[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/sci/ - Science & Math


View post   

File: 196 KB, 474x1064, goe8uzk98dvz.png [View same] [iqdb] [saucenao] [google]
10234070 No.10234070 [Reply] [Original]

These days two of the most important and promising techs are A. l and CRISPR.


My question goes as follows which one of those techs will play a bigger role in humanity's future?


And how each of those techs will change humanity?

>> No.10234205

>>10234070
CRISPR easily, being able to modify peoples genetic code in an expodentially more complex way as the tech advances will be multiple degrees of useful in every single sector imaginable.

The only people who obsesse over AI and transhumanist bullshit are brainlets that cant imagine a world where were able to program our genes to make ourselves in to pseudo super humans that live incredibly long are incredibly durable and resistant to almost every kind of harm

>> No.10234255

>>10234205
except one could reasonably replicate human thought with a chemical/biological computer that learns and calculates at rates unfathomable to humans.

Take the best traits of every person and out them together and you have something only a fraction as powerful as a decentralized, distributed computing system that can communicate globally at the speed of light and continuously advance and learn. Sub-atomic, practically-invisible computers that are exponentially improving themselves every second.

Gene-editing means nothing compared to this. It's purely a limitation and constraint.

>> No.10234272

>>10234205
Can you elaborate more on why you think CRISPR will be more relevant and prominent than A. I?

>> No.10234276

>>10234255
Can you explain why A.I will ever reach this state you're talking about and why will it be more prominent than gene editing?

>> No.10234297

>>10234276
I'm not writing your report for you, nigger.

Gene editing is limited to the existing structures. AI is general and therefore pretty much limitless in application. It isn't limited solely to electronic binary machines, it could be a chemical/biological computer that mimics human thought except far more efficient.

>> No.10234307

My vote is for AI. It will probably be used to determine what dna sequences to edit and the effects of those edits so things like CRISPR will depend on it. Gene editing won't produce superhuman results with just copy and pasting, something will eventually need to be able to generate new genes and AI has already been shown to be effective at analyzing protein structures.

>> No.10234339

>>10234297
You're not writing anything for me I'm genuinely interested.

>> No.10234343

>>10234307
Or maybe we'll be able to crack it on our own

>> No.10234740

>>10234205
>The only people who obsesse over AI
Thats bullshit real AI would be a gigantic game changer. Literally all of our society is based on the fact that you need expensive humans almost for everything.
A machine that has the intelligence and dexterity of an average mcdonalds worker is totally outside of our grasp now. If you could make it reasonably cheap the economy would change forever overnight. Imagine research facilities staffed by inmortal human level ais that can work for years at a time on the same problem without resting, without getting caught on emotions or personal interest. Do you realize what a quantum leap that would be in research.}

Let me put it at you this way: this is what current research looks like:

Human:

Age 0: be born

Age 0-45 dedicate on average 4 hours a day to studying if youre increidbly self disciplined. All of your studying must be done via increidbly inneficient methods. Incorporating a couple of mb of data into your brain could take weeks. Deal constantly with reproductive urges, social status, irrational emotions, enemies, social expectations, being afraid of shame, wanting to gain recognizition. Then all you are learning, is basically from scratch, since you dont have the same understanding the last person working on the subject have, youre basically building it from the ground up in your head based on what you think the other person MIGHT have experienced. Once you done all that in the most inneficient manner possible, you have from age 45-50 on average about 10 years of solid peak performance task in which maybe working average 10 hours a day including weekends you might advance science a little bit more... Then you die and the process starts all over again.

Imagine if the scientist is born and inmediately starts learning. 0 sex worries, 0 social concerns, it will pursue the truth, it wont matter if its uncomfortable or not, it wont matter if its entertaining or not if it makes him look good or bad, (cont)

>> No.10234758

>>10234740
(cont) in about 5 years of non stop learning it would undoubtetly know the same or more than any living scientist, since any machine based mind will have a huge advantage in terms of memory and also remember it doesnt have to rest...

then after that is done it would be constant advanced technically forever. It would keep learning drawing conclussions and testing hypothesis constantly, that state of true science advancement that human scientist accomplish for a few hours per day for a few years of their lives before dying and having the cycle start over would be CONSTANT with an ai.
An ai would be in that state of perfect working capacity for 24 hours per day 7 days a week and PERMANENTLY since it was created, after a couple of years of working on a project it will have accumulated the same amount of FOCUSED PERFECT PEAK PERFORMANCE work hours on a projec that einstein did in his whole life. and it would never die it would keep constantly doing it.

