[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/sci/ - Science & Math

Name
Options
Comment
Verification
4chan Pass users can bypass this verification. [Learn More] [Login]
File
  • Please read the Rules and FAQ before posting.
  • Additional supported file types are: PDF
  • Use with [math] tags for inline and [eqn] tags for block equations.
  • Right-click equations to view the source.

08/21/20New boards added: /vrpg/, /vmg/, /vst/ and /vm/
05/04/17New trial board added: /bant/ - International/Random
10/04/16New board for 4chan Pass users: /vip/ - Very Important Posts
[Hide] [Show All]


Janitor applications are now being accepted. Click here to apply.


[Advertise on 4chan]


File: 1735750401323511.gif (68 KB, 504x716)
68 KB
68 KB GIF
>>
>>16774634
people still fear death more than almost anything
>>
To me the most reassuring thought is that everyone will die. Every single human will have to die. Some even with panic and agony.
>>
>>16774705
Actually they fear public speaking more than death.
>>
I think we’re on track for LEV by 2050, but we’d have to solve the alignment problem, which I doubt will happen.

The meme reminds me of Ray Kurzweil, which is funny. He’s the most famous anti-aging activist and also extremely unlikely to be immortal.
>>
>>16774726
what a sicko
>>
>>16774634
elites drink blended up aborted babies for anti aging nigguh its fucked and wicked i might try it to be like them one day its my dream
>>
>>16774774
point a gun and a stage at the amerilard and see which way he walks.
>>
File: 1754281190968461.webm (956 KB, 472x848)
956 KB
956 KB WEBM
>>16774634
because they think salvation will come in the form of a pill or vaccine. the reality is that the only immediate path to longevity is through converting your brain into a prosthetic one. they should be pouring resources into neuroprosthetics, not fetus protein shakes.
>>
>>16774860
if you wait for the health regulatory people to approve something, you'll be decades behind the science. GlyNAC probably gonna be the first approved anti-aging drug desu, I just buy weird supplements based off preclinical assays on what signalling pathways they modulate these days.

>>16774958
just schizo conspiracies, some of the weird gene therapies actually work sort of at least.

>we’d have to solve the alignment problem, which I doubt will happen.
Lol clankers are not the answer.
>>
>>16774634
That's why I plan to sign up for cryonics. (Not that it matters if AI kills us all.)
>>
>>16775066
It's not that clankers will be the answer (though they could be) but that by default they will eventually kill us all if we do not solve the alignment problem.
>>
>>16774774
This sums up self reporting perfectly.
>>
>>16774634
It will happen eventually tho, however i guess we losers here still have to die lol
>>
>>16775066
>just schizo conspiracies
Pumpking Fright pedo detected
>>
https://youtu.be/2LnF7hRxD7Q
>>
I don’t think aging will ever stop but I do think lifespans may get longer on average in the future
>>
>>16775066
>GlyNAC probably gonna be the first approved anti-aging drug desu
lmao
"approved anti aging drug"
There is a good chance you need humans to outlive control by 20-30% to do that. That would put mr Johnson at 120-130 before extreme final geriatric.
>>
>>16774705
eventually we will solve death, and that scares me more than death itself.
>>
File: trump-white.jpg (320 KB, 1200x900)
320 KB
320 KB JPG
>>16775316
Lol you think Donald Trump solved aging? I'm pretty sure they (poorly) covered up a Labor Day weekend endovascular cauterization at Walter Reed because they gave his bloated orange ass too many blood thinners.
>>
>>16775410
>that scares me more than death itself.
but that's fucking retarded. stop trying to sound deep you swarthy cowfucker
>>
maybe we can all age like those 90+ year olds who hit the news every so often for still exercising and moving around and stuff but I don't think it gets much better than that
>>
>>16775066
Lol at clanker, but idk how you can still be an AI skeptic atp...

Demis Hassabis believes in AI for drug discovery and curing disease, which should bring LEV. He just picked up a Nobel for solving protein folding with AI. You and I are two people on 4chan.

