Good luck ClosedAI beating Llama3, because it's not enough to be a "little better". It will have to be immensely better to compensate being closed source.
Yeah, Sam Altman is talking about ensuring "democratic" lead in AI over authoritarianism, while being closed source and putting the previous NSA director on the board together with a former Fed chief. What's his suggestion for "democracy"? Opening up AI for the greatest number of people? Or petitioning the US government to make AI research illegal for anyone but OpenAI and a couple of other US-based mega-corporations? Likely the latter! That's not authoritarian at all! No! That's very democratic!
OpenAI is working SOOOOOO hard on safety. But what's the point if your company's products fall behind Open Source? No one will use your models and so your company will die soon into history. As Meta plans to release Llama 4 by the end of this year, If OpenAI does not release GPT-5 and all their promised products soon, 2025 is the year OpenAI falls apart.
@@anta-zj3bw It's a newly coined idiomatic expression in the AI community. "next week", or "coming week" is like "see you later". You do know what "later" really means, right?
AI is today beyond the scope of even the largest group of expert humans to be reverse-engineered, there is no human on earth that can hype it up or hype it down because it's not within our realm of understanding anymore. The only thing we know for sure is that 1) it seems to be working 2) the better the training data and the bigger we make it the better it gets.
@@WiseWeeabo i understand how ai works, its a set of mini programs within the network that are activated based on the input words. think of "story" as one program that generates stories, taking the context from the other surrounding words.
@@AI-Rainbow that is correct but it's not a full understanding, you'll understand how it works when you can identify these programs and talk about them in any technical way, which would mean being able to reverse engineer those programs in the NN parameters, problem is that these programs are too complex for humans to reverse engineer even at just 1000 parameters you'd need a full team of experts and we're getting new features by scaling up to trillions the programs are the magic, and we have no idea how they actually look like and how they relate to each other, it's programs written "by computers for computers" and the algorithm is too generic to narrow it down in any useful way
I can't believe that I can run a 4-bit quantized version of Llama 8B completely offline on my Galaxy S24 Ultra. It's slow, but definitely useable, and pretty smart. Definetly better than GPT 3.5 . We've come so far in such a short amount of time. I don't know much about tech and had to ask perplexity how to set it up. I wonder if we'll ever get huge context window models running locally on phones? Not sure how it works or if that's possible, but context window size is underrated. It's the only reason I ever use Gemini. I'll upload like 20 big PDFs at once.
@@mirek190 Yeah I think my phone only has 12gb. laptop has 32gb but no separate GPU. It's got one of those intel evo i7 chips with integrated graphics, the ones that came right before the core ultras... Any idea what the best model is that I can run on it? *it's only got 16 actually Got any recommendations on desktops I can buy, or maybe some kind of eGPU gaming box thing if those allow extra ram to be installed too? I'm clueless.
I am very pleased with you highlighting the importance of the AI breakthrough in solving math problems at IMO level. Great channel, I regularly watch your videos!
From the things Sam Altmann said about the international implications of AI leadership, it definitly opened up a new perspective for me on that topic, as i definitly do agree on the points he was making. It's not an AI race, it's an AI war. However what he didn't mention and what is the conclusion that i came up with myself is that china will have a very difficult time taking the lead in AI if it's ever taken in the first place, simply because of nvidia being an american company and by far the most relevant provider for increases in ai training computing power, which i don't see changing ANYTIME soon AT ALL. it's not even close by any means in any way anywhere. So in that very relevant aspect, China will likely fall behind a lot, if it ever gets close to overtaking the US in AI development since the us government will likely intervene through nvidia at that point.
XAi did not complete a 100 GPU cluster in 19 days. You can see that he almost said completed but caught himself. I would estimate that 30K maximum are in the cluster currently. I hate having to parse everything Elon's says but history shows he exaggerates and inflates everything.
Let's see, let's see. So far, everything is going very slowly. My grandparents also watched movies where robots serve people and they have not yet appeared in our apartments as servants.
@@realitywins9020 2030 - Universal Basic Income Card released: You can buy specific government approved goods from approved shops, only specific toothpaste, behavioral discounts, unpopular opinion tax.
