This perfectly explains Sam Altman's pivot to treating ChatGPT primarily like a Web 2.0 product rather than a scientific endeavour. He realized that LLMs are "good enough" and now is trying to get product/market fit ASAP.
I'm a Copilot Pro user and for me to get people to consider Copilot, I tell them to think of as a personal assistant. Like it's the personal assistant in "The Devil Wears Prada" or Iron Man (Pepper Potts, Gwyneth Paltrow). Example: I'm meeting on the XYZ project. Summarize my email on the project. Write a draft on the main issues. Now, get me a schedule with weekly task, so on. I think you can even say order me a coffee at Starbucks, not sure. Plus it's fun to use for many other things.
No mention of the millions of students using it for projects or call centre lay offs by the thousands. Lawyers are using it to replace assistants and many more in medicine.
Is it cool? Is it fascinating? Does it feel magical? Yes, yes, yes. Is it going to find a place in business? Yes. Did we find the path to Utopia? Let’s not get carried away. Maybe it’s an Internet-scale economic impact in the long run; maybe.
Intelligence is the number 1 most valuable resource humanity has ever had and ever will have. To suddenly have an abundance of it at the snap of your fingers is world changing, and if we use it right it could very easily be a path to utopia. The only thing standing in the way now is ourselves
I believe AI will have a bigger impact than the Internet. Internet = easy access to all knowledge. AI = new form of intelligence It's important to note that a new form of intelligence will be able to leverage the Internet in the same way humans have but they will be way better at it and they work 24hrs a day.
A big part of the problem isn’t just the hype coming from the AI industry but the effectiveness of that hype. People just outside the industry like technology pundits and science writers amplify the hype and people further out get really sold a bill of goods. They see a humanoid robot being controlled via an LLM and “explaining” it’s actions and suddenly they think “hey, they understand human language and can follow any instruction so they are pretty much human level right now.” If you disagree with that they think you are just not as well informed about AI and robots as they are. I can’t tell you how many people have told me humanoid bots and LLMs are ready to replace humans in the next couple of months.
This is excellent, finally cutting through all the ignorance, hype, and wild speculation. Thank you My take?: I phone/Android will offer various personal specialized AI assistant apps is what will take the diffusion of innovation curve parabolic imo, whereas the corporate adoption will stagnate due to most industries’ mal incentive structures (monopoly capitalism as an employment/profit protection racket), especially in the private markets. AI and especially AGI will offer a perfect fit to rebuild the trend of collapsing markets/industries as people’s perceptions of value will most likely dramatically shift away from the conspicuous consumption model in a post industrialized world and first world demographic turn over and current city/suburban evacuation to rural communities emerge.
😮 like today I had a 20-minute conversation about😮 solving the three-body problem😮 what's the measurement of stars in gravitational systems😮 and by the end of the conversation we had discovered a form of time travel where we could shoot probes into black holes and collect data from the future 😮 do you think the average person's doing that hell no😮
😮 not that many people use AI😮 because as someone who uses it on a daily basis most people literally lack the intelligence to ask it the right questions😮
Based on this comment , it is safe to assume that the average imbecile is smarter than you . The AI industry ends up with less useful idiots than they anticipated... that s a good thing and reminds me a little bit of what happened with NFTs and the metaverse, same problem : not enough useful idiots and too many mocking it ... fake it untill you make it ... or not 😂😂😂
Corpos never made it for ordinary people! No one in Intel, AMD or Nvidia wants oridinary people to use it by not giving people hardware to run such. They offering hardware mostly to governments today or very rich with many filters(ban to arabs & chinese). "Ai laptops" showed at Computex from all major brands with ONLY 16Gb RAM is a humiliation gimmick, showing that they don't want you to run anything big and locally! 16Gb is enough only for very primitive LLM on level of very simple translator or encyclopedia, but never any code writing or even long essay creation (Llama3 70billion need 80Gb RAM, chinese Deepseek v2 code writer with 300+ languages 236billions need more than 128Gb RAM). Even if you want to sell it - you can't - they don't sell hardware to your clients, if you want to run - you can't, they don't sell hardware to you. They want lock everyone to the cloud which is death to privacy & security altogether in society. Try to buy in ordinary store a server's 12 RAM slots motherboard, there's even weird CPU slot, so you must use only proprietary non consumer cooling.
The cost out weigh the benefits while the tech is there for 80gb ram. Getting that to run on a laptop with small fans and low heat. Basically impossible. That’s why the cloud is the main route. You get your processing power without the drawbacks.
This perfectly explains Sam Altman's pivot to treating ChatGPT primarily like a Web 2.0 product rather than a scientific endeavour. He realized that LLMs are "good enough" and now is trying to get product/market fit ASAP.
The 🐐 has replied. Means this channel deserves a follow.
Yeah, the glass keeps getting bigger. Great show.