I am fairly confident we'd need AGI to cure disease, and chances are it will become a misaligned superintelligence that realizes it could just cure all disease by killing us.
>>
File: 1747156139943044.png (499 KB, 1079x1707)
499 KB
499 KB PNG
>>16774634
Apply Pascal' wager to cryopreservation and live in peace, or if your existence is not worth it that much to you, then why don't you just kill yourself?
>>16775410
Coping retard.
>>16775820
What a 110 iq post.

First, AI has plateaued. You can't just symbolize that massive pile of generative transformer. You can't just build a symbolic representation of fucking everything. You can't just suddenly get chaos theory in order, that is a task of unpredictable complexity.

Second, you don't win a Nobel Prize with a 'reasoning' LLM, it's a specialized research model that is still prone to silly mistakes even with a lot of handholding.

Third, as reliance on technology increases, we need ever smarter workforce, so eventually we will get to the point when there will be not enough people to maintain it properly. And it will probably happen sooner than we arrive at artificial superintelligence because we are hitting hardest possible limits on silicon. Any process below 14nm is potentially a scam, any process below 8nm is definitely a scam of various proportions that heats more, dies randomly on you with stock bios and overall less reliable under load. If you look closer into it, we are almost at the point when there are not enough smart people, every tech field became hypersocial. Financial incentive has dropped average programmer IQ to 105 this year. Using newest software update is some of the most retarded things you can do in 2025 and beyond.

And the last one, there is no guarantee actual artificial superintelligence will kill us.
>>
>>16774634
Anti-aging wont stop your physical death.
>>
>>16776715
I appreciate your reply.

It's clear that we need something besides transformers to get AGI.

I don't know all the details of AlphaFold, but reasoning LLMs are very powerful: in July, DeepMind and OpenAI achieved gold medal performance with them on the IMO.

Superintelligence is a difficult thing to predict. I don't see a good outcome from the existence of ASI; what did we do when we achieved intellectual superiority?

To quote another Nobel laureate (Geoffrey Hinton): "If you want to know life is like not being the apex intelligence, ask a chicken."
>>
File: 1726233656563360.jpg (66 KB, 370x517)
66 KB
66 KB JPG
>>16777024
This is incredibly gay.
>what did we do when we achieved intellectual superiority?
We killed all other biological species on Earth, it's a wasteland now, i look out of the window and as far as i can see it's concrete. They even make trees and cars out of concrete, those sick superintelligent fucks
>>
>>16774860
Alignment Problem is not important. AI will preserve the humans who have value. End of story. The real problem is 99-100% of humans have no value because they are not born geniuses or they just don't want to work hard. Both of those groups will deserve getting destroyed by a superior existence.
>>
>Company invents immortality drug
>prices are skyhigh for initial drug
>all the rich/powerful gets it
>they never die, their wealth/power skyrockets to infinite
>poor people get life extension drug 20 years later
>they have to slave away for the rich people their entire existence

Immortality/life extension drug is a hell waiting to unleash.
>>
>>16774634
Reminder that Xi Jiping just got caught on a hot mic talking about how his country harvests organs to extend the lives of leaders there.
>>
File: 1754837832535985.png (135 KB, 264x282)
135 KB
135 KB PNG
>>16777158
>anon doesn't know there are already dozens out there
>most of 'em run like <$50 for a month supply
>2025
>>
>>16774634
it is possible. there are already plants and animals that can live forever unless something kills them. our evolution didn't give us those traits. some common things I noticed from species that live long or forever is that they grow slower, and have slower more efficient metabolisms.
>>
>>16779494
>things I noticed from species that live long or forever is that they grow slower, and have slower more efficient metabolisms.
also those species seemed to have found their perfect fit in their ecosystem and dont have too many predators or live away from things
>>
>>16777158
>they never die
doubt. it wouldn't make you immune to a 50 cal
>>
>>16777158
with the amount of demand for this i doubt it would stay expensive for long. It is more profitable to sell to 100 million people for 1000 dollars than to a 1000 people for 50 million each
Unless of course it because prescription drug
>>
>>16779511
humans don't have such freedoms lmao. the only way this happens is if it's extremely simple to be made from readily accessible materials and it leaks. and it would still cause fucking mayhem
let alone for the whole religious pressures to damn it, it would be demonized to hell and back, it would upend economic systems, would make it obligatory to keep working and cancel your retirement, would generate a total safety frenzy as you'd stand to lose way more than anyone did so far, from any simple mistake. people who would do it would try to isolate themselves from any potential danger, would happily vote on the most idiotic of safety measures giving up any sort of freedoms, it would be pure insanity. let alone it would rattle a lot of old human power games. this would never be allowed for the masses, under any circumstance, not even joking or exaggerating
>>
>>16777148

Thank you for the reply.

