I turned the GPT builder into a GPT that you can use for free here: chatgpt.com/g/g-jn619yw8Y-custom-gpt-builder Consider leaving a 5 star rating if you enjoy it. Thanks :)
at 4:48 in the line "# "Communication Sequence" outlines the step-b=y-step user interaction with GPTs." did you mean "# "Communication Sequence" outlines the step-by-step user interaction with GPTs." ?
The best prompt I ever used " Review your last response for improvements tell me what you have changed the reasoning why and rewrite it." The model will, as it says review slash response for improvements. It'll tell you the changes it made, the reasoning why. And then it will apply those changes and rewrite it very. powerful prompt. You should use it.
Ceasar was a dictator/consul of the republic. And the first modern version of dictator. Augustus, turned the roman republic into a dynastic/imperial state.
I built my own builder that I've successfully been using for months, but this is next level. Awesome mate. Thank you. Wish I could afford to sign up to your community. Exchange rate kills us Aussies.
What do you mean by "my own builder"? I've not messed with creating my own AI other than setting up a locally accessable iteration of Mistral, and that's only through PowerShell.
Negative prompts are really useful if you're getting generic outputs and want something more diverse. Open a new chat or edit the current prompt, repeating the original, and within that, tell it to specifically not mention a list of generic things it had mentioned in a past instance. This way, it won't have the history reinforcing its behavior.
I'd recommend leaving it. I clicked here before the bio, and it's really cool to see the creator helping us lowly commentors :P Also it actually made me look at the description. Just my two cents. Nice job on the video
I get fantastic results with this instruction in the customizations: Try to visualise steps of problems, as if you were to observe it through a camera/inner eye or as if drawing a sketch/picture of it
@@Kennyancat I read about a method to give LLMs reasoning where the model refines/checks the answere by an "inner perspective". Know, I don't know if this really does something compairable.
@@buzzm.8797 maybe you mean reflection? I gives an answer to itself, checks it, improves it and then gives it to the user. If it does this several times the quality of answers is always better.
Yes, it is possible to use a multimodal AI assistant that can both see your computer screen and respond to voice commands. This type of technology is often referred to as "vision-enabled voice assistants" or "visual voice assistants." Some popular examples of such assistants include Amazon Alexa with the Amazon Lookout for Gadgets service, Google Assistant with its Vision API integration, and Microsoft Cortana with the Windows Eye Control feature (which requires specialized eye-tracking hardware). These assistants can perform various tasks such as identifying objects on your screen, providing visual feedback based on voice commands, or even controlling your mouse and keyboard using only your voice. However, keep in mind that these features may require additional setup and configuration, including enabling accessibility settings, installing necessary software or services, and granting permissions to the AI assistant to access your computer's camera and microphone. Additionally, privacy concerns should be taken into consideration when using vision-enabled voice assistants, as they involve sharing more personal information than traditional text-based voice assistants. 😎🤖
Effective use of follow-up prompts and avoiding negative inputs are key to harnessing the full potential of AI, ensuring precise and valuable outputs for any professional setting. 🔑
I just want to say how much I appreciate this video and many others that you have done, along with all the work you put into helping others create their own GPT! It was super helpful and I have already created a custom GPT that is perfectly tuned to how I need to use it. Thank you! Also, your animated characters of yourself are so cool! Chat GPT never maintains the my actual look when creating animated versions and it also won’t edit real photos while keeping the original elements in them. Did you photoshop those yourself or use another AI software to help make them?
That's a great idea!Case studies are invaluable for understanding real-world applications. I'm particularly interested in the artificial intelligence area.
Much of these techniques should have already been taught from practiced communications in a business context, or any other context where thoughtful inquiry and discussion is concerned. Aside from that, knowing how to structure the prompts so that you use open and closing , referencing them as needed, and then embedding them down through as many layers as needed to organize prompts is another technique.
Superior content. You have Mad skills Igor. 👏🔥 The GPT builder is lit. You on a pink elephant 🐘💞🧞♀️😁. …. the penguin and the cat in a hat? Interesting menagerie 😁 Thank you 🙏
About Template, can you details with JSON ? Because in the JSON there is static attributes and dynamic variable. How do you detail the way to write these variable ?
