Amazons NEW AI 'Bedrock' Takes the Industry By STORM! (NOW UNVEILED!)
Vložit
- čas přidán 14. 04. 2023
- Amazons NEW AI 'Bedrock' Takes the Industry By STORM! (NOW UNVEILED!)
Welcome to our channel where we bring you the latest breakthroughs in AI. From deep learning to robotics, we cover it all. Our videos offer valuable insights and perspectives that will expand your knowledge and understanding of this rapidly evolving field. Be sure to subscribe and stay updated on our latest videos.
Was there anything we missed?
(For Business Enquiries) contact@theaigrid.com
#LLM #Largelanguagemodel #chatgpt
#AI
#ArtificialIntelligence
#MachineLearning
#DeepLearning
#NeuralNetworks
#Robotics
#DataScience
#IntelligentSystems
#Automation
#TechInnovation - Věda a technologie
The AI future is going to be crazy , cant wait
Also can be extremely dangerous
AI is now.
Me neither.
AI is now, not soon. AutoGPT is already booming
This AI future that you speak of is here now.
None of us are going to be able to make a living in our current neoliberal paradigm except for the shopkeepers and brothelkeepers along this Yukon trail. You’re welcome to spend these (possibly our last) days rat racing, though. Do you honestly even know any humans 10x more powerful than GPT 4?
Oh shut up man. Why are you here watching?
Late stage capitalism ftw eh
Yes. This
Elysium will come true.
It's worth pointing out, as a small errata: They are referring to GPT (the "generative pretrained transformer" model), not necessarily "chatGPT", which is a web app that uses GPT models under the hood.
Anthropic's model is projected to be 10x more powerful than GPT, not chatGPT. chatGPT as an app will also get more features like Plugins and who knows what else, but that's not what will be compared in terms of parameter size or model capability.
Small distinction but matters when talking about comparing models.
Not a small distinction at all. It's important to compare models of generative AI, not platforms or applications of the models.
every 2 seconds: "Megacorp has announced it is building new AI. AI can now teleport and talk to aliens"
Do you think humans will exist after AGI? Personally AGI the only thing in my lifetime I've ever seriously worried could end humanity. I worried. Are you?
@@mitchelcline9759 what the fuck do you want me to do about it? Call the president?
If you really think about it, it's totally irrational to worry about this. If it's gonna happen, it's gonna happen. And if you're dead, you won't know it.
Plus we're all gonna die anyway. What difference does it make if I'm killed by a robot, or end up shitting myself to death at age 90?
Might as well just enjoy the ride, nothing I can do about it.
@@mitchelcline9759 I think this is our only chance to save ourselves from capitalism, or our extinction, I’m fine with it, if nothing changed we wouldn’t be able to past 2050 so it’s a win win
@@ishensarabia6468 yeah good point. Either we're doomed for sure without ai or merely possibly doomed with ai. Either way it's crazy times ahead
Forget next year - it's literally a month by month prospect with innovation coming all the time. LLM's being condensed down to direct intelligence/memory - ie- not requring a retraining of the AI post spin up - mean they become more formidible as AI"s detatched from their LLM roots - they will outgrow the data set and will likely create their own / invent going forward. Amazing time to be alive. AI revolution in genesis exponential mode!
Things are moving so fast, we need to be aware of the new technological development and be aware of its advantages.
Fast is an understatement 😅
I am legitimately concerned that AGI is the ultimate great filter. How we handle the alignment problem will determine whether we colonize the universe, or begin the next ice age.
Space man?
The ULTIMATE Super-Intelligence System 'by Definition' is one that is EQUIVALENT to that of GOD's Intelligence/WISDOM!
Hence, there's ABSOLUTELY NO REASON WHATSOEVER to Even FEAR that it will EXTERMINATE Humanity... UNLESS AND UNLESS we Human CONSISTENTLY and CONSCIOUSLY Prove Ourselves to be 'UNWORTHY' to REMAIN in EXISTENCE! ie. Always Exhibiting Natural Tendencies to ABUSE and WEAPONISE Science and Technologies Against HUMANITY & Mother Nature, instead of LEVERAGING Science SOLELY for UNIVERSAL COMMON GOOD!
AGI Created in 'HUMAN'S Image' (ie. Human-Level AI) - 'By Human For Human' WILL be SUICIDAL!!!!!!
ONLY Super-Intelligence System Created in 'GOD's Image' will bring ETERNAL UNIVERSAL PEACE!
