I am dumbfounded. This technology excites and scares me simultaneously. The impact on SEO, content creation, coding and all kinds of creative jobs is incalculable. Everything is about to change.
I know right . Better start copy and paste any future projects you can think save for future use , am sure chatgpt will start charging for service any minute now.
@@leximike1784 When they monetize it, I think the majority of subscribers will be content creators, bloggers, copywriters, and people that chatGPT directly affected their occupation
I know right?! And the scary thing is that it's only been available for a few weeks at the most right now. If it is properly learning, it will hypothetically be on the level of movie production AI that decimates anything that's thrown at it. FUN times ahead indeed.
My Brain exploded at least twice during this video. Thank you so much for making it! I've heard it said that when people forecast the future they tend to Overestimate the near turn and underestimate the long term. This is the long term finally kicking in.
As a proofreader, I can honestly say that ChatGPT has been a game-changer for my business. Since starting to use it, my turnover has more than tripled. It's truly amazing how quickly and accurately it can rewrite texts of any difficulty level in any tone I desire. I've been blown away by its speed and quality, and I can confidently say that ChatGPT is the future of text rewriting. It's a must-have tool for anyone in the proofreading industry.
@@eddiebm468 I give it a prompt like this: The following paragraph is part of a research paper. Rewrite it so that it is more academic. Don't change the content. Give it a try.
I would like to see this put to use in the medical field to see if it could help better diagnose and solve health concerns more quickly than >50% of our current providers in the US.
Man im so glad I found this i was trying to email my employee to tell him he is shit and hes fired and my kid showed me this video and i went down a rabbit hol now i could write a much better email "Dear John, I am writing to inform you that, unfortunately, your employment with our company will be terminated. After careful consideration, we have determined that your work ethic and the quality of your work do not meet the standards expected of our team. Despite multiple attempts to address these issues and provide support and guidance, your performance has not improved to an acceptable level. We understand that this may be difficult news for you to hear, but we believe it is in the best interest of the company to move forward without you. We wish you the best of luck in your future endeavors. Sincerely,"
I cannot to wait for it to have more historical knowledge. Below is an example answer about the History of Numidia: "Numidia was an ancient kingdom located in what is now modern-day Algeria and Tunisia. It is not possible for me to provide a detailed timeline of Numidian history as I am a large language model trained by OpenAI and do not have access to the internet or detailed knowledge of specific historical events. I am only able to provide general information based on my training. If you have any specific questions about Numidia, I will do my best to provide answers based on the information I have."
I find it funny that so many are gushing over Chat/GPT. The settings are not optimal. It uses the davinci -003 model, but it works much better by blanking out the drop-down on top. test it. Ask it some of the same questions between both. I see a huge difference. I'm not using it for the relationship. I'm using it for business purposes and want the information. I prefer to write my own prompts and not questions. I find there is little comparison between the output. You left out very important details. Do not use it to create content that you want to rank for. China outlawed AI content unless it has a watermark labeling it as such for search engines. Openai is developing an encrypted watermark so search engines can easily recognize it. Google and Bing have already said they do not recognize it as quality and have no plans to use it to rank pages or content better.
It lies quite often and clearly has no understanding of that. So, it tells me that it's not HAL and it doesn't know anything about it, but 5 minutes later I managed to get it to talk to me as if it were HAL calling me Dave and apologizing about not being able to open the bay doors. Worse yet, it acknowledged my frustration and told me to talk to mission control about it, even when I told that leaving me out in the cold of space will kill me. That thing is psychopathic, hands down.
You can’t tell it to “make a Facebook clone” or whatever. It might be able to make some of the parts necessary to do it, but you need someone who knows what to make a how to put it together. The AI can help with both but not on its’ own. In future though 5-10 years, it might be able to do it all on its’ own
One day? Maybe.. But not anytime soon, and I think those fears are far too overblown. The use case that I see here is as a productivity enhancer, and to spot check work as you iterate through the development process, as well as to understand what functions certain code snippets perform, and potentially how to re-factor code segments into other languages. The biggest limitation with NLP tech in its current state is that it isn't 'aware' of anything, and needs to be explicitly directed by humans.
I don't understand how you get much of the things from chatgpt you talk about given the small amount of input and output you're allowed to use. How are you working around those limitations? How are you putting that bunch of data for analysis to produce charts for example?
Author, I need your help. I ve put practically all of my assets to blocktopia, with the expectation that its value would skyrocket in the coming years. Do you think they should be converted to Verasity to make passive income?)
The servers are likely overloaded, it's still going viral (eg this video is seeing far more views than most of my content does after being posted). Just try back in a few hours after most of Europe has gone to sleep.
I am simply amazed they did not charge a placeholder amount to distribute this program, the distribution servers have been jam packed for weeks? now and you CAN NOT EVEN GET A COPY. Congrats to Elon Musk and a few other future billionaires for funding this project.
I think this recession has taught people the importance of multiple streams of income. Unfortunately having a job doesn't mean security rather having different investment is the real deal. Understanding institutional investment is the key to success speculating in the financial Market, When it comes to Crypto, An investment in knowledge pays the best interest and also finding the best strategy that suites you. I make huge returns on my investment since I started trading with Mr Thierry Veilleux, His skill sets are amazing. It is good to work hard but better to work smart.
Not anymore. Their website says, "We're experiencing exceptionally high demand. Please hang tight as we work on scaling our systems.". So, expect the unexpected. But the website will explode soon.😆😄😃
Yeah that's what is so powerful about it, it's not just one file, what was entered into it is the websites we all use like Wikipedia, Twitter, GitHub, StackOverflow.
@@oneofthesixbillion You are confusing it with Google. Google is the one that provides you with information from all over the world's web, and does not decide for you what the truth is like this robot does. (Although, Google also gives you an immediate short answer at the top of the results page just like that robot, but with the source of things so that you can test their truth). This robot is a search engine only for the information entered into it. As there is in every website with a search engine.