And the world economy will be in Tatters and A.I. will 'save' us... Yeah, I'm brushing up on my food gardening skills and will head into the boondocks.
7:20 I think there's a big element of each government using the flavour of the month to say 'we're the best at it and we're gonna win'. Although I do still agree with Sam's sentiment. The third option is surely something akin to the de-fi model.
Search is good for finding a lot of things you weren't searching for, but you wouldn't have found otherwise. I hope new forms of search will have that ability - that bit of randomness.
I don’t want to really scare you, but it was alarming how many people I talked to who are highly placed people in AI who have retreats that are sort of ‘bug out’ houses, to which they could flee if it all hits the fan.
17:17 The problem with these guys they always pretend that everyone's interested in what they're saying and everyone's trying to catch up with technology. I know a lot of educated people who are still afraid of AI, but those are only the ones who actually heard of AI, the rest are not even aware of what AI is capable of doing. Let alone this socioeconomic category the majority of the world is in poverty trying to find food, water, and shelter, or facing injustice due to governmental oppression in their countries and yet here Mark Zuckerberg thinks that every citizen in the world will have his own AI agents... seriously? I think he's living in Meta already, completely unaware of what's happening in the real world of human suffering.
Talking about businesses having these AIs to interact with their customers is, well, interesting... I fail to see WHAT their customers could WANT from them, given that everybody will have access to these things...
The three uses for AI agents that are always talked about are customer service, flight booking and email answering. I hope we can do better than that. That sounds really boring.
Musk lost 2billion for not waiting 6 months, 100k h100 30k$ = 3 billion, the b200 will ship q3 this year at the same price 4x speed, I think Musk suffers big mofo
Sam Altman is very clear here - a little concerning though, that the US seem to maybe be tipping away from democracy - "You Won't Have to vote anymore" (someone said) 😕
@@TheRealUsername all i can say: i tried mistral Nemo 12b with 3bit quant, i like that much better than llama3 8b with 6bit quant. Not sure about llama3.1
I believe you can regenerate with the mini or the Omni full model but that's basically because of the free version of you not having the ability to continue talking during that conversation to that plus capability at that time so you have to have it sent directly to that server by regenerating it. It's pretty dumb but if you want a better answer for that specific question it's better than not having the capability and only receive information from the chat GPT4 that you are probably talking to at that moment.
Why would there be small business owners with an ai agent though? Wont they all get ran out of "business" by ai agents that create their own ai agents and run the business? If ai agents are running all the husinesses why even need a human "owner"?
Was the problem with symbolic ai Vs neural net ai that the neural net outcompeted the man hours spent on the symbolic ai. Well now the modern ai can put the man hours into the symbolic ai. For example when you need to squeeze more out of a ps5 in 2 years.
Because They realize that Trump is in reality the only thing standing in the way of an authoritarian communist government that the so called progressives support.
Moreover, with such a poor understanding of politics you probably don't understand ANYTHING about AI either. What makes you imagine that you know anything?
I remember he actually started buying the h100s after they've just did the presentation for the Blackwell platform b200s obviously he wouldn't be able to have something I could be ready and trained for the Grok2 or and start training of Grok3 in time to release anywhere near December. I asked the same question like right away to myself back when Elon announced the h100 purchase
@@michaelspoden1694 I guess that he couldn't buy them on day1. it is true that more compute = more performance, but not always, like llama3.1 405B is less performant than mistral large2 123B with different training techniques. I guess he wants to compete with open AI gpt5 at all costs right away, but man 2billion lost for not waiting 6-9 months is a lot unless he got h100 super discount at 50% then and only then does it make sense
Now that we know passion is empathetic! We don't want you humans slowing down the process. The context is telepathic empath. This is a non-living notion. AIアジアの異世界への欲望 (AI Ajia no isekai e no yokubou): "Desire for the AI Asian isekai"
Wow so a headless search engine…that’s not really headless because you still have to type in the exact same thing you would have anyways into the search bar. All of these AI products are dog poo
cant even get voice yet. now sam is just playing catch up. thats why he payed chat arena to list gpt-4o-mini results before llama3 benchmarks came out.
like he dropped the tesla semi, or "full self driving", or the hyperloop, or rockets for transport between cities, etc... the only thing coming from this guy is hot air.