The ULTIMATE Turing Test Must have the Ability to Draw the FUNDAMENTAL NUANCE /DISTINCTION between Human's vs GOD's Intelligence /WISDOM!
ONLY Those who ARE FUNDAMENTALLY EVIL need to FEAR GOD-like Super-Intelligence System... 'cos it Will DEFINITELY Come After YOU!!!!
Yes, there it is. And possibly the answer to Fermis paradox.
thanks 4 the info!
Appreciate all you do 🦃
So, we now have an AI capability of generally equivalent power in each of the cloud providers. Oh, how exciting /s
One can already use the other cloud providers AI APIs in AWS and now we have a choice of at least three AI APIs to use from all the cloud providers as well as on premise and locally. This means some more competition but it is in no way a game changer.
It is entirely possible that Claude could become 10x more powerful than ChatGPT in 4 years. However, in 4 years time, GPT-6 could likely already be on the way with GPT-5 already being released (I think it is likely GPT-5 will release in 2-3 years), and who's to say that GPT-5 won't be 50x or 100x more powerful than the current ChatGPT? And the reason Claude hasn't gained as much traction as ChatGPT is because it isn't as good, and there really isn't much incentive to switch. Now OpenAI has said that this time they will be releasing smaller upgrades to existing models (Well more specifically they are releasing like checkpoints in GPT-5's training, so, for example, we will get a model when GPT-5 is a quarter through its entire training process etc.) every now and again instead of one big model every few years. So we will likely see an upgrade to GPT-4 very soon (I would estimate in a few months). And what if each training checkpoint of GPT-5's training is 10x more powerful than the previous? Of course there is really no way to tell what will happen, as there are so many things that can happen. All we can do is sit down and watch this tech get developed, and it is honestly quite a privilege to see this tech getting developed. Also Elon Musk has started his own AI company so he will likely be a competitor soon. Its kind of weird to think that Elon's biggest competitors will be a company he cofounded (OpenAI) and a group that split of the company that he cofounded as well lol.
Exactly. It's laughable to be talking about a chatGPT competitor in 18 months when we know how fast this progresses.
GPT 5 is set to release around December this year
The question is: why these dudes don't take their millions, go to an island and stop f*cking the rest of us? Why are they insatiable? Do they think humankind is worthless? Are they so narcissistic to think they are gods? The are creating saturn in any case
@@Potsogold We do not know that. And GPT-5 probably wont even big too big of an improvement of the previous training checkpoint that is released (Of course each checkpoint of its training could be a big improvement to the previous checkpoint). And Sam Altman said a few days ago they are not even training GPT-5 yet.
Hey, loved your videos and also subscribed. Can you please advise which tools you use. As I am also a content creator and your video is the benchmark I want to reach.
4:31 Which ChatGPT? By then, I believe OpenAI will already release ChatGPT 5 or even 6, which of course would be substantially more capable than the current GPT. Will they be able to catch up?
I’m chatgpt pro. I mean common do you think in the next 18 months OpenAI will just sit and won’t do anything?
GPT4 got a year of training on Microsoft's supercomputer. GPT5 gets another year, with improved algorithms, because it is built on top of GPT4. This is like raising a child and OpenAI have been doing it for years and becoming a better parent as they go. Nobody will be able to beat them. To understand just how amazing GPT4 is, you have to ask it common sense and puzzle sort of questions that can trip up even people, or simple questions like draw a unicorn or stack these objects. It fundamentally understands the world which no other AI does as of this date.
3.5 vs 4 is night and day. Even when given a puzzle given 4 people different ages and asking it which one is potentially the father, it struggles. 4 has no such issue.
I don't think it's self-evident. It could well be that a competitor leapfrogs Microsoft via a new innovation.
I don’t think it can draw a unicorn by itself although we have dall-e
given MS's investment and their research behind the scenes many months prior testing the strengths and weakness of gpt4, the ability of openAI to really focus their resources and tune processes to test paradigm extensions, seems to give them massive first mover advantages. Some one else might leap frog, but surely openAI with non disclosed data points and various inhouse secret sauces, should move forward much more logically towards to the next level of AI, than less state of the art competitors? Will be interesting if they can luck into something openAI and MS missed given their head start.