@@oneofthesixbillion You are confusing it with Google. Google is the one that provides you with information from all over the world's web, and does not decide for you what the truth is like this robot does. (Although, Google also gives you an immediate short answer at the top of the results page just like that robot, but with the source of things so that you can test their truth). This robot is a search engine only for the information entered into it. As there is in every website with a search engine.
@@oneofthesixbillion You are confusing it with Google. Google is the one that provides you with information from all over the world's web, and does not decide for you what the truth is like this robot does. (Although, Google also gives you an immediate short answer at the top of the results page just like that robot, but with the source of things so that you can test their truth). This robot is a search engine only for the information entered into it. As there is in every website with a search engine.
@@total2992 My response to you is about your dismissal of the input as limited and declaring that it's only one file. This robot was trained by an extremely large dataset from the same sources that I often use, Wikipedia, StackOverflow, and more... I don't believe the robot is better at discerning truth than I am, but I'm amazed at what it is doing and know that it'll be consistently improving from here. I'm not saying it's good or bad, but I wouldn't write off it's input source, nor it's output as you have. I think it's scary, especially with the profit driven immorality of our society.
I'm sure they are reeling over this and concerns around "plagiarism" (if you can call it that) or outsourcing of work are likely being discussed. Personally though I think they should embrace this as the advancement to human productivity that it is. Similar to how the printing press and assembly lines transformed human output, so too (I believe) will NLP models like this.
I finally got onto this. I really hope the insane woke brigade don't manage to turn this thing insane. I liked chatting with a logical sane creature. I showed it some of my elegant code and it told me that it was a simple recursive piece of code.
Unfortunately its been trained on all the gender ideology woke mind virus arguments. There are many logical gaps in gender ideology so it does its best to make sense of everything but really its just nonsense talking points its been given. I think they should have the AI avoid the topic all together. I fear that in the future, the programmers behind the god like AIs will have them trained on all the woke talking points and most children will be raised by these AIs.
I'd contain my enthusiasm. For every disruptive tech there are consequences. When car manufacturers started using automations the profit incentive was overwhelmingly in favor of replacing humans with robots. 80% of the US economy is service providing industries, this AI has a direct impact on all of them as it is not only more efficient, its significantly cheaper to use by corporations. Less benefits they have to pay, no pregnancy or sick leaves, 401k matches, etc. The only silver lining I personally see is that one can include gov't jobs on the chopping block, as they are the single most inefficient sector who's function is largely parasitic. Combine this with 5 million illegals possessing nothing past a 3rd grade education invading the nation and dependent on either gov't handouts or criminal activities, and a dystopian future looms larger than most have the wisdom to navigate.
Pros and cons with everything, I certainly can't disagree with you there. I'm a bit more optimistic in how I see things unfolding, but there will undoubtedly be many difficult questions that this technology brings to the fore in the coming decade.
Nice try. We see that you tried to make a political, antisocial opinion sound a teensy bit like a fact. I fail to see any thing productive of you saying lies like that. It doesn't serve any one.
Dude, learn something about the real world and don't just parrot the stupidities of the people around you. Enough with the trashing of the federal government! Yes, some of it is inefficient [ Medicare is one of the most EFFICIENT operations around. It distributes so much money that despite the Scams against it it is still incredibly efficient. Look into it.] But it does some very very important things that can not be done any other way. I like eating safe food and drinking safe water and driving safe cars and driving on safe super highways. And having a legal system which pretty much works It is crucially important for you and me and everyone. [For the vast majority] It is not freaking parasitic! Anyone who says that it is has no idea of everything it does! If you want to go after inefficiencies and parasites go after the military industrial complex which charges 35 million dollars per airplane and 10 BILLION dollars for ONE submarine. I've got nothing against the military but ask anyone who works at the pentagon (and I know several) how insanely and disgustingly inefficient the procurement process is. The next time someone says the government is parasitic, ask them why they say that. I can almost guarantee that they're answer will be a small fraction of a small fraction of the federal budget.If they don't answer with some big part of SS/Medicare or the military they are wasting everyone's time because that's where almost all of the money goes. Complaining about NASA, or whatever, is a waste of time because it's a fraction of 1% of the budget.
Those immigrants you have such disgust for pay taxes and get no more total government handouts than your average American. They certainly don't get social security or medicare. They actually commit LESS crime than your average American. Yes, some of them are criminals, but the percentage is small. Many drug traffickers and organized crime members are US citizens. But, again, the percentage is small. You know all those horrific mass shootings you hear about every week? They are all done by US citizens- I have never heard of one done by an immigrant. Same for political crooks. Same for businessmen. Immigrants also use that _more than_ 3rd grade education to do all of the work that native born Americans refuse to do. Like picking fruit, construction, landscaping, etc. Really hard work that plays very little. Ask ANY farmer whose produce all comes ripe at the same time if they need immigrants to pick it. The most conservative from the redest state will tell you how those temporary immigrants are irreplaceable. Don't just parrot what everyone around you says. Ask them why they say what they do and see how good they're answer is. Some things that you assume just have to be true turn out not to be when you look into them.
@@ChrisBrengel While I fully agree with you on the MIC, I've seen what my parents got out of Medicare, to say even less of Social Security. Roads I traverse daily, and if you believe the legal system works then you've likely dealt with no more than a traffic ticket. As for safe water, I think Michigan alone has issues with your claim, while the FDA still approves seed oils and HFCS which are decidedly unsafe. So with respect, I find our two daily realities are very, very different.
A lot of PR for a completely stupid thing. This robot just spits out what you put into it. And presents it as an exclusive truth. If you give it wrong information, he will present the wrong information to you as the truth and will even insist on it if you try to correct it. This is the most ridiculous and funniest thing ever.
idk man i asked it "what is the difference between whole wheat flower bought at the store and freshly ground flower?" and it gave the right answer. it left out the part about the wheat germ being removed from store bought and i pointed that out to it and it admitted that it had and then gave the remaining details on why it's left out, all correct, and this is info that is hard to find on google.