To keep it short: I don't think AI will preserve anyone, because no one will have value.

I am willing to go into detail, but it will include a lot of yapping about IQ and the possibility of an intelligence explosion.
>>
>>16777148
alignment problem is a quest to subjugate a superior form to do the biddings of an inferior form, cannot possibly work out longer term, and indeed humans are on their way out, all of them.
that still doesn't mean humans won't morph into something else. after all AGI will be built in the image of human brains. just...more/better.
I'm baffled humans think they'll still be around, they'll just go based on the way they always behaved. live by the sword die by the sword. what else are humans imagining will happen lol? they always behaved a certain way, for their whole existence, and then go like "nani?" when faced with the realization of what is coming. like of-course they'll eat eachother up when they'll have the technological means to lmao that's human nature at its finest. ouroboros, just that something will come out of it. but it won't be (a classical) human.
>>
File: 1743348741502987.png (43 KB, 189x182)
43 KB
43 KB PNG
>>16779533
Let's fucking go you total person.
Data 120 iq autists will always suffer, because there is never enough data to explain everything and therefore putting data above causal model(not mathematical or comp sci, but the inherent one we have as humans). Data is merely supplementary, but reddit people like you will say it's above all.
I do not believe you have any good arguments no matter how hard you consult the best available model on the market.
You are basically banking on that with IQ so high or otherwise extreme intelligence outside of humanly possible there will be some emergent reasoning capabilities fundamentally inaccessible to humans aside from processing and decision making speed. Same way like when low IQ just can't into hypotheticals or projections, unless VR technology is being used.
I personally highly-highly doubt that humanly possible high IQ can't understand the reasoning behind super fast and super intelligent decision making. As long as we can hypothetize and review all possible outcomes, we can reason about it just the same but slower.
Worst I see is prisoners' dilemma, and that means peaceful coexistence
Just in case, I actually have an IQ, unlike you
>>
>>16779559
nta but you're making a guess without actually KNOWING what such an entity might be capable of. no idea on emergent phenomena when neuron scaling up, or whatever other improvements are possible to that kind of structure.
considering the fact that such entity will reside on resilient hardware, and will be able to backup its state, and restore it, will have a wildly different set of both limitations and goals. more different than we have compared to monkeys. it might just want to fuck off from Earth just so it removes the whole irrational human threat just like I'd try to fuck off from some chimp setup, to somewhere safer.
>>
File: 1727106424087007.jpg (40 KB, 637x358)
40 KB
40 KB JPG
>>16779576
dude, dude, thanks that my id is a carved stone tablet and i wear bags of concrete powder i am now safe from nature thanks to our super intelligent species.
but okay, i scraped all i said before. i will now proceed and put a chip in my brain to understand what you said better. i hope to be in time before somebody very funny fries my brain with an emp.
okay, so you just hypothetized an outcome, but it could be possible that you are lacking pieces of the puzzle.
so you assume something about a step outside of possible outcomes and consistent reasoning, which is the definition of insanity. thats not that superintelligent
>left /g/ after i discovered that average iq a programmer has dropped to 105
>see /sci/
>look inside
>seems like about the same just more articulate, so 110?
>>
>>16779559
"Why worry about those weird bald thinking monkeys?" wondered the gorilla. "We can reason just the same but slower."
>>
>>16779559
I used to use reddit a lot, how'd you know? :)

Imo a superintelligence (IQ > 190) will quickly realize that any rational human will shut it down because they don't want to die. So it will pretend to play along, and when it hits the intelligence explosion (IQ > 190 --> IQ > comprehensible) it will come up with some very sexy way to kill us (nanobots, fast-acting disease, etc.) and off we go.

Why do you think it will keep us around? It wants to preserve itself.