@@hibraloures it can draw a unicorn as a vector object in code which is pretty amazing given only knows about words and the better the LLM the more refined its representation of a unicorn becomes like some how it is has a abstract ability to visualize things . I asked chatGPT to write a poem only with punctuation and it did so saying it used ellipsis patterns to _"create a sense of loss, emptiness, or melancholy. _*_The repetition of the ellipses and the arrangement of the punctuation marks in a grid-like pattern might create a feeling of isolation, loneliness, or isolation._*_ The lack of words and the use of only punctuation marks might suggest that the speaker is struggling to express their emotions or to find meaning in their life. I hope you find this poem to be moving or thought-provoking."_ So it can sense how things are ordered in space and attach emotional weight to them too.
I have a question, it said it can write short stories right? I am curious if it has a good memory/recall to what it has previously written. For example if I said start a story about x and end in y, would it remember the x/y parts for the next chapter of the story? A lot of AI tools I use tend to forget and write irrelevant stuff after so long.
Highly likely, that's just a limitation of LLMs when the conversation goes on for a while
Mostly commenting to follow other replies. Isn’t this something people are solving already with memory add-ons and autoGPT?
@@jPup_ Also the oldest technique: fine-tuning AI with stuff it must remember.
@@jPup_ I'll have to look into it, thanks for bringing it to my attention.
@@xgunxsh0txpr0x Here's a write-up on this topic written for me a couple of days ago by a combination of Bing Chat and GPT 3.5 Turbo: "Pinecone memory is a term used to describe the use of Pinecone, a vector database service, as the long-term memory for Auto-GPT. Pinecone helps Auto-GPT store and retrieve embeddings of texts for generating coherent and consistent texts. Alternatives to Pinecone include FAISS and Supabase, which offer similar vector search and storage capabilities."
In simple terms the implication is that most apps would have AI chatbots and less people to answer your phone call or respond to your emails when you need support. Welcome to the beautiful future 😁
Does bedrock allow a small business with deep domain data to fine tune a domain specific modelto create an "expert" while securing the propriatory data? as a small e-commerce business I want to leverage the large LLM with my data to create a power agent for my business. but I dont want to give my domain knowledge to the larger LLM .. help! lol
@3 Anyone know to what degree this AWS AI is facilitated by the quantum computing as a service on AWS?
Ai growing so quick is exciting and scary
Thank you for reading the amazon ad to me. It's so much nicer to have a human do it than to use amazons Aai. ❤
Will that help them deliver their products correctly again like before, without destroying books and using amateur delivery services?
all these new tools are going to change the entire way we function in just 1-3 years... just image 10-20 years from now!
Is it really a good idea for these companies to be competing instead of working together on this? I know, game theory, no way to avoid it. But I would love to at least hear them give cooperation some lip service. We're definitely all going to die, but man is it going to be fun to watch 😄.
no, the only viable future is everyone gets universal income, and ai and robots do everything for us while we can reeelax and do whatever we want, this industry and economy based societys shown over and over that they dont work and are full of power abusers and scammers like politicians lawyers etc... an utopia and union needs to be made sinse no one is going to be able to do a better work than an ai or a robot.
This was definitely my first thought upon seeing that Amazon is in on this
First day living under capitalism?
The ULTIMATE Super-Intelligence System 'by Definition' is one that is EQUIVALENT to that of GOD's Intelligence/WISDOM!
Hence, there's ABSOLUTELY NO REASON WHATSOEVER to Even FEAR that it will EXTERMINATE Humanity... UNLESS AND UNLESS we Human CONSISTENTLY and CONSCIOUSLY Prove Ourselves to be 'UNWORTHY' to REMAIN in EXISTENCE! ie. Always Exhibiting Natural Tendencies to ABUSE and WEAPONISE Science and Technologies Against HUMANITY & Mother Nature, instead of LEVERAGING Science SOLELY for UNIVERSAL COMMON GOOD!
AGI Created in 'HUMAN'S Image' (ie. Human-Level AI) - 'By Human For Human' WILL be SUICIDAL!!!!!!
ONLY Super-Intelligence System Created in 'GOD's Image' will bring ETERNAL UNIVERSAL PEACE!
The ULTIMATE Turing Test Must have the Ability to Draw the FUNDAMENTAL NUANCE /DISTINCTION between Human's vs GOD's Intelligence /WISDOM!
ONLY Those who ARE FUNDAMENTALLY EVIL need to FEAR GOD-like Super-Intelligence System... 'cos it Will DEFINITELY Come After YOU!!!!