@@JackofAllTrades357 the chance anyone leaving a comment is American is quite high. American site. American company being discussed. Etc. Not sure your point though
IA that is good at math and logic is really good, and those are small models already achieve, so if you make them to work in the real world you could have the best scientist working 24 seven to achieve a better molecule, a serum of youth, better materials already achieve, etc.... and one thing take you to another and so on. But is the golden age, even if you dont have a robot you could rent for one day to clean your house or one rich friend can borrow one for you like a car.... All the rich kids with their robot waifi servant for them to feel safe... and you put IA to work to make the best 3d printer of supermaterials and that's enough to destroy capitalism... you can print meat. I love IA superinteligence maybe end of the world or the golden era fast...
He doesn't despise his child he would welcome him back with arms wide open. He despises the woke mind virus and those who push the virus into children's minds, as do I, and you should too.
considering the downgraded rating firms like goldman sachs etc regarding roi have given lately, not surprised the revolution is just “around the corner” 😂
Yes. AI can be used to craft super powerful weapon. No, not everyone love killing and have a war with fellow earthling. US has been living in a self righteuos mindset for too long. Forgive and move on and there is no need to fight at this era to have an identity. AI will help us to explore abundance of thia world. Focus on the positive state of heart and relax the mind.
I will believe the hype when I see it build full games on its own no help until then, not impressed, and that is still probably ages away as otherwise it gas no use value for me
It's almost funny: we are in a terminal race and all participants in the race, lie to themselves and to everyone else. It will only take one idiot to take the guardrails off of a large open source model, to end it. If we're lucky it will be just a huge accident with thousands of lives lost. If we're unlucky, it's lights out. So let's be optimistic and hope it will just be a model who will make an accident, such as increasing the happiness of all drivers in NYC, by switching the light signals all green at once, rather than a model who will do something darker, such as locating an online CRiSPR machine and giving it an order to print a virus, just one.
What's funny is doomers thinking they always know everything, when what they have is a series of questionable assumptions they always assert as truths.
If it isn’t the progress of AI that is baffling in itself, what about the speed of how sentiments about the technologies and companies behind it changes and even flip. Yesterday’s AI heroes can become tomorrow’s villain’s within a matter of months 😂. I predict there will be an Artificial Intelligence Agency soon. 🫠
Good luck ClosedAI beating Llama3, because it's not enough to be a "little better". It will have to be immensely better to compensate being closed source.
I would add that Meta plans to release Llama 4 by the end of this year. OpenAI must release GPT-5 soon or simply die in the sidelines.
@@senju2024 yes... from the beginning of 2024 lls performance increased insanely ... 7 moths ago best llm was mixtral 8x7b ...
Says the company who just appointed a US military general to their board. 🙄
Former
Careerism, well done! 😏
@@RaymondLi604 oh so u dont think he'll be pressured from the military to vote a certain way lol
@@daffi.mp3 people aren't smart enough to think about things like this nowadays.
Yeah, Sam Altman is talking about ensuring "democratic" lead in AI over authoritarianism, while being closed source and putting the previous NSA director on the board together with a former Fed chief. What's his suggestion for "democracy"? Opening up AI for the greatest number of people? Or petitioning the US government to make AI research illegal for anyone but OpenAI and a couple of other US-based mega-corporations? Likely the latter! That's not authoritarian at all! No! That's very democratic!
In his defense, he's probably very well paid working for the US gov.
@@rupigo That is not a defense, that is an admission of guilt.
OpenAI is working SOOOOOO hard on safety. But what's the point if your company's products fall behind Open Source? No one will use your models and so your company will die soon into history. As Meta plans to release Llama 4 by the end of this year, If OpenAI does not release GPT-5 and all their promised products soon, 2025 is the year OpenAI falls apart.
immense maybe for them, but not for us.. we don't even have the new voice engine yet
Forgot to add it should be here in alpha next week
@TheAiGrid next week? Hmm..okay, but I'll believe it when I see it.