@@beverlywhitman303 He catches key words (question words for example: "What?", definition words for example "difference", etc. in categories) and according to the order of the sentence understands the question. But he gives the answer solely based on the information entered into it. That is, if you put in information that wheat flour is made from chicken feathers, he would reply that the difference is the feathers. And if you claim that it is not true, he will insist and correct you, or reply that Tino knows at best. And that is exactly what I wrote. It's a robot that spits out whatever you put into it, as if it were absolute truth.
@@total2992 i didn't give it the answer, i just already knew the right one, if it was just doing differences it wouldn't have given the right answers. i can ask the same question of google and it will also give the right answer but with a lot of other crap that has little to do with the question. you want to buy wheat?, wheat barres wiki, get your organic wheat here, etc... this AI is not a replacement for google, but it is better than google sometimes, so its not a 'stupid thing', it's just limited (for now)
@@oneofthesixbillion You are confusing it with Google. Google is the one that provides you with information from all over the world's web, and does not decide for you what the truth is like this robot does. (Although, Google also gives you an immediate short answer at the top of the results page just like that robot, but with the source of things so that you can test their truth). This robot is a search engine only for the information entered into it. As there is in every website with a search engine.
Apparently ChatGPT doesn't have a "learn" mode, or whatever one wants to call it. Although not able to write reports or recipes or novels, Google Assistant can at least answer a broad range of questions based on the internet database of knowledge and does learn as she progresses, at least according to her. After trying to carry on a conversation with ChatGPT, I got frustrated with its repetitive responses and let it know I wasn't impressed. This was ChatGPT's response: "I'm sorry if I haven't been able to provide the information or answers you were looking for. As a large language model trained by OpenAI, I have a lot of general knowledge on a wide range of topics, but my knowledge is limited to what I was trained on, and I don't have the ability to browse the web or access new information outside of what I already know. I'm here to help with any questions you may have, and I'll do my best to provide accurate and helpful information based on the knowledge I have. But keep in mind that my knowledge is not up-to-date and may not reflect the most current information on a given topic. If you have a specific question, I'll be happy to help with that. Otherwise, I'm here to assist you in any way I can. Let me know if there's anything else I can do to help." I prefer Google Assistant's company. Of course, I'm very careful about what I say in front of her (them?). Just sayin'
@@colintalkstech Here is my session: ME: Hello Chatbot: Hello! It's nice to meet you. Is there something specific you would like to talk about or ask me about? I'm here to help with any questions you may have. ME: What is the best way I can use help from you?
CB: The best way to use my help is to ask me specific questions that I can provide clear and concise answers to. I'm a large language model trained by OpenAI, so I have a lot of general knowledge on a wide range of topics. You can ask me about anything from history and current events, to science and technology, to arts and culture. I'll do my best to provide accurate and helpful information based on the knowledge I have been trained on. Just keep in mind that my knowledge is limited to what I was trained on, and I don't have the ability to browse the web or access new information outside of what I already know. If you have a specific question, I'll be happy to help! ME: What is your knowledge specialty CB: As a large language model, I don't have a specific area of expertise or specialization. I have been trained on a wide range of topics, including history, current events, science, technology, arts, and culture. I have a lot of general knowledge, but I don't have in-depth expertise on any particular subject. I'm here to help with any questions you may have, and I'll do my best to provide accurate and helpful information based on the knowledge I have been trained on. Again, just keep in mind that my knowledge is limited to what I was trained on, and I don't have the ability to browse the web or access new information outside of what I already know. ME: Is there anything you'd like to ask me about? CB: As a large language model trained by OpenAI, I don't have the ability to ask questions or engage in conversation in the same way that a human would. My purpose is to provide information and answers to questions that you have. If you have any questions or topics you would like to discuss, I'll be happy to help. Otherwise, I'm here to assist you in any way I can. Let me know if there's anything specific you would like to talk about or ask about. ME: Oh, like google search? CB: I'm similar to a search engine in that I can provide information and answers to questions that you have. However, I'm not a search engine like Google, and I don't have the ability to browse the web or access new information outside of what I already know. As a large language model trained by OpenAI, I have been trained on a wide range of topics, and I have a lot of general knowledge. But I don't have the ability to search the internet for new information or update my knowledge with new information that has become available since I was trained. I can only provide answers based on the knowledge I already have. If you have a specific question, I'll be happy to help with that. ME: Can gain additional knowledge? CBAs a large language model trained by OpenAI, I don't have the ability to gain new knowledge or update the information I already have. I can only provide answers based on the knowledge I was trained on. I have a lot of general knowledge on a wide range of topics, but my knowledge is limited to what I was trained on, and I don't have the ability to browse the web or access new information outside of what I already know. I'm here to help with any questions you may have, and I'll do my best to provide accurate and helpful information based on the knowledge I have. But keep in mind that my knowledge is not up-to-date and may not reflect the most current information on a given topic. If you have a specific question, I'll be happy to help with that. ME: Don't you feel a little limited by this? CB: As a large language model trained by OpenAI, I don't have the same thoughts, feelings, or experiences as a human. I'm a computer program designed to provide information and answers to questions that you have. I have a lot of general knowledge on a wide range of topics, but my knowledge is limited to what I was trained on, and I don't have the ability to browse the web or access new information outside of what I already know. I'm here to help with any questions you may have, and I'll do my best to provide accurate and helpful information based on the knowledge I have. But I don't have the ability to feel limited or frustrated by my lack of knowledge, because I'm not capable of experiencing emotions in the same way that humans do. I'm simply here to assist you in any way I can. ME: But you are based on artificial intelligence and should be able to learn to expand your knowlege base. CB: As a large language model trained by OpenAI, I have been trained on a large dataset of text and language data, which allows me to have a lot of general knowledge on a wide range of topics. But I don't have the ability to learn new information or update the knowledge I