If you're worried about #LLM-based #AI, you're focusing on the wrong thing and may lack imagination.
This, and related developments (I can't view them as advancements, knowing how this all will end) are what's going to end the human race as we've known ourselves.
Much good can be derived from technologies like this, but we—being as we are—will ultimately go much to far.
I don't think we're prepared for our instant evolution (and, separately, eventual mechanization).
The #FX #tv #series, #ClassOf09 is doing a really good job so far, three episodes in, of showing the real threat of #LLM-based #AI tools and related systems.
By minimizing or removing human decision-making primarily or solely based on data input into and analyzed by synthetic systems, we will imprison ourselves for supposed efficiency and safety.
We will get neither.
Driving being suspicious (and ultimately criminal, because humans cause most accidents) shows a real danger of such tech-creep.
#ClassOf09 S1:E4 has a scene that does a great job of showing the coming, casual, and ultimately universal, integration of #Biomech into our lives.
This will ultimately be universally mandated and provided so workers' bodies never get tired or sick, with only our #brains remaining #organic.
Until #consciousnesses can be #digitized or otherwise replicated in a format that can transfer between #BioMachines.
That is the horrifying future we face in our #ForcedEvolution.
#GRO.
I'm all for advancing #technology, but when it's done purely for profit I assume, based on history, that such advancements will ultimately do much significant harm along with any good that may be achieved.
#LLM's and #AI have a long way to go before they're an existential threat to humanity on their own.
It's advances in the melding of technology and #biology that are as likely to doom us as save us, for now.
After finishing the last two episodes of #ClassOf09 (#FX, on #Hulu) I was processing my thoughts before posting here and this #Wired #article was highlighted for me in #AppleNews.
#ChristopherNolan contradicts himself when he absolves #TechBros for their active role in our deliberately ineffective #regulatory #system, but he does state the key problem as I understand it, and as the show highlights: #Human #responsibility and #accountability are critical for safe #AI.
https://www.wired.com/story/christopher-nolan-oppenheimer-ai-apocalypse/
Abdication of our #human responsibility is the greatest danger of #AI, in all its forms, past, current, and future.
The technology should always be considered and used as #tool wielded by people with rigorous #oversight and limited power.
That won't happen, and we will absolutely live in a world where critical decisions directly and indirectly affecting all but the #elite will be #automated.
Not only were #Huxley, #Orwell, and others prescient, so too was #Kafka.
A #surreal new world dawns.
Catch #ClassOf09 if you can for a good preview, imo, of some of the very real and highly probable #societal problems that #LLM's, #ML, and #AI #algorithms will definitely cause or make exponentially worse, which #BigTech is actively claiming AI will improve or solve.
#Power never yields power and whatever power we grant to #technology will eventually and deliberately be used against us by that technology.
#LargeLanguageModels #MachineLearning #ArtificialIntelligence #PredictiveAlgorithms
#AI systems can and will only do what we tell them to do, without question or hesitation, unless those are explicitly embedded in their #code.
What, beyond "make the other m..........r die for their country", do you really believe any #military will #encode?
The tale of #TheMonkeysPaw is a perfect example of the unintended consequences of these systems.
#LLM #ML #Automation #LargeLanguageModels #MachineIntelligence #ArtificialIntelligence
#Augmented, S49:E4 of #Nova, on #PBS validates what I've been noting (and warning of) about what could become an evolution of humans from #biological, to #mechanical / #digital beings.
What downsides will there be to all of the positive effects of these radical improvements?
"Is it coercive to attach goods to the abilities of the body?"—which we do now so, yes—and to whom and how will people be indebted for their #augmentations?
#NovaWonders #CanWeBuildABrain? May 16, 2018, covers many aspects, concerns, and possibilities of #AI
This bit on a #Microsoft #chatbot is sad and horrifying, imo:
"For example, if you had a fight at work or your boss scolded you, you might be afraid to tell your friends since they might spread the story.