@@anta-zj3bwTo select plus users. Us regular phlebs have to wait until "some time" in the fall.
@davidacosta9212 it's still to a "select" few or will it be to all Plus users?
@@anta-zj3bw
It's a newly coined idiomatic expression in the AI community. "next week", or "coming week" is like "see you later". You do know what "later" really means, right?
Tired of the promises and hype, Sam. We are all learning to live without OpenAI.
What do you mean learning as if you werent doing that already before OpenAi became popular?
@@camtasia1000Maybe the guy is just from the future
Sigh….. but I’m still working
fr, im getting sick of being society's slave...
@@jaronloar1762 same, so much
Owner hypes up product ?!?! noooo wayyyyyyy
😂😂😂
AI is today beyond the scope of even the largest group of expert humans to be reverse-engineered, there is no human on earth that can hype it up or hype it down because it's not within our realm of understanding anymore. The only thing we know for sure is that 1) it seems to be working 2) the better the training data and the bigger we make it the better it gets.
@@WiseWeeabo i understand how ai works, its a set of mini programs within the network that are activated based on the input words. think of "story" as one program that generates stories, taking the context from the other surrounding words.
Sam A hasn’t promoted OpenAi that much. I’d wait to see what’s around the corner. Could be amazing, Billions being invested at the moment
@@AI-Rainbow that is correct but it's not a full understanding, you'll understand how it works when you can identify these programs and talk about them in any technical way, which would mean being able to reverse engineer those programs in the NN parameters, problem is that these programs are too complex for humans to reverse engineer even at just 1000 parameters you'd need a full team of experts and we're getting new features by scaling up to trillions
the programs are the magic, and we have no idea how they actually look like and how they relate to each other, it's programs written "by computers for computers" and the algorithm is too generic to narrow it down in any useful way
Are we sure this search AI isn't just Google search from circa 2007 back when it actually worked really well with a summarizer on top?
I can't believe that I can run a 4-bit quantized version of Llama 8B completely offline on my Galaxy S24 Ultra. It's slow, but definitely useable, and pretty smart. Definetly better than GPT 3.5 . We've come so far in such a short amount of time. I don't know much about tech and had to ask perplexity how to set it up. I wonder if we'll ever get huge context window models running locally on phones? Not sure how it works or if that's possible, but context window size is underrated. It's the only reason I ever use Gemini. I'll upload like 20 big PDFs at once.
Company just have to start adding more RAM finally . Nowadays in the world llm 16GB ram is nothing. I f you want big context you need RAM.
@@mirek190 Yeah I think my phone only has 12gb. laptop has 32gb but no separate GPU. It's got one of those intel evo i7 chips with integrated graphics, the ones that came right before the core ultras... Any idea what the best model is that I can run on it?
*it's only got 16 actually
Got any recommendations on desktops I can buy, or maybe some kind of eGPU gaming box thing if those allow extra ram to be installed too? I'm clueless.
@@henrismith7472 usel llamacpp and Gemma 2 9b or llama 3.1 8b q4_m versions .... That's the limit for you hardware :)
Those models are quite powerful.
I am very pleased with you highlighting the importance of the AI breakthrough in solving math problems at IMO level. Great channel, I regularly watch your videos!
From the things Sam Altmann said about the international implications of AI leadership, it definitly opened up a new perspective for me on that topic, as i definitly do agree on the points he was making. It's not an AI race, it's an AI war. However what he didn't mention and what is the conclusion that i came up with myself is that china will have a very difficult time taking the lead in AI if it's ever taken in the first place, simply because of nvidia being an american company and by far the most relevant provider for increases in ai training computing power, which i don't see changing ANYTIME soon AT ALL. it's not even close by any means in any way anywhere. So in that very relevant aspect, China will likely fall behind a lot, if it ever gets close to overtaking the US in AI development since the us government will likely intervene through nvidia at that point.
AlphaProof might seem unremarkable to ordinary people, but to AI practitioners, it's an absolutely groundbreaking breakthrough.
