if "AI" is so smart then where are all of the breakthroughs? where are treatments for cancer and alzheimer's? where are room temperature superconductors? where is a usable linux desktop? where are self-driving cars and humanoid robots? where is the solution to the JQ?>navier-stokesautistic nonsense that changed nothing.
piss off
>>62746959this idiot convinced himself that a probability weighted sentence constructor is an "intelligence."
>>62746953ai is a kike scam
>>62746962Whatever you say materialist
is shorting anthropic easy money?
>>62746953AI has made enormous efficiencies in labor and manufacture. Haven't you noticed how the price of everything is down a lot?...wait
>>62746953Nigger, do you leave under a rock?They solved a decade long protein folding issues in a few days.The moderna skin cancer vaccine was with help of ai, many other cancer vaccines under development are also using aiIf you are a human with long term planning , invest in ai.
>>62747161>They solved a decade long protein folding issues in a few days.use case? sounds like more navier-stokes nonsense.>moderna skin cancer vaccine was with help of aisounds exactly like structure based drug design (an existing technique) by another name. companies rebrand everything as AI because it attracts the attention of 50 IQ nigger retard investors who click "buy" whenever they see BigCorp using AI™.>On the design end, a series of fully integrated AI algorithms takes next-generation sequencing data from tumor and blood samples, reviews their genetic mutations, and predicts up to 34 of those neoantigens that are most likely to elicit an immune response. https://www.modernatx.com/media-center/all-media/blogs/advancing-fight-against-cancer
>>62747180>use casevaccines and antibiotics
>>62746953>where are self-driving cars and humanoid robots?My car has full FSD. I plot in where I want to go and it goes there even if I fall asleep.Look up the chinese robot olympics. The developments in robotics has shot through the roof in the last 5 years. Some of the biggest companies in the world are currently pumping out a shitton of humanoid robots. You could buy a Unitree humanoid robot right now for like $13k and control it remotely. If you can't imagine how militaries across the world will utilize that technology you are the biggest smoothbrain on the planet
>>62747197you didn't meaningfully answer my question.
>>62747180The use case? We need specific protein structures to “insert” in cells to trigger specific conditions. Things that could have taken extreme amount of manpower (the R&D process) can be done rapidly using Ai. This means that you going to see literal miracle drugs rolling out and basically clog the entire FDA, until they upgrade that with AI.
>>62747283>Things that could have taken extreme amount of manpower (the R&D process) can be done rapidly using Ai.you don't understand drug research. research SBDD.
>>62746953The cancer treatments are unironically coming. 5-10 cancer will be almost totally cured.
>>62747378I hope you're right, anon. Cancer is scary. I'm not that afraid of dying but a shitton of people in my family have suffered cancer and even if you make it out, it's fucking hell.
>>62746953>where are all of the breakthroughsNot for the blind to see.
>>62747101The more efficient we get the more expensive everything isWe will all be homeless in 2030 because getting a job will be impossible and bread will cost a 100 dollars
>>62747390My dad has cancer. His doctors are some of the best in the world. They say 5-10 many cancers will either be cured or chronic non-terminal illness.
>>62747319Literally worked In labs before anon.I understand enough to see AI as fucking crazy significant. Medicine will become miraculous and we would have cure for everything, within your life time.>>62747390You are right, it’s not death, it’s the agonizing pain, and the random chance that you might sleep and tomorrow have a little tumor growing in you, even if you did everything right.It’s the just world fallacy on its head. That’s why it so scary.>>62747544Yeah, customized cancer vaccines are on their way; so also customized skin care products, customized vitamins, and even customized enhancers.I actually wanted to bite to bullet and hop on steroids for gains, but given that in 20 years we might be able to pass the longevity escape velocity, I rather not risk it.
It's going to cure our economic zone by putting the entire economy on AI controlled blockchains
>>62747568>Literally worked In labs before anon.in what capacity?
>>62747636I worked both on cancer cells and regular animal cells.And true, in vitro is not in vivo. But so what? Most of the things said can be super simple to run with ai. With AI, We can ensure that the animal testing phase almost always passes without any issue, and once animal testing passes, human testing becomes just a matter of ethics and long term side effects. No one said every medicine gonna go like covid with “speed of science TM” but you can be your ass, that it will get to human testing phase super fast.>>62747544Also, reminder that skin cancer pre 2014 was an actual death sentence with 80% chance of mortality within 5 years, and by 2015 it became almost trivial in comparison, with only 20% chance of death. It’s crazy how one medicinal change (something that can be way easier with AI) can have such drastic effect
>>62747202>My car has full FSD. I plot in where I want to go and it goes there even if I fall asleep.no you dont
>>62748236>I worked both on cancer cells and regular animal cells.you didn't answer the question. were you a $15/hour lab tech or were you a $200k/yr PhD scientist at a major pharma lab?>We can ensure that the animal testing phase almost always passes without any issue, and once animal testing passesi just answered my own question, never mind.