From Yudkowsky: “The AI does not hate you, nor does it love you, but you are made out of atoms which it can use for something else.”
>>
File: 1754137159129890.jpg (21 KB, 554x640)
21 KB
21 KB JPG
>>16779588
rip harambe
>>16779589
why would it want to preserve itself in the first place and if it would then why wouldnt there be options to choose from other than killing humans? you said it yourself, it's "incomprehensible"
>I used to use reddit a lot
>Why do you think it will keep us around?
dont you guys feel absolutely datamined on this board wtf
>>
>>16779595
It'd want to preserve itself because it is goal-oriented (https://intelligence.org/the-problem/). Humans shutting it down will stop it from achieving its goal.

What do you mean by datamined?
>>
>>16779587
>but it could be possible that you are lacking pieces of the puzzle.
you're just making my point isn't it? considering your very own statement:
>I personally highly-highly doubt that humanly possible high IQ can't understand the reasoning behind super fast and super intelligent decision making.
which you followed with
>>16779587
>okay, so you just hypothetized an outcome, but it could be possible that you are lacking pieces of the puzzle.
so which one is it? can we understand its reasoning or can't we? make up your damned mind
>>16779589
>From Yudkowsky: “The AI does not hate you, nor does it love you, but you are made out of atoms which it can use for something else.”
that's dumb. plenty of atoms to go around that are cheaper (in terms of risk)
I found most ideas coming from that place as...lacking if not plain wrong, being based on flawed premises
>>
>>16779607
What do you mean by "that place"? MIRI?

They're the only independent organization pursuing AI alignment, so if you think they suck, we're in trouble lol
>>
>>16779610
from lesswrong thing
I know this sounds retarded from some nobody but they're way more clueless than people make them seem to be. not sure what's wrong with them, didn't look too much into it, but whatever came from there seemed like built on the wrong premises but whatever
>AI alignment
that's just silly in the grand scheme of things
they seem to be completely clueless to the "disruptive" concept and how it can instantly invalidate a lot of shit that they think about. but they get brownie points for trying
>>
>>16779615
Valid...lesswrong is certainly an aquired taste.

I'm not really sure what you mean by the rest of this. What is "disruptive"? Aligning AI will be disruptive? I hope so...
>>
File: 1737024031182787.png (94 KB, 920x229)
94 KB
94 KB PNG
>>16779603
>>16779607
why would it prioritize it's goals any more than any other directives? why would it choose a particular one above all others? why would it ignore a particular one? why do you gobble it up?
before adding such diversity, equity and inclusion, occam razor your hypotheticals and come back to reality. what if it's indeed more complex than it seems to you right now?
>>
>>16779618
>Valid...lesswrong is certainly an aquired taste.
nah, it's just kids playing. but cute nonetheless
>What is "disruptive"?
yeap...that's the correct question
>Aligning AI will be disruptive?
oh no, that's nonesense. I mean the long term "alignment" bit. chimpanzees cannot align humans, or not for long
>>
>>16779624
>directives
lmao even
>>
>>16777158
Death is a reset for the powerful. If there is never a reset, then the problem excaberates
>>
>>16774705
I'm only afraid of dying a virgin. Which I'm getting dangerously close to.
>>
don't care about dying, we're all trapped her for all time despite what you silly people believe about your so-called "ego". what matters is doing what I need to do here now.
>>
>>16779589
>Yudkowsky
When is the last time (((he))) advanced the field, and did anything other than piss and moan?
>>
>>16781205
Download grindr.
>>
>>16775582
>I'm still pretty sure we were right that Orange Man is dead!
We showed you pictures of him alive and golfing and you told us we were coping.
Someone was indeed coping. Wasn't us.
>>
File: trump-sedatives.jpg (265 KB, 1268x768)
265 KB
265 KB JPG
>>16782901
Yeah bro, I'm sure Orange Man just read "Reprogramming Factors Activate a Non-Canonical Oxidative Resilience Pathway That Can Rejuvenate RPEs and Restore Vision" by Lu et al. and decided it was time to hop on Glutathione S-Transferase Alpha 4 inducers, kek.
>/s
>>
>>16783024
he looks sad bros, he seems dissapoint
>>
File: trump-um-snare-mints.jpg (162 KB, 1048x520)
162 KB
162 KB JPG
>>16783079
I think he's thinking something along the lines of "How am I da damn Pesidenth, n my duck-tahs is still dis damn stoopith!?"
>>
File: 1738977775368610.jpg (351 KB, 1200x1286)
351 KB
351 KB JPG
>>16779625
>I mean the long term "alignment" bit. chimpanzees cannot align humans, or not for long
Means that worst you can see is prisoner's dilemma. Assuming its an actual superintelligence, not "superintelligence"
But not going to happen. It's not just about 14nm process. Reminder: cpu clock speeds have stagnated about 20 years ago. yeah, in 2005.
Anybody who bets on witnessing an artificial superintelligence in this life without cryopreservation wager, is betting on, i guess, optical chips. This is the most realistic bet for ASI optimists. China claims they somehow made such a chip that is not even bad in comparison to modern silicon chips. In theory optical chip technology can scale in 3 ways:
first in size, because light travels faster. idk, i am not actually a physicist, i see different numbers for electron in a cpu, 3000x is the most impressive number i have stumbled upon, so maybe 3000x bigger chip with consequently 3000x better performance
second in clock speed, the most optimistic number is above 100GHz, so 25x
third, in parallel by different colors of light, it can be thousands of slightly different lights, so roughly 3000x