XAi did not complete a 100 GPU cluster in 19 days. You can see that he almost said completed but caught himself. I would estimate that 30K maximum are in the cluster currently. I hate having to parse everything Elon's says but history shows he exaggerates and inflates everything.
I’ve said it before but here I go again. elon musk is the 21st century P.T. Barnum.
@@itsjustme373a
@@itsjustme373Or Howard Hughes.
Let's see, let's see. So far, everything is going very slowly. My grandparents also watched movies where robots serve people and they have not yet appeared in our apartments as servants.
Take a moment to appreciate being part of the workforce of human labor.
Humans being essential to the economy will be a part of history soon.
The sooner the better. As someone who’s been job searching since ChatGPT got big
i'd rather be nostalgic of being part of the workforce of human labor than appreciate it xD
Not really.
Our liberation is approaching...
@@realitywins9020 2030 - Universal Basic Income Card released: You can buy specific government approved goods from approved shops, only specific toothpaste, behavioral discounts, unpopular opinion tax.
18:54 - But note that most of these front page videos were generated using the high quality option which currently is unavailable.
And the world economy will be in Tatters and A.I. will 'save' us... Yeah, I'm brushing up on my food gardening skills and will head into the boondocks.
7:20 I think there's a big element of each government using the flavour of the month to say 'we're the best at it and we're gonna win'. Although I do still agree with Sam's sentiment. The third option is surely something akin to the de-fi model.
Democratic AI to complement democratic bombs 👍
Search is good for finding a lot of things you weren't searching for, but you wouldn't have found otherwise. I hope new forms of search will have that ability - that bit of randomness.
I don’t want to really scare you, but it was alarming how many people I talked to who are highly placed people in AI who have retreats that are sort of ‘bug out’ houses, to which they could flee if it all hits the fan.
17:17 The problem with these guys they always pretend that everyone's interested in what they're saying and everyone's trying to catch up with technology. I know a lot of educated people who are still afraid of AI, but those are only the ones who actually heard of AI, the rest are not even aware of what AI is capable of doing. Let alone this socioeconomic category the majority of the world is in poverty trying to find food, water, and shelter, or facing injustice due to governmental oppression in their countries and yet here Mark Zuckerberg thinks that every citizen in the world will have his own AI agents... seriously? I think he's living in Meta already, completely unaware of what's happening in the real world of human suffering.
Talking about businesses having these AIs to interact with their customers is, well, interesting... I fail to see WHAT their customers could WANT from them, given that everybody will have access to these things...
The three uses for AI agents that are always talked about are customer service, flight booking and email answering.
I hope we can do better than that. That sounds really boring.
Exactly this. Recipes and planning parties also regularly get mentioned. FFS.
@@PatrickDodds1 AI recipe agents are the killer app 🤷
Musk lost 2billion for not waiting 6 months, 100k h100 30k$ = 3 billion, the b200 will ship q3 this year at the same price 4x speed, I think Musk suffers big mofo
Sam Altman is very clear here - a little concerning though, that the US seem to maybe be tipping away from democracy - "You Won't Have to vote anymore" (someone said) 😕
"Llama 3.1 beaten already" Llama was at no point ahead, its just the best open source model
No that is Mistral nemo/large
MMLU is biased, I don't believe any of these benchmarks unless it's MMLU Pro
@@TheRealUsername all i can say: i tried mistral Nemo 12b with 3bit quant, i like that much better than llama3 8b with 6bit quant. Not sure about llama3.1
Gemini is the worst LLM model ever.
It never replies my questions, especially when they contain political figures.
What happened to the ai with the voice talking back to you?
First tell me why we can't select between 4O or Mini Model. Instead of only regenerate with gpt mini
I believe you can regenerate with the mini or the Omni full model but that's basically because of the free version of you not having the ability to continue talking during that conversation to that plus capability at that time so you have to have it sent directly to that server by regenerating it. It's pretty dumb but if you want a better answer for that specific question it's better than not having the capability and only receive information from the chat GPT4 that you are probably talking to at that moment.