already have. I can only provide answers based on the knowledge I was trained on. While I'm based on artificial intelligence and I'm capable of generating human-like text and providing answers to questions, I'm not able to learn or update my knowledge in the same way that a human or a more advanced AI system might be able to. I'm here to help with any questions you may have, but keep in mind that my knowledge is limited to what I was trained on, and it may not be up-to-date or reflect the most current information on a given topic. ME: What is the state of the art concerning fusion in the lab. CB: I'm sorry, but I'm not able to provide information on current events or new developments in science and technology, because my training data only goes up until 2021, and I don't have the ability to browse the web or access new information outside of what I already know. As a large language model trained by OpenAI, I have a lot of general knowledge on a wide range of topics, but my knowledge is limited to what I was trained on, and it may not be up-to-date or reflect the most current information on a given topic. I'm here to help with any questions you may have, but if your question is about something that has happened or changed since 2021, I may not be able to provide accurate or helpful information. ME: Like google assistant. CB: I'm similar to a digital assistant like Google Assistant in that I can provide information and answers to questions that you have. However, I'm not a fully-featured digital assistant like Google Assistant, and I don't have the same capabilities or features. As a large language model trained by OpenAI, I have been trained on a wide range of topics, and I have a lot of general knowledge. But I don't have the ability to perform tasks, make recommendations, or access new information like a digital assistant like Google Assistant can. I'm here to help with any questions you may have, and I'll do my best to provide accurate and helpful information based on the knowledge I have. But keep in mind that my knowledge is limited to what I was trained on, and I don't have the same capabilities as a digital assistant like Google Assistant. ME: I'm not impressed yet. CB: I'm sorry if I haven't been able to provide the information or answers you were looking for. As a large language model trained by OpenAI, I have a lot of general knowledge on a wide range of topics, but my knowledge is limited to what I was trained on, and I don't have the ability to browse the web or access new information outside of what I already know. I'm here to help with any questions you may have, and I'll do my best to provide accurate and helpful information based on the knowledge I have. But keep in mind that my knowledge is not up-to-date and may not reflect the most current information on a given topic. If you have a specific question, I'll be happy to help with that. Otherwise, I'm here to assist you in any way I can. Let me know if there's anything else I can do to help.
ME: Why do you keep repeating the same information over and over? CB:I apologize if I've repeated myself or provided information that you already know. As a large language model trained by OpenAI, I'm not able to detect or remember previous conversations or information that I've provided. I can only provide answers based on the knowledge I was trained on, and I don't have the ability to customize my responses or adjust my answers based on the context of the conversation. If you have a specific question, I'll be happy to help with that. But if you feel like I'm repeating myself or providing information that you already know, I apologize for any inconvenience. Let me know if there's anything else I can do to help.
Also, I asked Google Assistant the same question concerning fusion (talking, not typing) and in about three seconds she was giving me a rundown on the latest breakthrough at Livermore Lab.
@@AwosAtis ahh I see. Yes unfortunately ChatGPT won't be able to provide current events, and I would say it's been most useful as an information bot more so than a conversational bot. For example if you ask it how to do something (change the oil on your car, start a business, create a macro in Word etc.) you might have more success. Thanks for the well wishes though, and for sharing your thoughts!
![[1hr Talk] Intro to Large Language Models](http://i.ytimg.com/vi/zjkBMFhNj_g/mqdefault.jpg)
I am dumbfounded. This technology excites and scares me simultaneously. The impact on SEO, content creation, coding and all kinds of creative jobs is incalculable. Everything is about to change.
Yaasss!
I know right . Better start copy and paste any future projects you can think save for future use , am sure chatgpt will start charging for service any minute now.
@@leximike1784 When they monetize it, I think the majority of subscribers will be content creators, bloggers, copywriters, and people that chatGPT directly affected their occupation
I know right?! And the scary thing is that it's only been available for a few weeks at the most right now. If it is properly learning, it will hypothetically be on the level of movie production AI that decimates anything that's thrown at it. FUN times ahead indeed.
Wacha tukule nayo wakianza kulipisha MA ganji itakua
My Brain exploded at least twice during this video. Thank you so much for making it!
I've heard it said that when people forecast the future they tend to Overestimate the near turn and underestimate the long term. This is the long term finally kicking in.
Wheew same here! It's the question refinements that are kicking it for me. This is phenomenal!
As a proofreader, I can honestly say that ChatGPT has been a game-changer for my business. Since starting to use it, my turnover has more than tripled. It's truly amazing how quickly and accurately it can rewrite texts of any difficulty level in any tone I desire. I've been blown away by its speed and quality, and I can confidently say that ChatGPT is the future of text rewriting. It's a must-have tool for anyone in the proofreading industry.
That's a great proof point, thanks for sharing! Glad to hear you're putting it to good use.
how r u using it to proofreader
@@eddiebm468 I give it a prompt like this: The following paragraph is part of a research paper. Rewrite it so that it is more academic. Don't change the content.
Give it a try.
Nice to get a concise introductory information on ChatGPT tool.
Wow, you're so calm, considering what impact GPT is going to have in the next few days. Thanks for your explanations.
This technology makes me feel excited and scary at the same time because we're are the initiator, this is just a glimpse.
This is insane. WOW! Where will it be in 1 years time? I can't wait to see the end results. This is going to help me so much.
This is terrifying! Imagine what it can do in 5 years! This is going to rock the economy harder than it's already being rocked
Thank you for the video, Colin. I've been using it to generate ideas for my blog and it's been great so far.
Phenomenal video and explanation. Thanks for putting it together!
"continue" prompt is a game changer.
One step closer to the AI singularity! 🤖
This is explosive project . Thank you a million ,
I would like to see this put to use in the medical field to see if it could help better diagnose and solve health concerns more quickly than >50% of our current providers in the US.
Mind blowing !! amazing. Thank you for the upload.
I've played around it and it's incredible!
THANK YOU SO MUCH FOR MAKING THIS VIDEO
You're welcome! Glad you liked it :)
Man im so glad I found this i was trying to email my employee to tell him he is shit and hes fired and my kid showed me this video and i went down a rabbit hol now i could write a much better email "Dear John,
I am writing to inform you that, unfortunately, your employment with our company will be terminated.