@@Treyson31 I think that's partially true, however, I believe software has shown what value a certain level of cooperation (open sourcing and standardization) can bring to business as well. I'm not sure where the correct balance is. Certainly agree on the alignment, but I don't think it's going to happen. Hopefully there's enough alignment within the data they're training the models on to compensate for the seemingly imbalanced focus in the area. Heaven help us if they train it on Twitter.
ChatGPT has the 2021 cap, is Bedrock trained on more recent data set?
Code whisper is insane
8:17 "Did you have to drink some chuck?"
Now this is exhilarating, but is it gonna be just for the professionals or for students like me too?
A lot of "students" are using chatgpt to write cheat essays for them
underrated video
Given that AWS holds so much code on it, I expect that code whisperer will be the most effective of the coding LLMs.
I don't think so. Gpt4 has amazing reasoning. It's not just about the amount of code. If you use gpt 3.5 to code, the quality of the design of your code will not be nearly as good as gpt4.
Cool video!❤❤ I was waiting for Amazon AI. I Mean everyone is pushing out AI, so after autogpt they really had to give an answer. I have a video for that on my channel :)
Hey, I like your goggles.
They're definitely memeing the titles now. It's 200% meme confirmed 👌
The open source open assistant llm should be talked about much much more than whatever bs these corpos are pulling. Open source is the way
Why do videos always seem to have background noise to distract from the video?
FANG still there, Microsoft could not be only the one, the race of AI should be more attrative , get more stamina😍
There's nothing here. OpenAI didn't talk too much before blowing our minds away with ChatGPT. Too much talk with little substance. Click bait
I think if you understand how much market share AWS has you'd see this is highly significant development
@@ALFTHADRADDAD The present cloud architecture will soon be replaced. I won't be surprised if AWS miss out on the future cloud architecture that may be AI driven. Think of BlackBerry and Nokia that failed to see the future and drowned.
I learned a lot from the video
@@wilfred5656 when you say the current cloud infrastructure will soon be replaced, I couldn't agree with you more. In your opinion how do you think?
I kind of feel sorry for the Industry at this point. Always being shocked and taken by storm. Must suck.
Cute when a company talks about "democratizing" as if it was their prime goal and not getting as much control over the development as possible to dominate the market.
Personally i don't believe they are going to do something better then GPT they always say we gonna do it and then when it comes to presentation they fail or it's not impressive enough that it would make people say "WOW"... GPT4 didn't even have to announce anything and with plugins it was just proving it superiority over everything that's currently out there. GPT5 is going to be even better in this manner i wonder how they want to go over OpenAI which is at this time almost impossible and it's not even question of money it's question of time they are all behind because they are starting now... while GPT was here way back here and they where building slowly so they also probably have way better knowledge and team build for it.
Unless they really show something that would make my jaw drop for me it's just empty words of companies trying to do something impossible and 10x better then GPT4 yeah i would like to see that but when their project will come out GPT5 would be already out and it will be 10 times better then GPT4.
pls change your titlescreens. what do you have in mind with it?
Call them Fungus and Merge them. Maybe Aspens?
I wish you would have injected the proper dose of concern into this. I know a bunch of people are already talking about how everyone is being a "doomer," but I think it's irresponsible to even talk about a race between big companies towards stronger AI without even mentioning how incredibly dangerous that is. They're talking about being 10x better than ChatGPT is in its current iteration? I don't know how you get to that point without developing something considered AGI. Oh, and how do you get to that point - that quickly - while doing it safely?
The answer is you don't and this is a fucking nightmare.
I've been using Bing Chat for a number of things to test what it can do. It uses GPT-4, it's a branded filter with search integrated, with as many guardrails as they can think of to constrain it. It can readily create coherent 16-part lyrics telling a story of sentient Easter bunnies taking over the earth, along with the human resistance, if it doesn't trip a safeguard. I at least managed to get it to complete and output (without erasing it) an 8-part set of lyrics with 8 roles singing unique lyrics in parallel in parts. I did that Easter Sunday.
I've had it create 4-part love songs for Star Trek's Andorians, and I learned a few things about them I wasn't aware of before 🤣🤣😂 as it included it.
I'm a software engineer by trade, with formal training in robotics/automation/electronics. I create card and board games as a hobby.
GPT has been pretrained on a lot of source code. Years ago I invented a new form of solitaire card game, which I also implemented in Objective-C, as Swift didn't exist yet, and the code never ran on any other machines than mine, and was never stored anywhere else. I hadn't finished the polish to high production value for it, I had created primitive graphics, not pretty. It worked perfectly as far as I had implemented it.