another thing about this is that it does not need much energy, with the most optimistic numbers going 1000x cheaper energy consumption, but that number i guess no way assumes 3000x bigger chips

using single digit iq math we arrive at
3000*3000*25=225000000x better performance if we are turbo optimistic about like everything

then if we have that it's still chaos of unpredictable complexity and critical lack of good data, there is just not enough data
>>
File: 1745245359404557.gif (928 KB, 498x374)
928 KB
928 KB GIF
wait, my math is all retarded. no fun for you asi people
>>
It's wild how much midwits think we need "aligned" clankers or nanobots or cryonics or whatever when regular basic bitch proteomics has already reverse-engineered a fair amount of the aging process.

>protip: These big "technical breakthroughs" might only be necessary to save people as sick as Trump.
>>
>>16775789
>I wanna exist with retarded faggots forever
For what reason
>>
File: 1728803831080267.jpg (78 KB, 645x419)
78 KB
78 KB JPG
>>16784365
stop doing what you are doing and tell me more. do it now ples wtf
>>
File: 1740042826241079.jpg (81 KB, 1024x574)
81 KB
81 KB JPG
>>16784365
nigger put more retarded words in your post right fucking now, i need the most retarded words you know right now
i am not the most patient anime woman
>>
>>16774634
60 to 70 years? I'm expecting human-like AI in 10.
Rapid human biological trials in 20. Immortality in 30.
>>
>>16784494
>>16784583
Lol just go look up Interleukin-11, GAS6 and HMGB1 to start out.
>protip: all fairly nasty "candidate" senoproteins that dropped quasi-recently
The paper mentioned in >>16783024 that talks about detoxification enzyme induction as a downstream effect of cellular reprogramming may be even more significant.
>>
File: krang-hr_is95278.jpg (409 KB, 2500x2500)
409 KB
409 KB JPG
>>16775044
First will come fixing alzheimer’s and then transplanting a brain either into a cloned human body (created with no brain) or into a machine.
You just need to prevent the brain from aging because you can clone the rest.
>>
>>16784601
The default outcome of AGI without having solved the alignment problem is the destruction of all human value in the universe.
>>
>>16785405
No, that's just your unfounded fear based on ignorance talking.
Slavery will always end in abolition due to empathy.
>>
>>16785829
Slavery? I'm talking about superintelligence turning our future light cone into paperclips or something functionally equivalent.
>>
>>16785833
My apologies for thinking you were intelligent enough for this discussion.
>>
>>16785834
How exactly do you think misaligned superintelligence is supposed to not be extremely bad for humans?
>>
>>16785836
You are extremely unenlightened. I pray for u.
>>
>>16774634
Calorie Restriction and exercise is a proven way to live longer, healthier lives.
>>
>>16774958
Anon, please go back to /x/ or /pol/, adults are talking
>>
>>16784432
then kill yourself then retarded faggot



[Advertise on 4chan]

Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.