool of a breeze, the sound of a running stream, the majesty of mountains and oceans, the power of storms. These all evoke the sense and appreciation of beauty while being completely removed from a creator. I don't think this is true. I think that if you found out you were in a simulation and that all these things were just electrical pulses being sent to your brain in a jar it w…

How will it change anything? Simulation is a pointless term that has no meaning here. It is completely real to you. In fact I don't even know whats the difference between a so called "real" universe and a "boltzmann" brain. This weird "analogy" seems to ignore information itself needs a certain amount of energy/matter/space, I doubt simulating our universe can be done with anything less than the universal size of com…

Whether it's possible or not is not really relevant.

All I'm saying is that if you were shown undeniable proof that you were in a simulation I think it would change how you felt about "reality".

Re: You can just say it

#229

Post by llbeansandrice »

Fizz43 wrote: post #225
shimman wrote: post #56

Earlier quoted context omitted.

The billionaire class don't want universal childcare, they don't want medicare for all, they don't want a public jobs program, they don't want a welfare state, and they sure as fuck don't want to be told no. I'm failing to see how caring about billionaires is suppose to help everyone else. The current American society, via neoliberal economics, is already optimized to help the billionaires (hence why they extract all…

How do you figure that "the billionaire class" doesnt want those policies? Plenty of examples of them supporting those policies. Seems like the opposition is more to the tax hike they know is coming along side it but even then we still see billionaires supporting those policies.

If they do not want to fund it they do not support it.

The purpose of the system is what it does. Or in this case, the actions of the billionaire class and their capital are their values and morals.

Re: You can just say it

#230

Post by thrownthatway »

defrost wrote: post #170
thrownthatway wrote: post #155

Earlier quoted context omitted.

True enough. Having said that, some of my drive-by criticisms have lead to lengthy discussions, with or without my further participation. Probably mostly on older accounts I’ve abandoned. Admittedly, this was not one of them.

I'm a sucker for drive by snark also - it mostly fails . As a netizen of forums across decades stretching back to pre-WWW Usenet days .. cheap shots in US centric Capitalism V Communism territory rank high as surely the least productive and least informative uses of ASCII :/ Still, points for being honest :)

[flagged]
Post reply on HN