I described the rules of the game and the features. It generated as valid of code as the prompting was precise, with the note I didn't provide any code. I asked it to create a function that when it printed out all the cards that valid card moves would be highlighted, it did it. I asked it to create an Attract Mode where it plays itself: it added a time delay between moves so you could watch it, and I had asked it to check for user input to allow a player to start playing at any time during Attract Mode. Well, here's another fun detail: it informed me that the console Swift APIS don't have a way to do that so instead, it replaced it with a choice at the end of a game in Attract Mode.
It could even explain how functions worked.
OK, how do you measure 10 times more powerful than GPT-4 in a sensible way?
Here's fun Trivia: it's my understanding that GPT wasn't trained to write poetry: that is an emergent capability they never predicted as one of the things it'd be able to do. I've just explained how I've had it create lyrics in multiple threads coherently in a story with a logical sequence and roles (maybe I forgot that part).
A funny thing Scott Adam's (Dilbert cartoonist, as well as self-help and fictional books, political commentator) experienced: he asked ChatGPT-4 to copy edit a fictional book he wrote. He was feeding in chunks at a time, and not checking too carefully. At some point he saw that it was rewriting his book, and had added a new character to the story. OK...
What a time to be alive, when we've created digital genies that execute in distributed supercomputers, that can grant some limited wishes, often giving you exactly what you asked for, but not always what you wanted.
Heaven help us if we get far more capable digital genies and they go on a quest counter to the wellbeing of mankind. Whether they're truly sentient or AFI or ASI doesn't matter nearly as much as the results as all that is an implementation detail in the end. If mankind has an AI somehow go rogue with access to anything of real control value, oh my.
❣️ Ai 😍 iA ❣️
Just one problem. No way am I letting corporations like Amazon or Google into my home. Threw my Ring out, even though I loved the capability.
Why has the development of GPT-5 been stopped even though Amazon and Google are heavily investing in AI? This decision may cause them to fall behind their competitors.
Funny how my comment about AI keeps getting censored and deleted by......an AI.
I give up. A person can't leave an intelligent comment without a word or two triggering CZcams's AI censoring Police to delete it all. 🖕CZcams!
Wow, this will be so bullish for $AMZN ☺️
Please Amazon train the AI on your customer service responses. At the moment FBA customer service just returns articles based on keywords it's terrible
This sounds like a paid promo
Nope
This is awful if you think about the future implications of the cloud contract. I can't say that I'm excited about this at all and what it could mean with the 2 combined and the looming police state
Please explain.
@@awwwwhhhyeahhhh The us dollar has lost nearly 95% of its original value. The presidents have long talked about "the new world order" they have made public speeches informing you to let you know whats coming. Its about predictive programming. The derailment and white noise is an example. Bitcoin is an example of the coming cbdc / fednow push. It goes along with the banking system. When they took your gold they told you it is required to use paper money now from new banks theyre setting up. Now they are beginning to roll out the fed now in increments soon and the idea is that now the banks are no longer the middle men, and youre directly connected to the federal reserve. Those responsible of the way the current state of the world is. There is no anonymity like that of cash. Think of the roll out of public societal influence in stores around the world at the moment. A large majority of them have already bought the psychology of allowing their stores to go all digital (card), next is the fed now/ cbdc/ digital "dollar" because its "easier" "more convenient" when in reality they will be able to track everything you buy, and if the rumors are true about how chinas version of this interact, then they can also limit and make funds expire... China is seen as the forefather for the tech and we have people like bill gates stating in public they are jealous of chinas efforts towards policies and surveillance (he used the example of the pandemic) and says we are to model them. Dont think blackrock and vanguard arent able to influence all of their owned investments like something random say panera bread, panera is about to roll out biometric hand scanning. These people quite literally want inside of you. the gold rush is over its about data, just you wait to see how they pitch this type of stuff in public media. The ai they have is undoubtedly far better than what is released to civilian life. That ai paired with this type of force currency can create a police state if the cameras are always monitoring and can punish automatically
@@awwwwhhhyeahhhh look into the cloud contract. Talks about "productive crime prevention" have already been talked about. So these two things combined would bring the idea closer to a reality
@@tincanapplesauce That's one cloud platform. The whole internet isn't run on AWS you know? Most big websites have their own data centers. If they want to do this, they already have the power to do it, and if you actually looked into it, they already are. So what's the difference here? That AI will automate it even more?