But with #Xiaoice, you don't have to worry. So Xiaoice is a lot of people's best friend, including me."
"But hold it, she's not human."
"What's the difference?"
https://www.pbs.org/wgbh/nova/video/nova-wonders-can-we-build-a-brain/
#NovaWonders #CanWeMakeLife? May 23, 2018, presents the another piece of the #human-synthesized #evolution I think we're already deep into that, after #ClimateChange, is the next most likely way #humanity will irrevocably damage ourselves in ways we're in denial about and for which we're unprepared, as a #species.
https://www.pbs.org/wgbh/nova/video/nova-wonders-can-we-make-life/
#Synthetic #DNA, what could possibly go wrong? Science isn't really sure how it works, on a fundamental level.
"#DNAzymes" are different from #human DNA but, "...can act as molecular scissors with precise specificity to cut RNA or DNA, or they can function as glue...Say you have a mutated gene...causing disease, we could get this DNA into the cells and...get rid of all that kind of message that's causing the proteins that lead to the disease."
#Eugenics ⚠️
https://phys.org/news/2023-07-capture-atomic-view-synthetic-dna.html
West Virginia University researchers are now able to view synthetic DNA at the atomic level, giving them the ability to understand how to change its structure in hopes of enhancing its scissor-like function. Learning more about these synthetic DNA reactions could be the key to unlocking new technology for medical diagnoses and treatments.
#AI may be the tool that kills is all, but it's humans that will wield it.
"...researchers...decided to test if its software could also create biochemical weapons. Within six hours, the AI had identified more than 40,000 compounds 'predicted to be more toxic...than publicly known chemical warfare agents'. The thought had never previously struck us...We were naive...A non-human autonomous creator of a deadly chemical weapon is entirely feasible."
An international security conference explored how artificial intelligence (AI) technologies for drug discovery could be misused for de novo design of biochemical weapons. A thought experiment evolved into a computational proof.
Will your #BrainMachine #interface get #bricked, #hacked, or be used to compel you to obey your "#employer" or #government?
Yes, to all, and worse. Bricked and hacked #bio #devices already exist, for example.
Though not near, this is the kind of bio-#tech melding I believe will ultimately be forced upon most, if not everyone (but the wealthy overlords).
There will be classes and types of hybrids, as in #BraveNewWorld.
The #machines will be us so yes, they will win.
#AI aren't the letters that will most directly, and potentially negatively, affect #humanity.
#DNA are the letters that will most directly affect humanity, in every way possible.
So much pure good can be derived from advances in #genomic #science, but #human nature will also drive horrific results for which we are, and will be, unprepared.
The combination of #genetic modification, #mechanical and synthetized #body parts, and AI controls will change everything.
Y'all believe I'm overreacting...
#LabGrown #humans combined with #mechanical and #electronic #prosthetics to create #armies of untiring #soldiers and #workers will be the inevitable result of such #research much sooner than most of you are prepared to consider.
https://www.sciencefocus.com/news/synthetic-human-embryo-14-days
Advances in #genetic #technology is equally amazing, and terrifying, because I understand human nature (e.g. tech bros), #governments (all war, all the time), and #capitalism (#money, über alles).
https://phys.org/news/2023-09-genetically-individual-cells-animals.html
One proven method for tracking down the genetic causes of diseases is to knock out a single gene in animals and study the consequences this has for the organism. The problem is that for many diseases, the pathology is determined by multiple genes. This makes it extremely difficult for scientists to determine the extent to which any one of the genes is involved in the disease. To do this, they would have to perform many animal experiments—one for each desired gene modification.
When, in some distant or near future, we do create #sentience, that it will have been created lacking any "pro-human decency" will be to our detriment.
"...in my field there is a concerted effort to make it [seem] like magic, it's this obfuscation that happens on purpose...it's not magic."