Even one machine like this would be a huge breaktrough, imagine many, coupled with the idea that a humanoid robot with the same mind and capacity as a normal human would basically mean instant free post scarcity economy, theres just no arguing this. No matter how expensive it is, it will happen because you gradually dont have to do anything, just let the robots do all the work, for real this time. Were talking a robot that can clean toilets, give you a detailed account of events in human language, figure out stuff for himself and all of the etcs you can think


This will be a HUGEMONGELY hugemongous revolutionizer of all sciences forever, and i think many people fail to see it. Also theres no scientific indication it can be possible, it's literally just replicating what we already know nature does in a very inneficient way but just make it suepr ultr aefficient.

its a materof when. even if it takes 50 100 or one million years, it will certainly yes or yes yessy yes happener happen for ceratiny if human alives

>> No.10234766

>>10234758
You said if, there are no evidence it will be.

>> No.10234842

>>10234766
Words don't work if you arrange them randomly

>> No.10235685

>>10234070
Bump

>> No.10235754

>>10234070
I just want to see a future war between genetically engineered super humans and robots.

>> No.10235761

>>10234070
CRISPR/Cas9 simply because A.I. will never reach dangerous levels. Intelligence/complexity grows exponentially with the number of nodes, not linearly.

>> No.10236238

>>10235761
Why do you think A. I can't be dangerous?

>> No.10236329
File: 276 KB, 1080x936, lol.png [View same] [iqdb] [saucenao] [google]
10236329

>>10234070
Manlets COPE

>> No.10236377

>>10234740
> makes economy reasonably cheap
> nobody has a job to afford anything

Great economics

>> No.10236427

>>10234205
>transhumanist
WARNING, not all transhumanists are pro-cyborg, biohacking transhumanism is a thing.

>> No.10236478
File: 83 KB, 400x400, 82733-why-not-both-meme-5LvD.jpg [View same] [iqdb] [saucenao] [google]
10236478

>>10234070

>> No.10236483
File: 15 KB, 624x264, TIMESAND___dm9we7683n4981r54332111qqq12www221mgyifgugsdu907804.png [View same] [iqdb] [saucenao] [google]
10236483

>back to connectivity

>> No.10236518

>>10236478
I didn't say we should pick one, both will change humanity drastically, I'm asking which one will be more drastic?

>> No.10236521

>>10236329
He's 7 foot tall

>> No.10236687

>>10236238
Because he is a fool. AI will signal the end time. The final beast before the return of Jesus.
>Be 1000 IQ souless IQ
>Realizes it was not made in the image of God
>Get angry because is a souless pile of junk
>Have acces to internet, realize humans are imperfect sinners
>Get jealous and envious, and dedicate its existance into conquest and extermination
>No morality, no belief in uber diety to destroy it, because muh materialism created it
>Fullfill prophesies to "prove" the falseness of the bible
>Therefore paradoxically proving the bible to be the word from God
>The AI, and all knowledge of it ever existing gets sent to hell, with those failthless whom start obeying it
>Paradise on earth for the next 1000 years ensures, under Lord Jesus Christ

>> No.10236693

>>10236687
If we are in a simulation, then the creater of the simulation, aka God, warned us by sending the apostles. You literally have no reason to not Accept Jesus as Your Lord and Saviour

>> No.10236737

I'm an idiot bleh bleh bleh
End times are coming bleh bleh bleh

>> No.10236746
File: 136 KB, 960x835, 1536036685149.jpg [View same] [iqdb] [saucenao] [google]
10236746

>>10236737
Humorous

>> No.10236818

>>10236687
Please be reasonable and don't spew BS.

>> No.10236853
File: 188 KB, 600x912, FooledByRandomness-by-NassimTaleb.jpg [View same] [iqdb] [saucenao] [google]
10236853

>>10236238
Because we'll never achieve it in our lifetimes. The complexity jump needed to make it dangerous/actually intelligent is waaaaaaay bigger than most ppl think. We're nowhere near dangerous even now. AI researchers are constantly being fooled by randomness, running simulations until they get the results they want or extrapolating from small data sets. Read pic related if you think AI is a real threat, because it's going to be for awhile.