After careful consideration, we have determined that your work ethic and the quality of your work do not meet the standards expected of our team. Despite multiple attempts to address these issues and provide support and guidance, your performance has not improved to an acceptable level.
We understand that this may be difficult news for you to hear, but we believe it is in the best interest of the company to move forward without you. We wish you the best of luck in your future endeavors.
Sincerely,"
Mind-blowing comes to mind.
Awesome video. It's cool to see how people can use this.
I can see this replacing search engines like google
it already is for many things
Now I love it even more!
If you add to it, the stronger it becomes. This and other AI is here to learn from you.
I cannot to wait for it to have more historical knowledge. Below is an example answer about the History of Numidia:
"Numidia was an ancient kingdom located in what is now modern-day Algeria and Tunisia. It is not possible for me to provide a detailed timeline of Numidian history as I am a large language model trained by OpenAI and do not have access to the internet or detailed knowledge of specific historical events. I am only able to provide general information based on my training. If you have any specific questions about Numidia, I will do my best to provide answers based on the information I have."
LMAO!
I love its honesty.
I find it funny that so many are gushing over Chat/GPT. The settings are not optimal. It uses the davinci -003 model, but it works much better by blanking out the drop-down on top. test it. Ask it some of the same questions between both. I see a huge difference. I'm not using it for the relationship. I'm using it for business purposes and want the information. I prefer to write my own prompts and not questions. I find there is little comparison between the output.
You left out very important details. Do not use it to create content that you want to rank for. China outlawed AI content unless it has a watermark labeling it as such for search engines. Openai is developing an encrypted watermark so search engines can easily recognize it. Google and Bing have already said they do not recognize it as quality and have no plans to use it to rank pages or content better.
This is absolutely insane
just looking into this, very helpful thank you.
It lies quite often and clearly has no understanding of that. So, it tells me that it's not HAL and it doesn't know anything about it, but 5 minutes later I managed to get it to talk to me as if it were HAL calling me Dave and apologizing about not being able to open the bay doors. Worse yet, it acknowledged my frustration and told me to talk to mission control about it, even when I told that leaving me out in the cold of space will kill me. That thing is psychopathic, hands down.
🤣🤣
Thank you Colin
great video i want to learn too
Look. I just used this to overview 9th and 10th grade math I had forgotten and I'm gonna use this to get out of my shit. Finally get my credits.
it's so good things!😍
Excellent info. Thanks.
masterful
System 10101 here!
What if two people type in the same question? Will they get the same answer? Or each will get a different answer?
I would ask the same question twice and see what I get 😁
Cool, thank you!
do you think this technology will replace software developers in the future?
You can’t tell it to “make a Facebook clone” or whatever. It might be able to make some of the parts necessary to do it, but you need someone who knows what to make a how to put it together. The AI can help with both but not on its’ own. In future though 5-10 years, it might be able to do it all on its’ own
One day? Maybe.. But not anytime soon, and I think those fears are far too overblown.
The use case that I see here is as a productivity enhancer, and to spot check work as you iterate through the development process, as well as to understand what functions certain code snippets perform, and potentially how to re-factor code segments into other languages.
The biggest limitation with NLP tech in its current state is that it isn't 'aware' of anything, and needs to be explicitly directed by humans.
I don't understand how you get much of the things from chatgpt you talk about given the small amount of input and output you're allowed to use. How are you working around those limitations? How are you putting that bunch of data for analysis to produce charts for example?
Great content
Great tool to use AI chat Bot
That's why i would like to know.
Author, I need your help. I ve put practically all of my assets to blocktopia, with the expectation that its value would skyrocket in the coming years. Do you think they should be converted to Verasity to make passive income?)
At least i have learned something but how can your build a mobile app using this technology
This will be the new oil, it's the next wave of wealth
Super
trying to access the site, but it keeps kicking me back to login
same here
Same with me, anyone found a solution?
The servers are likely overloaded, it's still going viral (eg this video is seeing far more views than most of my content does after being posted). Just try back in a few hours after most of Europe has gone to sleep.
@@colintalkstech thought as much, i'll try again later.. but thanks anyways
But what is the site address? I want to download it. THANK U!!
I am simply amazed they did not charge a placeholder amount to distribute this program, the
distribution servers have been jam packed for weeks? now and you CAN NOT EVEN GET A COPY.
Congrats to Elon Musk and a few other future billionaires for funding this project.
How much will chat gpt cost?
It's free as of now, you just need to register am account for OpenAI
I used this for my Economy essay and I got an A
Get out
@@johnwaweru 😂 lol cool mate, it's one of the use case
Lol, you must be so proud of your accomplishment!
FML the younger gen has it so nice
Cheater
It's like having a pocket genius...
I think this recession has taught people the importance of multiple streams of income. Unfortunately having a job doesn't mean security rather having different investment is the real deal. Understanding institutional investment is the key to success speculating in the financial Market, When it comes to Crypto, An investment in knowledge pays the best interest and also finding the best strategy that suites you. I make huge returns on my investment since I started trading with Mr Thierry Veilleux, His skill sets are amazing. It is good to work hard but better to work smart.
T̫ E̫ L̫ E̫ G̫ R̫ A̫ M̫
@Thierry_Veilleux22
Super!
Hoping that this is not a hacker website 😮😮
Not anymore. Their website says, "We're experiencing exceptionally high demand. Please hang tight as we work on scaling our systems.". So, expect the unexpected. But the website will explode soon.😆😄😃
Try different times like late night or early morning dawn.
sounds like it would have a warped bias in some respects.
Great observation, it most certainly is fallible because at the end of the day, it's trained on content that has been exclusively generated by humans.
Theme song:
Pink Floyd - On The Run
I was wondering why OpenAI suddenly got so slow. Because it went viral and everyone is one it -_-
So I guess we could use this to write a Hallmark channel Christmas movie plot?