I have made a custom GPT called Smart Vision image/text analysis it is 5x + smarter than base-model, it understands the complex Its unlike anything before and is ready completely for new vision features for 4o specifically, it will drastically improve how accurate the details seen can be conversed, and also an example i’ve been \using is upload an image of a cloud that looks like multiple things but it can be interpreted, the one i have made recognised it was a rabbit but not even a random human guessed it every time now on 1st shot so it knows when something is unusual about an image even if you dont say anything is, it can also do iq test image reasoning pattern questions relatively well It kind of even understands real logic games when giving good instruction but is limited there by the model not the instruction *IMPORTANT* just gotta follow the instructions given to get the right seed its 1 in 2 chance or so i have absolutely no idea why it needs that. Just paste the conversation starter and you’ll understand what to do.
I would like to give it a long list of addresses and have it search the web the pull back property info such as beds and baths. It does maybe 1 or2 addresses, but then tells me to manually look up the rest. How can I just have it look up all the addresses?
Hey there, great video but I can't create the GPT as its limited to 8000 characters wheras the instructions are around 12000 characters. Any advice please?
With that recommendation about negative prompts, I might need to reconsider my custom instructions, in which I give a list of words to avoid - those ones that ChatGPT tended to overuse which resulted in a big "this was made by AI" flag.
I have found the preface: “for the following (question, request, search,…) please omit results for…” and it cleaned up the disappearing Ceaser’s for me mostly in my searches.
I believe you are a leader in my opinions presented in this case it would behoove anyone else is unauthorized just wanted super professional leaders in my life 🧬
Can I use your prompts and especially your builder but later use it in my native language? Going to try of course but was wondering if anybody knew this already
Your "application to the business world" prompt appears to cause Chat GPT to hallucinate rather than give entirely accurate information. At the very least, if the area you put in those brackets is obscure.
@@aiadvantage A little specific but "christian indie games" and well it made up three companies that dont exist. "faithful interactive", "redemption games", and "Beacon Studios" Interestingly it seemed to tailor these made up companies to my interests that it has in its memory, very specifically. I question it a little further and it just gave me the one case study that's kind of obvious for the topic, That Dragon, Cancer by Numinous games, and a kids bible app which I wouldnt consider an indie game.
The templating technique does not work for me. The AI did not produce generic placeholders. Strange. Perhaps I have to be more concise. But good hint anyway. Thanks.
Thank you for offering tips and prompts to use, but your video is not very accessible for people who have vision problems. If I cannot see the video or read the screen, I have no way to know what prompts you are talking about or what the prompt says. You really should vocalize the words for the prompt, or at a minimum have a text available in the description or a link
Love the video title, insinuating that there aren't already dozens of videos about doing all the same things this guy is saying 99% of AI users don't know about. 🤣
Gen AI automates "low value" tasks to be able to invest more time in high value tasks, I do not worry about this. Sure, there is an impact in job loss but society has lack of skills and employees in many areas like healthcare, clean technologies, education and more. I think it makes society more productive
More Jobs are beign created with A.I as we become more efficient and there expose gaps in our society. We must Solve more space issues which is infinite so there are more Jobs if you really want to work, like mining ⛏️ on the moon 🌚 in a teleportic robot.
I don't understand the concept of "prompts" Isn't the whole idea to talk to the AI like a normal person? If we have to learn specific prompts, it's not real AI.
Much like talking to a person, the quality of the results you get depends on the quality of the communication that you input. If you don't have a healthy, structured conversation, the results aren't going to be as good, and the same goes for prompting AI.
Fun fact, if you want an image with Giger's style, you can't do that. But if you ask chatgpt to describe Giger's style as thoroughly as it can WITHOUT mentioning Giger's name, and then copy it and ask it to make a picture with that style, it will work :)
Keep up the great work on this channel are the winners of the Futur Chris Do The Right Thang 🙏 Poetic AuTexticTikTokJusticeBeeKeepersKeepinITReal Coding and it's far rightful owners daily and monthly Y B A Smukwich and I have taken the appropriate person responsible 😔🔙
I turned the GPT builder into a GPT that you can use for free here:
chatgpt.com/g/g-jn619yw8Y-custom-gpt-builder
Consider leaving a 5 star rating if you enjoy it. Thanks :)
at 4:48 in the line "# "Communication Sequence" outlines the step-b=y-step user interaction with GPTs."
did you mean "# "Communication Sequence" outlines the step-by-step user interaction with GPTs." ?
The quotation feature has allowed me to isolate points among a general response, and explore the details in complex subjects, thanks bro.
That's very generous of you and it's much appreciated. Glad I could help.