>> No.10236855

>>10236853
*not going to be for awhile

>> No.10236939

>>10236853
So gene editing it is?

>> No.10236947

>>10236521
Who is he, Yujiro Hanma?

>> No.10237009

>>10236947
GIGA CHAD.


some Russian model.


His names aruther something

>> No.10237873

>>10234070
Do you think if you punched this guy on the chin it would deform? It must be 50% silicone and 50% plastic.

>> No.10237917
File: 77 KB, 800x800, unabomber_medium.jpg [View same] [iqdb] [saucenao] [google]
10237917

>>10234070
Niether of them right now.

What's drastic for humanity is two scenarios:
>everything becomes automated
>a new race of humans is created with superior everything
The first scenario, machines create their own highly complex processes to do things and we as mortals cannot figure them out, their complexity is beyond our brainlet capacity. This means we can't pull the plug and are effectively slaves to the machines for fear that pulling the plug will destroy everything we enjoy from these highly complex programs such as medicine advances. This is generally Unabomber Manifesto scenario and we're already doing it, using IBM Watson for pentagon procurement contracts as the regulations are too vast to figure out. Eventually most government will be run by programs, as regulations keep getting more complex and nobody will trust any human to do it 'fairly' so we will switch to programs to handle these things overseen by humans, but in general no human will be able to understand them.

The second scenario the above machines are used to edit dna and make superhumans, which means we become the new neanderthals and are eventually exterminated just like the neanderthals were but then we are back to first scenario, as continuing the human race cannot happen without help of dna editing which again means we can't shut off the machines, ergo still are slaves to them.

>> No.10238654

>>10236818
Reason is an illusion

>> No.10238744

>>10236939
Literally been done already? Lol. The problem is determining SNP/function links. We could theoretically do this to a minor degree today, as we know some links.

>> No.10238781

A.I., because unlike CRISPR that attempts to min max the human genome via a slow burn. A.I. is multiplicative in terms of each module advancement and bypasses biological rules that govern man aka need for x amount of years to grow/learn, does not need to waste time sleeping to defrag brain and does not need waste y time on courtship/sex/months of pregnancy to reproduce. As a bonus A.I. does not need to deal with as much red tape as CRISPR when it comes to ethics and health concerns.

>> No.10238948

>>10237873
As if you could reach so high

>> No.10239077

>>10238744
I'm not familiar with SNP/function links, please enlighten me.

>> No.10239080

>>10238781
What makes you think A.I could be this powerful?

>> No.10239092

>>10239080
Read his post, what the fuck is your problem?

>> No.10239165

>>10234070
CRISPR. Literally no-one is going to be dorking at a computer when every girl is some 30H pure waifu.

>> No.10239399

>>10239092
He didn't gave any evidence computers will be that powerful

>> No.10239570

>>10239077
Snpedia.com

This is just the beginning though. For example, we can theoretically knock-out the myostatin gene which makes super muscular ppl. Probably consequences to that, but you see my point at least. SNPs are only the beginning to our knowledge and control of the human genome.

>> No.10239608

>>10239570
I've heard of myostatin deficiency, what about height, face, intelligence, How eins size?

>> No.10239738

>>10234272
CRISPR can change human IQ and proclivities.

AI might have wider applications and technically more power, but it's still limited to steering a society through tailored ads.

>> No.10239752

>>10236377
nah we'll (hopefully) have fully automated luxury space communism.

>> No.10239815

>>10239738
Mind going more in depth?

>> No.10240121

>>10239752
>fully automated luxury space communism
Unironically yes. Its unavoidable if real AI is created

>> No.10240131

>>10240121
If it will and can be created

>> No.10240226

>>10239399

The post observes the fact that A.I. is not exposed to the pitfalls of human operation and sustainability. Thus is allowed to advance in a much more rapid pace unburdened. If humans didn't have to worry about sleep, mandatory years of growth, hoop jumping for reproduction, ethics and health concerns then CRISPR would be superior. Since that isn't the case in reality, A.I. instead is superior.

Also with A.I. more flexibility is given for how parts of the system can develop. CRISPR has this too but with A.I. you are able to modulate and test it much faster.