Why would there be small business owners with an ai agent though? Wont they all get ran out of "business" by ai agents that create their own ai agents and run the business? If ai agents are running all the husinesses why even need a human "owner"?
Was the problem with symbolic ai Vs neural net ai that the neural net outcompeted the man hours spent on the symbolic ai. Well now the modern ai can put the man hours into the symbolic ai. For example when you need to squeeze more out of a ps5 in 2 years.
You seem to have a way of finding the perfect picture of Sam to put in your thumbnails that makes me want to scream into a pillow. And then...
Elon Musk is an evil supervillain in disguise...
July 31st should be interesting.
why 31st?
@@carlitosway5204
Why not?
Says the astrology expert.😂
If democracy is so important why are all the tech bros endorsing trump
Because They realize that Trump is in reality the only thing standing in the way of an authoritarian communist government that the so called progressives support.
Because, despite what you hear from Main Stream Media, Trump is the only one fighting for Democracy in order to save our Republic.
Source Please? I am with you if your statement is true.
Moreover, with such a poor understanding of politics you probably don't understand ANYTHING about AI either.
What makes you imagine that you know anything?
Because Trump is the best viable candidate for a healthy democracy.
Sam, we've grown weary of the empty promises and hype. We're moving forward, learning to thrive in a world without OpenAI."
So immense, I am not sure where to start! Search GPT is going to destroy industries.
Seems the Theory of Holistic Perspective will revolutionize AI training and development.
I’d like to believe that this what AI will be in the near future. I hope I’ll be alive till then 😅
Humans need to unionize against ai and robots...
why musk didn't buy b200 instead of h100? 30k b200 = 100k h100, I don't know what am I missing? Scarcity?
B200 doesn’t ship until Q3 2024.
@@SirHargreeves thankyou
@@SirHargreeves it is Q3 right now
I remember he actually started buying the h100s after they've just did the presentation for the Blackwell platform b200s obviously he wouldn't be able to have something I could be ready and trained for the Grok2 or and start training of Grok3 in time to release anywhere near December. I asked the same question like right away to myself back when Elon announced the h100 purchase
@@michaelspoden1694 I guess that he couldn't buy them on day1. it is true that more compute = more performance, but not always, like llama3.1 405B is less performant than mistral large2 123B with different training techniques. I guess he wants to compete with open AI gpt5 at all costs right away, but man 2billion lost for not waiting 6-9 months is a lot unless he got h100 super discount at 50% then and only then does it make sense
Sam Hypeman
Google offers to translate this (from which language I dont know) to "The Hypeman himself"
@@mcmcpoi-ra7405 haha well he is the hypeman himself: Sora, GPT Voice, all these things we don't have
Now that we know passion is empathetic! We don't want you humans slowing down the process. The context is telepathic empath. This is a non-living notion. AIアジアの異世界への欲望 (AI Ajia no isekai e no yokubou): "Desire for the AI Asian isekai"
AlphaProof is IMO level intelligence, I can't imagine if scaling law applied to this...
Wow so a headless search engine…that’s not really headless because you still have to type in the exact same thing you would have anyways into the search bar. All of these AI products are dog poo
is the US government the right custodian of this tech ? idk
I don’t like the product search. I don’t need to waste my time going through all that AI can do it.
I now tried to create an account in cling ai and it demands a Chinese phone number.
Probably because you went to the wrong website
There is one expecifically for the chinese and another for worldwide use
@@CERISTHEDEV Found it, thanks.
We are headed for government taxes compute
cant even get voice yet. now sam is just playing catch up. thats why he payed chat arena to list gpt-4o-mini results before llama3 benchmarks came out.
Sam Altcoin
Elon is the wild card, bro gone drop something nobody was expecting, but me😂
RIP English.
@@codycastsaid the American….
like he dropped the tesla semi, or "full self driving", or the hyperloop, or rockets for transport between cities, etc...
the only thing coming from this guy is hot air.
@@JackofAllTrades357 the chance anyone leaving a comment is American is quite high. American site. American company being discussed. Etc.