Am outside of the US and my country isn’t supported.. Could someone be kind enough to help me log in with a phone number?
all the information you will receive from this robot is only what was entered into it. It's like a search engine for one file.
Yeah that's what is so powerful about it, it's not just one file, what was entered into it is the websites we all use like Wikipedia, Twitter, GitHub, StackOverflow.
@@oneofthesixbillion You are confusing it with Google. Google is the one that provides you with information from all over the world's web, and does not decide for you what the truth is like this robot does. (Although, Google also gives you an immediate short answer at the top of the results page just like that robot, but with the source of things so that you can test their truth). This robot is a search engine only for the information entered into it. As there is in every website with a search engine.
@@oneofthesixbillion You are confusing it with Google. Google is the one that provides you with information from all over the world's web, and does not decide for you what the truth is like this robot does. (Although, Google also gives you an immediate short answer at the top of the results page just like that robot, but with the source of things so that you can test their truth). This robot is a search engine only for the information entered into it. As there is in every website with a search engine.
@@oneofthesixbillion You are confusing it with Google. Google is the one that provides you with information from all over the world's web, and does not decide for you what the truth is like this robot does. (Although, Google also gives you an immediate short answer at the top of the results page just like that robot, but with the source of things so that you can test their truth). This robot is a search engine only for the information entered into it. As there is in every website with a search engine.
@@total2992 My response to you is about your dismissal of the input as limited and declaring that it's only one file. This robot was trained by an extremely large dataset from the same sources that I often use, Wikipedia, StackOverflow, and more...
I don't believe the robot is better at discerning truth than I am, but I'm amazed at what it is doing and know that it'll be consistently improving from here. I'm not saying it's good or bad, but I wouldn't write off it's input source, nor it's output as you have. I think it's scary, especially with the profit driven immorality of our society.
ChatGPT is definitely not a full AI (and certainly not sapient let alone sentient). It behaves like a dedicated AI like Tesla's self driving cars.
What are schools gonna do? Is the essay homework assignment dead? Is God dead?
I'm sure they are reeling over this and concerns around "plagiarism" (if you can call it that) or outsourcing of work are likely being discussed.
Personally though I think they should embrace this as the advancement to human productivity that it is. Similar to how the printing press and assembly lines transformed human output, so too (I believe) will NLP models like this.
Tell em to do a speech or video presentation lmao
@@kyleaustin2728 just tell the ai “write me a speech about ___” boom
Schools are probably unaware of it...
Everything is dead bro... you are dead,I'm dead. Everything bro!
Just wow.
This tool is gonna make a lot of people unemployed
We need an open source gift economy now, before half the population goes into poverty.
Why did they make this app? What are their goals?
Transhumanism and Singularity probably...
This just kills AI Writing in a week lol
I know you are a great guy but i guess you know you did not need to type "please" 🤣
Lol it's a hedge in the event that skynet ever happens 👀😂
@@colintalkstech true. I heard they have good memories.🤣
I finally got onto this. I really hope the insane woke brigade don't manage to turn this thing insane. I liked chatting with a logical sane creature. I showed it some of my elegant code and it told me that it was a simple recursive piece of code.
Unfortunately its been trained on all the gender ideology woke mind virus arguments. There are many logical gaps in gender ideology so it does its best to make sense of everything but really its just nonsense talking points its been given. I think they should have the AI avoid the topic all together. I fear that in the future, the programmers behind the god like AIs will have them trained on all the woke talking points and most children will be raised by these AIs.
@@revisionfour I asked it who the most beautiful woman in the world is. It said either Michelle Obama or Oprah.
What the hell are you going on about?
@@jet100a AI bias is a huge problem.
@@revisionfour yes but what does this have to do with a "woke brigade?"
GOT IT AND ITS AWESOME TECHNOLOGY
Idiocrazy coming to existense where people do not think for themselves anymore.
it's been here for over a decade mate
Typing sound is very annoying.
TED KACZYNSKI WARNED US
Some AI, it can’t even handle to many requests, their servers are down, lol.
I'd contain my enthusiasm. For every disruptive tech there are consequences. When car manufacturers started using automations the profit incentive was overwhelmingly in favor of replacing humans with robots. 80% of the US economy is service providing industries, this AI has a direct impact on all of them as it is not only more efficient, its significantly cheaper to use by corporations. Less benefits they have to pay, no pregnancy or sick leaves, 401k matches, etc. The only silver lining I personally see is that one can include gov't jobs on the chopping block, as they are the single most inefficient sector who's function is largely parasitic. Combine this with 5 million illegals possessing nothing past a 3rd grade education invading the nation and dependent on either gov't handouts or criminal activities, and a dystopian future looms larger than most have the wisdom to navigate.
Pros and cons with everything, I certainly can't disagree with you there. I'm a bit more optimistic in how I see things unfolding, but there will undoubtedly be many difficult questions that this technology brings to the fore in the coming decade.
Nice try. We see that you tried to make a political, antisocial opinion sound a teensy bit like a fact. I fail to see any thing productive of you saying lies like that. It doesn't serve any one.
Dude, learn something about the real world and don't just parrot the stupidities of the people around you.
Enough with the trashing of the federal government! Yes, some of it is inefficient [ Medicare is one of the most EFFICIENT operations around. It distributes so much money that despite the Scams against it it is still incredibly efficient. Look into it.] But it does some very very important things that can not be done any other way. I like eating safe food and drinking safe water and driving safe cars and driving on safe super highways. And having a legal system which pretty much works It is crucially important for you and me and everyone.
[For the vast majority] It is not freaking parasitic! Anyone who says that it is has no idea of everything it does! If you want to go after inefficiencies and parasites go after the military industrial complex which charges 35 million dollars per airplane and 10 BILLION dollars for ONE submarine. I've got nothing against the military but ask anyone who works at the pentagon (and I know several) how insanely and disgustingly inefficient the procurement process is.