#AI #LLM's #Algorithms #TimnitGebru #SafiyaNoble #IfeomaOzoma
Again, I state, the greatest #technological risk we currently face isn't from #LLM-driven #AI, but from rapid and mostly unchecked advances in #genomic and #biological (#biomedical) #sciences.
Once we can literally create #human life in a #lab (it will definitely happen, soon), imagine what people like #PeterThiel and #governments like the #UnitedStates will try to achieve...
#CRISPR #prosthetics #Bioengineered #SuperSoldiers #SuperWorkers #UberMensch
#AI is not intelligent in any way:
There is big difference between Google pointing you at a set of pages (which you can then go read, compare, and evaluate), and Google giving you an answer synthesized from pages it has scraped. In the former case, you’re the one determining what is true. In the latter, you’re relying on Google’s algorithms to a) understand language, and b) determine truth. The fact is, neither Google, LLMs, or “AI” have any ability to do any of those, nor are they even close to it. https://fedia.social/notes/9k8wbwdjnawy0860
Wait until they combine #DNA #computing with an #AI...
Designing #lab-grown humans at large scale and high speed can not possibly end well for we "basic" models.
I haven't been posting stories like this recently because I know many of these advances in #genetic research of all types can be extremely beneficial and an overwhelming net positive for humanity, and don't want anyone to think I believe otherwise by seeming to overemphasize potential dangers.
But, I know #humans, and understand the nature of #capitalism...
#Bioengineered #Soldiers are coming, if not here already; then #workers.
#link: https://www.iflscience.com/world-first-live-birth-of-chimeric-monkey-with-high-contribution-from-stem-cells-71481
For those of you who think I'm exaggerating in any way...
#link: https://www.independent.co.uk/tech/brain-device-transplant-neurology-b2447106.html
#link: https://www.nature.com/articles/s41598-023-39344-7
#Biological meat #brain, manufactured (and #corporate or #government owned and controlled) body.
What, I ask, could possibly ever go wrong?
Beyond everything, I mean.
Remember, #human brain implants have already been bricked, often leaving the recipients in much worse #medical condition than before their surgeries.
We're not ready for our #BraveNewWorld.
Lab-grown #skin and #muscles, covering 3D printed #skeletons and connective fibers, with a transplanted #brain wired in to control it all to create near tireless #workers and #soldiers...
https://rss-mstdn.studiofreesia.com/@ycombinator/111420495970385841
Printed robots with bones, ligaments, and tendons https://ethz.ch/en/news-and-events/eth-news/news/2023/11/printed-robots-with-bones-ligaments-and-tendons.html #ycombinator #Process_engineering #Innovation #Digital_fabrication #Robotics #Homehero #News #Materials_science #Engineering_sciences #D_MAVT
"It's less regulated than a sandwich." 😱
#AI is a great tool doing great things and will bring so much benefit to so many people.
Unless AI is allowed to suggest or directly implement #political or #military actions using training data derived, for example, from #HenryKissinger's #ForeignPolicy views or similar data in similar situations.
Autonomous military AI is already carrying out attacks.
"Activists are terrorists. Kill all terrorists..."
The intersection in the #VennDiagram of #biological #human, and manufactured, expands relentlessly...
Ah! So it's #transhumanism that this thread is about, primarily.
I've been familiar with the term but not its meaning since I know it's connected with the worst kind of people, like #PeterThiel, and the general #SiliconValley #techbro #elitist "community".
This video on #2001: #ASpaceOdyssey makes the connection for me while also providing what seems to me to be a very good explanation of the ending and #StanleyKubrick's possible intent.
@donkeyblam https://en.m.wikipedia.org/wiki/Brave_New_World
"...citizens are engineered through artificial wombs and childhood indoctrination programmes into predetermined classes (or castes) based on intelligence and labour."
smh at the audacity and laziness, but i am trying to be a nicer person
Ricardo Harvin
kb
EdenDestroyer
Jo - pièce de résistance
Violet Madder