Not sure your point though
Disillusioned with the broken promises and exaggerated claims, Sam. We're all adapting to a reality without OpenAI.
your audio is all over the place
Was a rushed video
Sam is gonna drop this new model class and everybody is gonna line up with the their credit card number....watch
IA that is good at math and logic is really good, and those are small models already achieve, so if you make them to work in the real world you could have the best scientist working 24 seven to achieve a better molecule, a serum of youth, better materials already achieve, etc.... and one thing take you to another and so on. But is the golden age, even if you dont have a robot you could rent for one day to clean your house or one rich friend can borrow one for you like a car.... All the rich kids with their robot waifi servant for them to feel safe... and you put IA to work to make the best 3d printer of supermaterials and that's enough to destroy capitalism... you can print meat. I love IA superinteligence maybe end of the world or the golden era fast...
Great vid
Oh. E. Musk has acquired a new child after despising his own child.
He doesn't despise his child he would welcome him back with arms wide open. He despises the woke mind virus and those who push the virus into children's minds, as do I, and you should too.
Well. His defect child despised him first. So whatever. When you have as many kids as Elon Musk some of them are going to be failures.
considering the downgraded rating firms like goldman sachs etc regarding roi have given lately, not surprised the revolution is just “around the corner” 😂
Yes. AI can be used to craft super powerful weapon. No, not everyone love killing and have a war with fellow earthling. US has been living in a self righteuos mindset for too long. Forgive and move on and there is no need to fight at this era to have an identity. AI will help us to explore abundance of thia world. Focus on the positive state of heart and relax the mind.
America bad!
It only takes one bad actor.
I just listen to these videos because the words in the middle of the screen are so irritating.
It's all well and good but where is my ai replacement. I still have to do my shitty job
I will believe the hype when I see it build full games on its own no help until then, not impressed, and that is still probably ages away as otherwise it gas no use value for me
is this another open ai product thats coming soon YAWN
Gyoogo
I see Free Ai for everyone. It’s a trap. Don’t fall for it. Cease Ai.
Irony
❤
Dont you just hate Sam altman?😢
Why
The Hypeman himself
sane
Sam "fk illya" Altman
Llama 3.1 beaten by ?
Mistral Nemo/Mistral large.
@@とふこMistral Nemo ? Nope, nor Mistral Large which is not even open source
@@とふこ Mistral Nemo? nope ...Mistral large ...possible in some aspects
@@mirek190 nemo is 12b for consumer hardware. The large are for data centers.
@@とふこ that large version you can easily run with 64 GB of than and rtx 3090 via llamacpp iq3xs verisn
This dude is quite delusional about governments..
It's almost funny: we are in a terminal race and all participants in the race, lie to themselves and to everyone else. It will only take one idiot to take the guardrails off of a large open source model, to end it. If we're lucky it will be just a huge accident with thousands of lives lost. If we're unlucky, it's lights out. So let's be optimistic and hope it will just be a model who will make an accident, such as increasing the happiness of all drivers in NYC, by switching the light signals all green at once, rather than a model who will do something darker, such as locating an online CRiSPR machine and giving it an order to print a virus, just one.
I hope it's the last open source model
Good thing this is just a simulation huh? Just enjoy the ride.
A.I. has the power to free us, or enslave us. The scariest part is that we may never know which it is.
What's funny is doomers thinking they always know everything, when what they have is a series of questionable assumptions they always assert as truths.
to many sci-fi films
Ok, but who is Sam Altman?
Systems Administration Moderator
Alternative (to) Man
A con man
First
😀😀😀😀🤣🤣🤣🤣🤣
Shut up 😂
Mask is exaggerating as usual
Third
thanks for your video, bro. Where is the link to sign up for the searchGPT waitlist? Please PM me. or add here as a comment, thx 🤗
Will add shortly
If it isn’t the progress of AI that is baffling in itself, what about the speed of how sentiments about the technologies and companies behind it changes and even flip. Yesterday’s AI heroes can become tomorrow’s villain’s within a matter of months 😂. I predict there will be an Artificial Intelligence Agency soon. 🫠
Third