I'm a Copilot Pro user and for me to get people to consider Copilot, I tell them to think of as a personal assistant. Like it's the personal assistant in "The Devil Wears Prada" or Iron Man (Pepper Potts, Gwyneth Paltrow). Example: I'm meeting on the XYZ project. Summarize my email on the project. Write a draft on the main issues. Now, get me a schedule with weekly task, so on. I think you can even say order me a coffee at Starbucks, not sure. Plus it's fun to use for many other things.
No mention of the millions of students using it for projects or call centre lay offs by the thousands. Lawyers are using it to replace assistants and many more in medicine.
Is it cool? Is it fascinating? Does it feel magical? Yes, yes, yes. Is it going to find a place in business? Yes. Did we find the path to Utopia? Let’s not get carried away. Maybe it’s an Internet-scale economic impact in the long run; maybe.
Intelligence is the number 1 most valuable resource humanity has ever had and ever will have. To suddenly have an abundance of it at the snap of your fingers is world changing, and if we use it right it could very easily be a path to utopia. The only thing standing in the way now is ourselves
I believe AI will have a bigger impact than the Internet.
Internet = easy access to all knowledge.
AI = new form of intelligence
It's important to note that a new form of intelligence will be able to leverage the Internet in the same way humans have but they will be way better at it and they work 24hrs a day.
A big part of the problem isn’t just the hype coming from the AI industry but the effectiveness of that hype. People just outside the industry like technology pundits and science writers amplify the hype and people further out get really sold a bill of goods. They see a humanoid robot being controlled via an LLM and “explaining” it’s actions and suddenly they think “hey, they understand human language and can follow any instruction so they are pretty much human level right now.” If you disagree with that they think you are just not as well informed about AI and robots as they are. I can’t tell you how many people have told me humanoid bots and LLMs are ready to replace humans in the next couple of months.
I definitely consider myself an AI user, but even I don't use it daily. Sometimes I have something to do with it, sometimes I don't, and that's fine.
This is excellent, finally cutting through all the ignorance, hype, and wild speculation. Thank you
My take?: I phone/Android will offer various personal specialized AI assistant apps is what will take the diffusion of innovation curve parabolic imo, whereas the corporate adoption will stagnate due to most industries’ mal incentive structures (monopoly capitalism as an employment/profit protection racket), especially in the private markets.
AI and especially AGI will offer a perfect fit to rebuild the trend of collapsing markets/industries as people’s perceptions of value will most likely dramatically shift away from the conspicuous consumption model in a post industrialized world and first world demographic turn over and current city/suburban evacuation to rural communities emerge.
That take is going to age like milk if not already obsolete, but as a parody it is funny af
@@Dystopian84 You mean like sour cream or yogurt? Like that? Well then, thank you…I think…
@@dalerohling5989 parody it is , when the skills are poor the comedy needs to be solid
😮 like today I had a 20-minute conversation about😮 solving the three-body problem😮 what's the measurement of stars in gravitational systems😮 and by the end of the conversation we had discovered a form of time travel where we could shoot probes into black holes and collect data from the future 😮 do you think the average person's doing that hell no😮
I dont think AI is any less acurate than a person. Their value will take time to exploit but I think its clear what the value is.
😮 not that many people use AI😮 because as someone who uses it on a daily basis most people literally lack the intelligence to ask it the right questions😮
Based on this comment , it is safe to assume that the average imbecile is smarter than you . The AI industry ends up with less useful idiots than they anticipated... that s a good thing and reminds me a little bit of what happened with NFTs and the metaverse, same problem : not enough useful idiots and too many mocking it ... fake it untill you make it ... or not 😂😂😂
🎵🎹🎶 💃 OpenAI My Sugar Papi by PEACHY da WHUUPi on UA-cam NOW.
Hey competition coming guys !
Luna used here
Corpos never made it for ordinary people! No one in Intel, AMD or Nvidia wants oridinary people to use it by not giving people hardware to run such. They offering hardware mostly to governments today or very rich with many filters(ban to arabs & chinese). "Ai laptops" showed at Computex from all major brands with ONLY 16Gb RAM is a humiliation gimmick, showing that they don't want you to run anything big and locally! 16Gb is enough only for very primitive LLM on level of very simple translator or encyclopedia, but never any code writing or even long essay creation (Llama3 70billion need 80Gb RAM, chinese Deepseek v2 code writer with 300+ languages 236billions need more than 128Gb RAM).
Even if you want to sell it - you can't - they don't sell hardware to your clients, if you want to run - you can't, they don't sell hardware to you. They want lock everyone to the cloud which is death to privacy & security altogether in society.
Try to buy in ordinary store a server's 12 RAM slots motherboard, there's even weird CPU slot, so you must use only proprietary non consumer cooling.
The cost out weigh the benefits while the tech is there for 80gb ram. Getting that to run on a laptop with small fans and low heat. Basically impossible. That’s why the cloud is the main route. You get your processing power without the drawbacks.