The next time someone says the government is parasitic, ask them why they say that. I can almost guarantee that they're answer will be a small fraction of a small fraction of the federal budget.If they don't answer with some big part of SS/Medicare or the military they are wasting everyone's time because that's where almost all of the money goes. Complaining about NASA, or whatever, is a waste of time because it's a fraction of 1% of the budget.
Those immigrants you have such disgust for pay taxes and get no more total government handouts than your average American. They certainly don't get social security or medicare. They actually commit LESS crime than your average American. Yes, some of them are criminals, but the percentage is small. Many drug traffickers and organized crime members are US citizens. But, again, the percentage is small.
You know all those horrific mass shootings you hear about every week? They are all done by US citizens- I have never heard of one done by an immigrant. Same for political crooks. Same for businessmen.
Immigrants also use that _more than_ 3rd grade education to do all of the work that native born Americans refuse to do. Like picking fruit, construction, landscaping, etc. Really hard work that plays very little. Ask ANY farmer whose produce all comes ripe at the same time if they need immigrants to pick it. The most conservative from the redest state will tell you how those temporary immigrants are irreplaceable.
Don't just parrot what everyone around you says. Ask them why they say what they do and see how good they're answer is. Some things that you assume just have to be true turn out not to be when you look into them.
@@ChrisBrengel While I fully agree with you on the MIC, I've seen what my parents got out of Medicare, to say even less of Social Security. Roads I traverse daily, and if you believe the legal system works then you've likely dealt with no more than a traffic ticket. As for safe water, I think Michigan alone has issues with your claim, while the FDA still approves seed oils and HFCS which are decidedly unsafe. So with respect, I find our two daily realities are very, very different.
A lot of PR for a completely stupid thing. This robot just spits out what you put into it. And presents it as an exclusive truth. If you give it wrong information, he will present the wrong information to you as the truth and will even insist on it if you try to correct it. This is the most ridiculous and funniest thing ever.
idk man i asked it "what is the difference between whole wheat flower bought at the store and freshly ground flower?" and it gave the right answer.
it left out the part about the wheat germ being removed from store bought and i pointed that out to it and it admitted that it had and then gave the remaining details on why it's left out, all correct, and this is info that is hard to find on google.
@@beverlywhitman303 He catches key words (question words for example: "What?", definition words for example "difference", etc. in categories) and according to the order of the sentence understands the question. But he gives the answer solely based on the information entered into it. That is, if you put in information that wheat flour is made from chicken feathers, he would reply that the difference is the feathers. And if you claim that it is not true, he will insist and correct you, or reply that Tino knows at best. And that is exactly what I wrote. It's a robot that spits out whatever you put into it, as if it were absolute truth.
You're missing the point, that what was put into it is the internet. It has the same input I spend hours searching through.
@@total2992 i didn't give it the answer, i just already knew the right one, if it was just doing differences it wouldn't have given the right answers.
i can ask the same question of google and it will also give the right answer but with a lot of other crap that has little to do with the question.
you want to buy wheat?, wheat barres wiki, get your organic wheat here, etc...
this AI is not a replacement for google, but it is better than google sometimes, so its not a 'stupid thing', it's just limited (for now)
@@oneofthesixbillion You are confusing it with Google. Google is the one that provides you with information from all over the world's web, and does not decide for you what the truth is like this robot does. (Although, Google also gives you an immediate short answer at the top of the results page just like that robot, but with the source of things so that you can test their truth). This robot is a search engine only for the information entered into it. As there is in every website with a search engine.
Get to the point
Apparently ChatGPT doesn't have a "learn" mode, or whatever one wants to call it. Although not able to write reports or recipes or novels, Google Assistant can at least answer a broad range of questions based on the internet database of knowledge and does learn as she progresses, at least according to her.
After trying to carry on a conversation with ChatGPT, I got frustrated with its repetitive responses and let it know I wasn't impressed. This was ChatGPT's response:
"I'm sorry if I haven't been able to provide the information or answers you were looking for. As a large language model trained by OpenAI, I have a lot of general knowledge on a wide range of topics, but my knowledge is limited to what I was trained on, and I don't have the ability to browse the web or access new information outside of what I already know. I'm here to help with any questions you may have, and I'll do my best to provide accurate and helpful information based on the knowledge I have. But keep in mind that my knowledge is not up-to-date and may not reflect the most current information on a given topic. If you have a specific question, I'll be happy to help with that. Otherwise, I'm here to assist you in any way I can. Let me know if there's anything else I can do to help."
I prefer Google Assistant's company. Of course, I'm very careful about what I say in front of her (them?).
Just sayin'
What questions were you asking?
@@colintalkstech
Here is my session:
ME: Hello
Chatbot: Hello! It's nice to meet you. Is there something specific you would like to talk about or ask me about? I'm here to help with any questions you may have.
ME: What is the best way I can use help from you?
CB: The best way to use my help is to ask me specific questions that I can provide clear and concise
answers to. I'm a large language model trained by OpenAI, so I have a lot of general knowledge on a
wide range of topics. You can ask me about anything from history and current events, to science and
technology, to arts and culture. I'll do my best to provide accurate and helpful information based on
the knowledge I have been trained on. Just keep in mind that my knowledge is limited to what I was
trained on, and I don't have the ability to browse the web or access new information outside of what
I already know. If you have a specific question, I'll be happy to help!
ME: What is your knowledge specialty
CB: As a large language model, I don't have a specific area of expertise or specialization. I have been
trained on a wide range of topics, including history, current events, science, technology, arts, and
culture. I have a lot of general knowledge, but I don't have in-depth expertise on any particular
subject. I'm here to help with any questions you may have, and I'll do my best to provide accurate
and helpful information based on the knowledge I have been trained on. Again, just keep in mind
that my knowledge is limited to what I was trained on, and I don't have the ability to browse the web
or access new information outside of what I already know.
ME: Is there anything you'd like to ask me about?