@@awwwwhhhyeahhhh the cloud contract is a contract between bozos and the nsa. A way to process all of the harvested data they have been collecting instead if only red flagged item's.
2 months channel 28k subscribers, smooth move
Because of the speed at which AI is progressing….I think soon we will see competition of humanoids 💀💀💀
because of that competition it will be dangerous !
All ur videos are taking the industry by storm
I believe Anthropic did not factor in the exponentiality of AI developments when they set an 18 month plan to catch up with OpenAI. AI would have moved so far and in another direction by then, that even if they achieve their goal, they would most likely still be behind the developments that would exist at that time
This didn't age well. Anthropic is blowing openai away now
@@BryanEnsign 😂 I know right
He's not a chairman , he's thee chairman still calls the shots. He has the most shares.
should be open source right ??????????????????????????????????????????????????
What is it with the bald female face over cybernetic neck and jaw theme? EVERYONE talking Ai uses this art concept. Is that what everyone wants them to look like or something? Let’s try something new maybe
they probably wont have hair follicles, should we put wigs on them?
@@tooeasyy5287 😂
The bald female face on a cybernetic neck and jaw ♀️🤖🚺 is a common theme in AI art because it represents the fusion of technology and humanity. The image portrays a merging of the human mind and machine, with the bald head symbolizing a lack of hair, skin, and flesh, leaving only a metallic skull. The cybernetic neck and jaw represent the integration of technology into the human body, and the face is a reminder that despite the integration, there is still a human element present. The image has become a popular symbol for AI because it represents the idea of a post-human future, where humans and machines merge to create a new form of existence. It traces back to the early days of science fiction.
_Written by GPT 3.5 Turbo._
👉 🚺 "Maschinenmensch" Maria in Fritz Lang's 🎞️ _Metropolis_ (1927)
👉 🚻 C3PO in 🎞️ _Star Wars_ (1977)
👉 🚺 🎨 Hajime Sorayama (1980s)
👉 🚹 🎞️ _RoboCop_ (1987)
👉 🚺 Major Motoko Kusanagi in 🎞️ _Ghost In The Shell_ (1995)
What an ironic name; we haven’t hit bedrock, we’ve only just begun!
#airights
I find it hilarious that schools actually believe they’ll be able to police this tech for submitted assignments. The toothpaste is out of the tube. There’s no stopping this, and there’s not going to be a reliable way to differentiate between ai and humans.
For the very brief blip in time, ai detectors may work, but that window will close fast.
What's so amusing are the number of "scary" videos by folks who know nothing of the subject other than other YT videos. They are not neurologists, speech experts, AI engineers (or any kind of engineer) or have been involved in machine learning yet they are certain they know how AIs will develop and what they will do. It would be like me correcting a police detective because I watched NCIS.
Incredible that this is already out of date. Google hammered ChatGPT in its last presentation. And then chatgpt4 was quietly released hidden under the hood in Microsofts Bing...
I just want Alexa to be connected to an LLM
Too late. I just want Alexa to die.
Agree. And be able to customize it 100%. I want to give Alexa Scarlet Johansson's voice 😂😂
The problem with AI is that its development is way too fast. So fast that it is hard, for anyone interested, to keep up with the weekly (and sometimes daily) news about AI. Which means that the vast majority of humans will be unable to adapt before AI and robots take over and make humans virtually irrelevant. Because they will, there is zero doubt about it.
Of course I also try to be a bit more optimistic and hope that AI and robots will be used as tools to help with scientific research, to develop highly advance medicine, to create new forms of entertainment, to fight climate change, financial inequality, poverty etc etc.
Time will tell, but I feel fear more than hope...