>>62746953I don't like this aids skrillex pepe he looks like he smells
>>62746953They’re coming dude they’re definitely coming, they just need a few more trillion in market capitalization and they’ll get there I promise
>>62748270>>both cancer cells AND regular animal cellsNo need to ask, clearly just an undergrad who spent a semester volunteering in a lab.>t. working with AD mouse modelFuck this disease, it’s fucking impossible to work with in vivo and even when you get good results it barely translates into clinical care.
>>62748236>And true, in vitro is not in vivo. But so what? Most of the things said can be super simple to run with ai. With AI, We can ensure that the animal testing phase almost always passes without any issue, and once animal testing passes, human testing becomes just a matter of ethics and long term side effects.Holy kek the level of ignorance on this one. Most mouse results don’t translate to human results, let alone in vitro translating to in vivo, let alone in silico translating to in vitro. I’m sure AI is going to help accelerate some aspects of drug discovery, but even the CEO of Lilly said it himself that the impact of AI has been hugely overstated. I work at a top-10 pharma company and so far there’s no dramatic results, just a lot of yapping and spending
>>62748325How do I transition into pharma form academia anon, I can’t spend another year in this doomed field. Where should I look for internships?
>>62746953this pepe looks like JA Westenberg
>>62748385I’m really not great for advice, I’ve stumbled my way through my entire career so far. But I would just say, focus on building employable skills if you can. Analytical jobs like mass spec and HPLC always exist but they generally suck afaik. Just don’t get pigeonholed into a niche area of science and you should be ok. What are you working on now?
>>62748385>pharmado you like to scam people`?
>>62748303>it barely translates into clinical care.That's because the models aren't real anon. They are all overexpressing amyloid/tau to comical levels (3xTg LOL) just to see some "AD-like" symptoms. Not even the mutations that cause 95+% of AD cases (APOE4). If you are even using transgenic mice and not just straight injecting amyloid. You are modeling amyloidosis/tauopathy/extreme brain proteinopathy. No animal gets true AD in the way it develops slowly and sporadically in humans over 50+ years. If anything, Alzheimer's research might be THE field where AI (as in ML and bioinformatics) is actually useful, because we can stop FAKE in vitro/in vivo work and instead focus on REAL human data. Not trying to discourage your work, but this seems to be the general state of the field atm. Enlighten me if I'm wrong.
>>62746953he doesnt know
>>62748855>phone screenshotit's like visual ebonics.
>>62746953You're assuming the goal is to improve your lifeThat's an interesting assumption given the fact that your comfort has never been a priority before
>>62748869oh sweet honey boo boo summer child
OP still unrefuted
>>62748325Yeah yeah please keep this up. Pretend that I think mouse results 1:1 works on humans even though we both know animal testing 100% need to be verified on humans. That’s all I am saying. In silico make it more likely to Design medicine that can pass in vivo easier, however that doesn’t mean that in silico going to replace human testing anytime soon. We still gonna have that 10 year human test phase, but instead being some minor medicine it’s gonna be either custom ones or really miraculous ones that were created by AI.
>>62746953in terms of medicine, personalized highly granular treatments (like targeted CRISPr, specialized mRNA vaccines, etc) could be where AI shines, since in theory it will one day be trivial for a trained LLM to look at your genome and biomarkers and generate whatever you need. Such as a course of regenerative vaccines that will unfuck your aging organs without giving you cancer. With AI, the cost of this kind of analysis would be very low. But this is still scifi stuff, the treatments themselves do not exist yet, and AI cannot really accelerate the testing process, it can only really make suggestions for things to be tested.
>>62746953Medicine is a practical field, you can't solve it in a lab. Not until human body is 100% digitalized, then AI can solve everything. Not happening any time soon though
>>62748855Since I'm not sure where else to ask: What would you recommend for starting to understand agents and working with AI in a more project command-like capacity?
>>62749085Yeah yeah please keep this up. Pretend that you didn’t say exactly what you said kek
>>62748239You do have to keep your eyes open and occasionally look out the windshield, but you can prop up a book and read the whole time. It's totally capable of self driving once the nanny police back off
Timeline of AI Derangement Syndrome (AIDS)2022 >yeah its cool and all but it gets a lot wrong2023 >ok its wrong a lot less but its not really a professional tool2024>ok it's a professional tool but it's not going to take my job2025 >ok it took my job but it won't take the hard jobs2026>ok it took the hard jobs but it didn't heckin cure cancer and solve nuclear fusion therefore it's heckin useless!!!!
>>62746953There are none because it's only good for monkey work like summarising emails
>>62747283No, we won't.In 20 years people will still be bald, infertile, have cancer, have hiv, be weak, be old, have diabetes, be paralysed, have small dicks, have bad tit and flat assesAI is solving nothing. It's just another scam, like musks space x crap
>>62747568>, but given that in 20 years we might be able to pass the longevity escape velocity, I rather not risk itCongratulations, you are a Reddit tier retard
>>62749509Except we're at>People won't shut the fuck up about this new predictive text that constantly gets things wrong and talks like a faggot