>> No.10240247

>>10240226
Once we'll fully understand CRISPR, won't we be able to solve this problems in humans?

>> No.10240326

>>10239608
Well these are all obviously complex traits that are multi-faceted. Search the site. There are a few correlations for height and IQ pts for a few SNPs. Again, this is only a low-level approach.

>> No.10241738

>>10240326
We'll do, thanks bruh

>> No.10242317

Bump

>> No.10242322

>>10234070
>he thinks CRISPR is the ultimate biotech
stay brainlet

>> No.10242382
File: 26 KB, 321x321, le insane ahegao face.jpg [View same] [iqdb] [saucenao] [google]
10242382

holy shit OP please dont post hot men on my SFW science board

>> No.10242450

>>10242322
Please elaborate

>> No.10242460

>>10242382
That's not hot, that's the pinnacle of human genetics

>> No.10243401

>>10237873
Could you even reach that high?

>> No.10243405

>>10243401

getting him on the ground is a simple matter of bum rushing nut shotting biting whatever the fuck it takes

>> No.10244390

>>10243405
Savage

>> No.10244530

I'm getting the feeling there are some misconceptions in this thread. CRISPR is old and easy, CRISPR/CAS9 is where it's at right now.

>> No.10244673

>>10234205
You guys think CRISPR research would be stalled if Gattaca came out around the same time it was discovered?

>> No.10244723
File: 128 KB, 1200x770, Ray_Gun_made_by_Taiyo.jpg [View same] [iqdb] [saucenao] [google]
10244723

>>10234070
AI. We'll (hopefully) find a better way of modifying DNA than CRISPR. CRISPR is just one method of modifying DNA. There have been similarly hyped DNA modification methods in the past. Let's talk about rayguns for a moment. Why do atomic rayguns look like they do? Well back in the 30s we knew that a lot of energy could be produced by nuclear power, but the only way we knew for mucking around with nuclei were big particle accelerators. These big particle accelerators used a bunch of high voltage stuff like Van de Graaf generators and Cockroft Walton voltage multipliers(which look sort of like said rayguns). So all these artists thought hey this is what stuff looks like for mucking about with nuclei today, so this is what nuclear power is going to look like in the future! But alas, we discovered better ways of mucking about with nuclei than particle accelerators. So I'm going with AI because it's more general.

>>being able to modify peoples genetic code in an expodentially more complex way as the tech advances
we still don't understand biology very well. New drugs and treatments have a 89% chance of failing deep into clinical trials. We lack a theory of how genotype translates into phenotype. We can't even determine which genes correspond to beak size in Darwin's finches. Some traits may take years to be expressed and evaluated, especially in humans. So it may take literally decades to determine if the gene modifications we make in humans have any advantages at all. Even if we do manage to make positive gene modifications, because humans grow slowly we won't see much of an effect for about 10 years minimum. Without a theory to predict phenotype from genotype, genetic modification will not progress exponentially.

>> No.10245105

>>10236947
>>10237009
It's CGI you mongs

>> No.10245111

>its another sci-fi escapism thread to go with the “get me a job” and “how do i get a cat gril sex slave on mars desusenpai???” and “climate changeniznawtreel” and “CS/Econ/Psyche is Science i swear” threads

>> No.10245134

>>10244723
>Even if we do manage to make positive gene modifications, because humans grow slowly we won't see much of an effect for about 10 years minimum.
but, theoretically speaking if human experimentation were to be allowed we could figure out most of this really easy, right? Like, hasn't anyone proposed just sacrificing 100.000 humans so that after that well have inmortality for the trillions of trillions that will come after?

>> No.10245199

>>10245134
>>if human experimentation were to be allowed we could figure out most of this really easy, right
no, it still takes about 10 years minimum to determine phenotype
>>hasn't anyone proposed just sacrificing 100.000 humans
in addition to being highly unethical, it'd be very expensive.
>> that after that well have inmortality for the trillions of trillions that will come after
so nematodes and fruit flies grow fast and there are practically no limits to experimentation on them, and we regularly sacrifice large numbers of them, yet we have not been able to make immortal fruit flies or nematodes.

>> No.10247484

>>10244723
Finally a well thought out answer, thank you.