The best prompt I ever used " Review your last response for improvements tell me what you have changed the reasoning why and rewrite it." The model will, as it says review slash response for improvements. It'll tell you the changes it made, the reasoning why. And then it will apply those changes and rewrite it very. powerful prompt. You should use it.
I tell the LLM to "ask me 5 clarifying questions" at the end of my initial prompt. The quality of the outputs has improved as a result.
10:30 To be fair Cesar wasn't an Emperor. The first Emperor was is adopted son (August) so it's not a a bug or a negative consequence
Ceasar was a dictator/consul of the republic. And the first modern version of dictator. Augustus, turned the roman republic into a dynastic/imperial state.
I built my own builder that I've successfully been using for months, but this is next level. Awesome mate. Thank you. Wish I could afford to sign up to your community. Exchange rate kills us Aussies.
What do you mean by "my own builder"? I've not messed with creating my own AI other than setting up a locally accessable iteration of Mistral, and that's only through PowerShell.
Negative prompts are really useful if you're getting generic outputs and want something more diverse. Open a new chat or edit the current prompt, repeating the original, and within that, tell it to specifically not mention a list of generic things it had mentioned in a past instance. This way, it won't have the history reinforcing its behavior.
Geat Post! 👍
Thank you RUclips for teaching millions of people how to turn 2 minutes worth of usable information into a 30-minute video!
you need those views lol
Actually thank chat gpt😂 I'm sure they wrote him this long script with some kind of method to entice more views
I can't grasp the 2 minutes of info. There's a lot of filler talk in this video
This comment should have hundreds of thousands of likes.
You should put the prompts somewhere easily accessible, if you are going to suggest we can copy paste them
They are all in the description no?
@@aiadvantage I guess the description didnt update for me when the video dropped.
@@Yipper64 oh, not sure how that happened. Is it ok for you if I delete the comment then to reduce confusion for others?
I'd recommend leaving it. I clicked here before the bio, and it's really cool to see the creator helping us lowly commentors :P
Also it actually made me look at the description.
Just my two cents. Nice job on the video
@@bardnchill Alrighty, appreciate that you might be right
I get fantastic results with this instruction in the customizations:
Try to visualise steps of problems, as if you were to observe it through a camera/inner eye or as if drawing a sketch/picture of it
Uhh that is an interesting one! Thanks for sharing
can you give an example - I am curious
I think this functions the same as "go step by step", which has been proven to give better results.
@@Kennyancat I read about a method to give LLMs reasoning where the model refines/checks the answere by an "inner perspective". Know, I don't know if this really does something compairable.
@@buzzm.8797 maybe you mean reflection? I gives an answer to itself, checks it, improves it and then gives it to the user. If it does this several times the quality of answers is always better.
Best channel for Ai. Thanks for not wasting our time.
Yes, it is possible to use a multimodal AI assistant that can both see your computer screen and respond to voice commands. This type of technology is often referred to as "vision-enabled voice assistants" or "visual voice assistants."
Some popular examples of such assistants include Amazon Alexa with the Amazon Lookout for Gadgets service, Google Assistant with its Vision API integration, and Microsoft Cortana with the Windows Eye Control feature (which requires specialized eye-tracking hardware). These assistants can perform various tasks such as identifying objects on your screen, providing visual feedback based on voice commands, or even controlling your mouse and keyboard using only your voice.
However, keep in mind that these features may require additional setup and configuration, including enabling accessibility settings, installing necessary software or services, and granting permissions to the AI assistant to access your computer's camera and microphone. Additionally, privacy concerns should be taken into consideration when using vision-enabled voice assistants, as they involve sharing more personal information than traditional text-based voice assistants. 😎🤖
Thank you so much for these tips, and sharing the custom GPT setup instructions. I just used it to automate some emails that I send to new clients.
Effective use of follow-up prompts and avoiding negative inputs are key to harnessing the full potential of AI, ensuring precise and valuable outputs for any professional setting. 🔑
Julius Caesar never was Roman Emperor.
Wait a minute.... you are right! The point still works but the details of his rule escaped me. Shame on me
Farts
@@aiadvantage I do believe the dislikes on this video came from this particular mistake
I just want to say how much I appreciate this video and many others that you have done, along with all the work you put into helping others create their own GPT! It was super helpful and I have already created a custom GPT that is perfectly tuned to how I need to use it. Thank you!
Also, your animated characters of yourself are so cool! Chat GPT never maintains the my actual look when creating animated versions and it also won’t edit real photos while keeping the original elements in them. Did you photoshop those yourself or use another AI software to help make them?