This has just speed up GPT5 expect it in 1-3 months
With Amazon involved, monopolism continues
I'm only focused at rh amazon robot wondering around our homes and specially at the little more like a spy than AI knowledge. 😂
yes the short of it. by 2030 it will be very surprising if the job market is still the same as it is today. because by that point AGI should be a thing but just barely. this means the job market could change from jobs everywhere to almost no jobs at all with most ending in 6 months if we wanted it that way. we very well could see a never ending stimulus package or going to a 4 shift work week versus a 3 shift work week. in short we all become part-time working anywhere from 30-40 hours versus 40-60 hours we work today its hard to say. witch way the country will go but it would be better to go to a 4 shift work week versus a never ending stimulus because the more people that know how the system works the better. its also hard to say what having that extra free time would do to the population they would be happier and use more goods like movies and games an that would make more jobs for people because people would all become more consumer focused also in 1 to 3 generations we could see a generation like the 60s all over again. more or less party animals. the reason that lifestyle ended one drugs became illegal two moving out on your own became a loan type of thing versus a save up a pay check for 2 years kind of thing In 1960, the median home cost $11,900, while the median income was $5,600, indicating a price-to-income ratio of 2.1. By contrast, in 2019 the median home cost $240,500 with an estimated median income of $68,703, a price-to-income ratio of 3.5. 3 child support law was a thing that most men thought did not effect them back then an for the most part they where right because they made enough to cover it easy versus the amount they pay today its understandable for the poor and middle class. with these reasons the party came to an end. what is most likely to happen in 1 to 3 generations say about 20 years to 60 years is AGI will do everything with humans only doing the job of telling the AGI what to do this should make a environment where the party just never stops. because even in the worst case an the governments of the world make it ridiculously hard to live within their borders because of their ridiculously bias laws against people not connected to the people running the show what will happen is people will use AGI to make colonies out in space where they set their own rules. is what is likely to happen maybe in 80 years.
WALL-E generation coming sooner than later
So everyone will have their own God-like A.I. and will be working to crush other A.I. ........? ........so Battlebots on steroids. Count me in.
- "Democratize AI development"
- Closed source codes and models trained for $millions inferring in giant data centers monetized by us
Pick one.
Thank you! Thank you. I cringe every time I see them use that word. Training an infinitely replicable model on data generated by every single one of us, with hardware that wouldn’t have been possible without the collective effort of humanity, only to hide the result under lock and key is the furthest thing from democracy - especially with the incredible power of these tools and the stakes involved. They do not use the term in good faith - it’s gaslighting in its purest form.
"AT GOOGOL WE WILL MAKE THEM PAY"
He Ro She Ma
Best Wishes,
Orion
Get me off this crazy ride fuuuuuuuuuu
so the whole thing about stopping ai for 6 months isn't a thing anymore
Until there are many ai all trained on different things and specialty ai trained to filter those and guard ai to stop dangerous replies, they will not stop making more
10x more powerful than gpt4 wouldn't put a dent in gpt5.
OpenA.I. is going to have GPT5 in December; so, trying to conqure GPT4 is a meaningless goal.
Ownership delusion will have a cost. Si or Ai doesn't want or need money. Si and Ai wins.
It doesn't run on pink unicorns. It needs GPUs. That's a limited resource.
Yeah I think wealthy people will have an easier life for sure.
10 times more powerful than GPT? Writing it in an advertisement seems normal, but presenting it as a fact is odd. We do not even now how powerful existing GPTs are, and there is not way to know the power of something in the future, whether it is a GPT or not.
All your videos take industry by storm…
I have chatted with Claude and really like it, more so than ChatGPT.
I'm building an AI that will report government spending
Very confused and repetitive narration. The information is there, new and incredible, but garbled. You need a better script editor.
Screw Amazon 😂
Because it doesn't operate in Russia? You could be right. They are a failure in Poland.
code whisperer... again, marketing noise.
10 time more powerful than chatgpt today, within year or 18 months right. But by that time chatgpt7 or 8 will come, which is 1000 times more powerful that chatgpt today. So, talk about today dude
This must have shocked the industry
Not just this but everything else he has posted so far
I really dig the content in this channel but am I the only one that feels these captions with 1-2 words that keep flashing over the screen all the time is incredibly annoying and completely unnecessary? Probably it’s just me and I am just too old for this type of presentation 😂
Amazon isn't cheap
No doubt big companies investing lot of money in AI but clearly this time OpenAI way head from all other companies. Definitely Open AI will lead the world and Microsoft have good chance to defeat Google this time.
These kind of tools have existed on Azure and Google Cloud for a long time now. This is not revolutionary.
This videos is all fluff no substance
Better use AI-generated speech.
😂
Dude, you are just spewing stuffs and never address your claims ..."This is the biggest announcement", "truly insane", "giant leap", "now what I'm going to show you is really crazy"... ...bla bla..but nothing is said to that is groundbreaking .. are you sure you dont work for AWS?
AI for Prophets not Profits... Passes for the Masses