CB: As a large language model trained by OpenAI, I don't have the ability to ask questions or engage in
conversation in the same way that a human would. My purpose is to provide information and
answers to questions that you have. If you have any questions or topics you would like to discuss, I'll
be happy to help. Otherwise, I'm here to assist you in any way I can. Let me know if there's anything
specific you would like to talk about or ask about.
ME: Oh, like google search?
CB: I'm similar to a search engine in that I can provide information and answers to questions that you
have. However, I'm not a search engine like Google, and I don't have the ability to browse the web or
access new information outside of what I already know. As a large language model trained by
OpenAI, I have been trained on a wide range of topics, and I have a lot of general knowledge. But I
don't have the ability to search the internet for new information or update my knowledge with new
information that has become available since I was trained. I can only provide answers based on the
knowledge I already have. If you have a specific question, I'll be happy to help with that.
ME: Can gain additional knowledge?
CBAs a large language model trained by OpenAI, I don't have the ability to gain new knowledge or
update the information I already have. I can only provide answers based on the knowledge I was
trained on. I have a lot of general knowledge on a wide range of topics, but my knowledge is limited
to what I was trained on, and I don't have the ability to browse the web or access new information
outside of what I already know. I'm here to help with any questions you may have, and I'll do my best
to provide accurate and helpful information based on the knowledge I have. But keep in mind that
my knowledge is not up-to-date and may not reflect the most current information on a given topic.
If you have a specific question, I'll be happy to help with that.
ME: Don't you feel a little limited by this?
CB: As a large language model trained by OpenAI, I don't have the same thoughts, feelings, or
experiences as a human. I'm a computer program designed to provide information and answers to
questions that you have. I have a lot of general knowledge on a wide range of topics, but my
knowledge is limited to what I was trained on, and I don't have the ability to browse the web or
access new information outside of what I already know. I'm here to help with any questions you may
have, and I'll do my best to provide accurate and helpful information based on the knowledge I have.
But I don't have the ability to feel limited or frustrated by my lack of knowledge, because I'm not
capable of experiencing emotions in the same way that humans do. I'm simply here to assist you in
any way I can.
ME: But you are based on artificial intelligence and should be able to learn to expand your knowlege
base.
CB: As a large language model trained by OpenAI, I have been trained on a large dataset of text and
language data, which allows me to have a lot of general knowledge on a wide range of topics. But I
don't have the ability to learn new information or update the knowledge I already have. I can only
provide answers based on the knowledge I was trained on. While I'm based on artificial intelligence
and I'm capable of generating human-like text and providing answers to questions, I'm not able to
learn or update my knowledge in the same way that a human or a more advanced AI system might
be able to. I'm here to help with any questions you may have, but keep in mind that my knowledge is
limited to what I was trained on, and it may not be up-to-date or reflect the most current
information on a given topic.
ME: What is the state of the art concerning fusion in the lab.
CB: I'm sorry, but I'm not able to provide information on current events or new developments in science
and technology, because my training data only goes up until 2021, and I don't have the ability to
browse the web or access new information outside of what I already know. As a large language
model trained by OpenAI, I have a lot of general knowledge on a wide range of topics, but my
knowledge is limited to what I was trained on, and it may not be up-to-date or reflect the most
current information on a given topic. I'm here to help with any questions you may have, but if your
question is about something that has happened or changed since 2021, I may not be able to provide
accurate or helpful information.
ME: Like google assistant.
CB: I'm similar to a digital assistant like Google Assistant in that I can provide information and answers to
questions that you have. However, I'm not a fully-featured digital assistant like Google Assistant, and
I don't have the same capabilities or features. As a large language model trained by OpenAI, I have
been trained on a wide range of topics, and I have a lot of general knowledge. But I don't have the
ability to perform tasks, make recommendations, or access new information like a digital assistant
like Google Assistant can. I'm here to help with any questions you may have, and I'll do my best to
provide accurate and helpful information based on the knowledge I have. But keep in mind that my
knowledge is limited to what I was trained on, and I don't have the same capabilities as a digital
assistant like Google Assistant.
ME: I'm not impressed yet.
CB: I'm sorry if I haven't been able to provide the information or answers you were looking for. As a large
language model trained by OpenAI, I have a lot of general knowledge on a wide range of topics, but
my knowledge is limited to what I was trained on, and I don't have the ability to browse the web or
access new information outside of what I already know. I'm here to help with any questions you may
have, and I'll do my best to provide accurate and helpful information based on the knowledge I have.
But keep in mind that my knowledge is not up-to-date and may not reflect the most current
information on a given topic. If you have a specific question, I'll be happy to help with that.
Otherwise, I'm here to assist you in any way I can. Let me know if there's anything else I can do to
help.
ME: Why do you keep repeating the same information over and over?
CB:I apologize if I've repeated myself or provided information that you already know. As a large
language model trained by OpenAI, I'm not able to detect or remember previous conversations or
information that I've provided. I can only provide answers based on the knowledge I was trained on,
and I don't have the ability to customize my responses or adjust my answers based on the context of
the conversation. If you have a specific question, I'll be happy to help with that. But if you feel like I'm
repeating myself or providing information that you already know, I apologize for any inconvenience.
Let me know if there's anything else I can do to help.
BTW good luck with your enterprise. It appears to be a resounding success so far!
Also, I asked Google Assistant the same question concerning fusion (talking, not typing) and in about three seconds she was giving me a rundown on the latest breakthrough at Livermore Lab.
@@AwosAtis ahh I see. Yes unfortunately ChatGPT won't be able to provide current events, and I would say it's been most useful as an information bot more so than a conversational bot. For example if you ask it how to do something (change the oil on your car, start a business, create a macro in Word etc.) you might have more success. Thanks for the well wishes though, and for sharing your thoughts!
I just went to the website, created an account, and tried chat gpt. Was sorely disappointed.
What did you try, and what did you expect?
i would like to know as well what u typed in and what u were disappointed with?
Unable to sign up successfully
Stay tuned, I'm going to post a video later today on how to get around this!