Wow! Informative and important for all. Its very helpful for all traders. Specially who are beginner.
That's a great idea!Case studies are invaluable for understanding real-world applications. I'm particularly interested in the artificial intelligence area.
Any idea when this is coming to Microsoft Copilot? I thought MS Copilot was always supposed to get these first but it still doesn't have GPT-4o...
Julius was not an emperor… if that matters.
My male ego is in shambles... you are right 🥲
Wow you are good. Impressive catch 👏💯
Much of these techniques should have already been taught from practiced communications in a business context, or any other context where thoughtful inquiry and discussion is concerned. Aside from that, knowing how to structure the prompts so that you use open and closing , referencing them as needed, and then embedding them down through as many layers as needed to organize prompts is another technique.
Superior content. You have Mad skills Igor. 👏🔥 The GPT builder is lit.
You on a pink elephant 🐘💞🧞♀️😁. …. the penguin and the cat in a hat? Interesting menagerie 😁
Thank you 🙏
Turning RUclips into a zoo one video at a time 😁
The GPT builder works very well. Uploaded D&D adventures and does a great job of easily finding the right information for me.
Love to hear it :) I ran dozens of tests between this and the OpenAI builder and my builder is better every time. Enjoy
@@aiadvantage really appreciate all the effort! Looking forward to what will come in the future for AI.
nice ideas - thanks
If I have built a gpt myself can I use the case study prompts to provide real world examples or do I need to use the full chatgpt4o
Great input again Igor! Thanks
Love your video. How do you record yourself transparently in front of your screen?
About Template, can you details with JSON ?
Because in the JSON there is static attributes and dynamic variable.
How do you detail the way to write these variable ?
is there a gpt that could help you manage other gpts and be the base?
I have made a custom GPT called Smart Vision image/text analysis
it is 5x + smarter than base-model, it understands the complex
Its unlike anything before and is ready completely for new vision features for 4o specifically, it will drastically improve how accurate the details seen can be conversed,
and also an example i’ve been \using is upload an image of a cloud that looks like multiple things but it can be interpreted, the one i have made recognised it was a rabbit but not even a random human guessed it every time now on 1st shot so it knows when something is unusual about an image even if you dont say anything is, it can also do iq test image reasoning pattern questions relatively well
It kind of even understands real logic games when giving good instruction but is limited there by the model not the instruction
*IMPORTANT* just gotta follow the instructions given to get the right seed its 1 in 2 chance or so i have absolutely no idea why it needs that. Just paste the conversation starter and you’ll understand what to do.
Thank you very much for your sharing!
I would like to give it a long list of addresses and have it search the web the pull back property info such as beds and baths. It does maybe 1 or2 addresses, but then tells me to manually look up the rest. How can I just have it look up all the addresses?
Love the thumbnail - well done!
Does the reply to specific text in response work in mobile app? And how?
Hey there, great video but I can't create the GPT as its limited to 8000 characters wheras the instructions are around 12000 characters. Any advice please?
With that recommendation about negative prompts, I might need to reconsider my custom instructions, in which I give a list of words to avoid - those ones that ChatGPT tended to overuse which resulted in a big "this was made by AI" flag.
I have found the preface: “for the following (question, request, search,…) please omit results for…” and it cleaned up the disappearing Ceaser’s for me mostly in my searches.
you're so right bro
useful af. thanks
Brilliant video Igor
Is your email GPT available to the public? I love the idea of it.
I believe you are a leader in my opinions presented in this case it would behoove anyone else is unauthorized just wanted super professional leaders in my life 🧬
Can I use your prompts and especially your builder but later use it in my native language? Going to try of course but was wondering if anybody knew this already
This is tremendous…and it works with so many different subjects/inputs. Great vid and thanks for sharing. 🙏🏻💯
So glad to hear that as this video took waaay longer than expected to create. 🙏🏼
How to make thumbnail like that ?
Stable Diffusion Dreambooth + Photoshop for text
Thank you
this channel is always like watching b-movie horrors turning into c-movie :D
I am sad I didn't get to join today's events....😢 I was looking forward to meet you..
I replied to your comment, it is scheduled for thirsday not wednesday so you had the right time but 24 too early.
Thank you.
Aw man, that price was a bummer: I love your videos and was excited to join a tight community of enthusiasts, but it's way out of my budget.
take part in the public challenges... the first prize is 1 year free membership. second and third are a 3 months membership
te acompanho a um tempo e so agora percebi, voce é br?