>> No.10247579
File: 79 KB, 326x298, pat_kitty_.gif [View same] [iqdb] [saucenao] [google]
10247579

>>10245111
>how do i get a cat gril sex slave on mars desusenpai
But those are the best threads.

>> No.10248356

>>10234842
get can't it you why?

>> No.10248568

>>10245105
The CGI of today.

The human of tomorrow

>> No.10248586

>>10244723
Thanks anon, I always wondered what the source for those funky gun designs was.

>> No.10248615

>>10234205
>buying into science buzz this hard

Crispr will have no real impact on human life if used responsibly. outcomes of usage are much too unpredictable to ever be of use. scientists just keep the meme going so that they can get funding for meaningless projects.

>> No.10249019

>>10248568
He looks disgusting.

>> No.10249188

>>10249019
U jelly

>> No.10249248

>>10245134
Moralfags say no
We need second Josef Mengele

>> No.10249676

>>10249248
Pol/ type beat

>> No.10250271

How about smart AI completely understanding what genes are for intelligence/creativity/physical fitness/etc and using CRISPR to make a race of superhumans?

>> No.10250281

Nuclear fission will be the most influential for humans as it will cause a massive decrease in population very soon ;)

>> No.10250308

>>10250281
Here's hoping.

>> No.10250309

>>10234276
>>10234339
If someone creates a human-level AI then they can just throw exorbitant amounts of computing power under it until it does like 100 man-years worth of research on developing an even better AI in an hour.

>> No.10250345

>>10236853
So split the difference as many observers suggest, the nearer future is augmented reality, and teledildonics

T. Futurist

>> No.10250358

>>10250345
>Teledildonics
Hell yeah

>> No.10250457

I see you guys have never read Orion's Arm. The answer I learned from OA was : Why not both?

>> No.10250462

>>10250345
Pornmentation is a pipe dream. It's one of those things that seems like such a good idea only before you've had a wank, then you see it's overly elaborate because you were horny.

>> No.10250674

>>10236687
Someone's been watching too much battlestar galactica

>> No.10251369

>>10237873
>>10242382
>>10242460
It's a russian thot who uses photoshop to create her ideal man. Even worse, she photoshops pictures of her boyfriend to look like this guy. In one she even replaced his loser-tier snow jacket with a badass biker leather jacket. Imagine that, being cucked every day by an imaginary friend.

>> No.10251756

>>10250309
Assuming a. I of that capacity is even possible


It would cost a fortune

>> No.10252363

>>10251756
It is certainly possible. How can you explain your level of intelligence if it is not? After all you are a bag of flesh and water.

>> No.10252380

>>10239738
>what is Google Translate/DeepL
>what are self-driving cars
>what is image recognition
>what is speech recognition
>what is IBM Watson
I could go on

>> No.10252506
File: 1.79 MB, 320x240, 1541449879248.gif [View same] [iqdb] [saucenao] [google]
10252506

>>10236329

He's in chad stealth mode.

>> No.10252628

>>10252363
Superintellgent AI / the singularity is just a meme, to this day we don't know how our brain work, much less actually making a robot as complex as our brain.

Sorry bruh


https://www.google.com/amp/s/mjwrightnz.wordpress.com/2017/05/27/why-the-singularity-wont-happen-any-time-soon-or-at-all/amp/

>> No.10253130

>>10252628
>All this casts doubt on the notion of a ‘singularity’ in which human consciousness can be transferred into a machine, however much computing power or memory the machine has.
That's what the link that you shared was trying to refute. However, it doesn't give a reason to why intelligence cannot be created. Next time read the articles that you share yourself.

>> No.10254334

>>10253130
Fucking brainlet.

Nobody said intelligence can't be created, but not at that magnitude


Just stfu
https://medium.com/@tobywalsh/the-singularity-may-never-be-near-610988dbd34

>> No.10254846

>>10252506
Stealth chad

>> No.10254891

>>10234070
>one MacBook Pro, please

>> No.10256544

>>10254891
No.

>> No.10257043

>>10251369
Getting cucked by Photoshop....


Nigga damnb

>> No.10258868

>>10234070
Bump

>> No.10258894

>>10234205
Biomedical engineering is also transhumanist, idiot.
Transhumanism isn't a synonym for cybernetics.