No, slovakian and grew up in Austria. Live in Portugal now though if that matters.
The GPT Builder seems broken today. Worked beautiful yesterday.
Good Video
Your "application to the business world" prompt appears to cause Chat GPT to hallucinate rather than give entirely accurate information. At the very least, if the area you put in those brackets is obscure.
I’d be curious to hear what you are using as the variable ahahaha
@@aiadvantage A little specific but "christian indie games" and well it made up three companies that dont exist.
"faithful interactive", "redemption games", and "Beacon Studios"
Interestingly it seemed to tailor these made up companies to my interests that it has in its memory, very specifically.
I question it a little further and it just gave me the one case study that's kind of obvious for the topic, That Dragon, Cancer by Numinous games, and a kids bible app which I wouldnt consider an indie game.
They sound pretty legit to me 😂
@@Yipper64 you really need to redifine hallucinations or where you expecting it to call it Marrio 😀
@@onekycarscanners6002 I was expecting it to give me real companies?
The templating technique does not work for me. The AI did not produce generic placeholders. Strange. Perhaps I have to be more concise. But good hint anyway. Thanks.
just try re-running it.
Thank you for offering tips and prompts to use, but your video is not very accessible for people who have vision problems. If I cannot see the video or read the screen, I have no way to know what prompts you are talking about or what the prompt says. You really should vocalize the words for the prompt, or at a minimum have a text available in the description or a link
We have all the prompts in the description
Love the video title, insinuating that there aren't already dozens of videos about doing all the same things this guy is saying 99% of AI users don't know about. 🤣
What does GPT stand for?
Cool, this will solve the work flow l needed to create. Multiple Templates.
Thx
On the black screen I can see nothing on my mobile phone...
Thanks so much for the help 🙌🏼
"users don't know what they're doing" seems fitting, because these models don't know as well... :)
With swell robotics everywhere, Ai jobloss is the only thing I worry about anymore. Anyone else feel the same?
Gen AI automates "low value" tasks to be able to invest more time in high value tasks, I do not worry about this. Sure, there is an impact in job loss but society has lack of skills and employees in many areas like healthcare, clean technologies, education and more. I think it makes society more productive
My Mom
More Jobs are beign created with A.I as we become more efficient and there expose gaps in our society. We must Solve more space issues which is infinite so there are more Jobs if you really want to work, like mining ⛏️ on the moon 🌚 in a teleportic robot.
great video! thanks!
What about orange Julius Caesar salad? Dang I’m hungry.
youre very presumptive about what people know or dont know :)
Super helpful!!
Julius Caesar was never officially emperor, maybe that why he wasn't there 😅
The GPT Builder is great but incredibly slow.
🔥
Lmao when cat appear and "huh"?
I don't understand the concept of "prompts"
Isn't the whole idea to talk to the AI like a normal person?
If we have to learn specific prompts, it's not real AI.
Julius Caesar was never emperor ^^
first
😁
"reply" is not a new feature
you look so much like Iman Gadzhi
Julius Caesar was not an emperor
If they don’t know how to use gpt4o then why so many active users. Lmao
From, my experience many people show up once a week and summarize something or use it to correct their writing with the free version.
Why do you have to talk so fast? If you think your target i to communicate a certain Message you Are failing!
third
Prompts don't exist, just talk to your new overlord like a person.
Much like talking to a person, the quality of the results you get depends on the quality of the communication that you input. If you don't have a healthy, structured conversation, the results aren't going to be as good, and the same goes for prompting AI.
Second
Forth
0th
Hu? 😂
Fun fact, if you want an image with Giger's style, you can't do that.
But if you ask chatgpt to describe Giger's style as thoroughly as it can WITHOUT mentioning Giger's name, and then copy it and ask it to make a picture with that style, it will work :)
Wow, generic Google responses from an LLM!
Who else finds GTP-4o costly to subscribe?
If you find $20 a month costly, then I’d suggest you find a better means to making more income.
@@eachoneteachone4989 I would definitely appreciate a lead. I need a guide or a mentor
I've been watching Iman Gadzhi for a while now, and it is refreshing to see him being interested in AI. Keep it up bro
another mumbling talker
Keep up the great work on this channel are the winners of the Futur Chris Do The Right Thang 🙏 Poetic AuTexticTikTokJusticeBeeKeepersKeepinITReal Coding and it's far rightful owners daily and monthly Y B A Smukwich and I have taken the appropriate person responsible 😔🔙
Thank you