Chad gbt released their new 01 model about a week ago and in this video I'm going to be giving you five prompts that make it Unstoppable now this model is unlike any other model so these prompts that I'm going to show you and send off to 01 in this video will not be the same in something like GPT 40 or even Claude they go much more indepth there's much more deeper connections tailored responses and solutions to your problems that you bring to chat GPT so in this video I just want to show you five of
the many prompts that I have in order to make this new model respond much better now before we get into this like always the AI Foundation Community is a spot to be if you want to master efficiency and productivity with artificial intelligence I'll leave a link to join in the description or the topend comment we have this awesome Community here where we can share a bunch of different posts with each other we have a network of 337 members who are all trying to leverage AI in their industry sharing insights feedback and so much more the
network is the most powerful part here but we also have a classroom with plenty of different courses on mastering Ai and the calendar where we are starting to do four live calls per week so again if that interests you this will be in the description or the topend comment below now what makes this model different well the responses are much better and much more thought out some of these prompts you might be seeing in this video you might ask yourself why can't I just type that out in gp4 or Claude or perplexity and get the
same answer well that's because you will not get the same answer because what they've done is they've trained it with something called reinforcement learning to perform complex reasoning and not only that but the part that really makes this model Stand Out is it thinks before it answers now the technical thing it's doing in the background is long internal Chain of Thought prompting before actually responding so it's mapping out its response it's checking its constraints ensuring that the solution is tailored and adaptable to your problem now we could go a lot more in depth on that
but that's topic for another video let's get into the fun part where we're learning all of these prompts it's making a one Unstoppable that I've been using members of the community have been using and I found that these are very awesome we're going to have some use cases and some more practical prompts that can be applied to any output that it gives you as well prompt number one is the brainstorming Paradigm that I've developed now if you need to brainstorm anything I would recommend using 01 because it's not going to be giving you any more
canned responses you can type out really long paradigms with deep detailed information and get very tailored responses so let's take a look at this brainstorming prompt that I've developed I'll be leaving all of these prompts in the description for you to use if you're interested but this is a brainstorming prompt that has seven steps essentially it's a paradigm and not only does it have seven steps but I have a loop that it runs through at the end so before the model responds it's going to run through three steps three different times let's take a look
at what this brainstorming prompt entails and this is a huge prompt right and 01 does an amazing job it actually going through all of the steps GPT 4 does not even come close to doing this right so step number one you have to list your problem and you're also giving it a system function or a role now this is just basic prompt engineering skills uh taught in our large language model Mastery class and also we have tons of YouTube videos going over defining roles uh even on this channel so I recommend learning what role prompting
is if you haven't already but basically you'd give it a role to fill fill so it can act and think like that professional next you need to list the problem you have or the solution you want an answer to or the topic that you want brainstorm Med right here and those are really your only inputs and you're just listing your problem here and step number two that we're giving the large language model is to brainstorm Solutions we want three solutions you can pause the screen and read this more in depth but I say when providing
these Solutions think about why you've selected each one unless three components or factors that went into choosing that solution step number three we have a probability evaluation so it's not only going to just give me Solutions but I wanted to think about the solutions it gave me and give it a probability of success score from 1% to 100% that way it can kind of gauge itself on how well it thinks that that solution would work based on the problem that I want solved and this is just an awesome step of the process because right away
we're telling it go back look at those uh give us a probability of success percentage based and give me reasons for it too I'm saying list the pros and cons difficulty to perform challenges outcome expectations and so much more step number four is kind of fun because after it takes into account the probability evaluation I want it to look at those percentage scores that it gave us and I wanted to exclude the bottom two percentage scores and isolate the winner so I only want the best solution step number five is where this Loop begins so
what I want you to do before you even send off this prompt is just copy the problem that you input up here and then paste it right here so we can reiterate the problem I say take a look at the winning solution you found from steps 1 through four I need you to brainstorm two more winning ideas that could potentially have better results than our first winning solution I say when providing these new Solutions provide three components that go into making each solution effective and right now I'm just having it brainstorm the competitive Solutions and
then it's kind of like we're repeating these steps so I'm saying next for step six uh give me another probability evaluation for the two new ones the two new solutions that we just got from step five and then step seven we're going to exclude the losers and isolate the winner there might be a new winner there might be the same winner regardless we're having it go back and try to generate two more winning Solutions and not only that okay we could end it here and have an amazing result but what we're going to do is
tell it to repeat this Loop steps 5 through seven three more times so it's going to keep trying to find two- winning Solutions re-evaluate the probability of success exclude the losers and isolate the winner three different times okay so we're reiterating this thing so many different times and it's doing all of this in the background and it's following my Paradigm that I've developed let's say I'm an e-commerce or I'm trying to velop a unique product that I can at least get some blueprints out for and just have a mockup idea let's say I have an
idea of the product that I want but I want help brainstorming solutions for my product or different ideas for my product I can give it a rle so I can say you are a professional product developer that would be an example of role prompting and highlighted in yellow here is my problem I say I need you to develop a unique product that uses the power of augmented reality in large language models with multimodality so now what do you do well just copy The Prompt let it go to work and have some fun so I'm going
to copy this entire thing and then I'm going to paste it into chat gbt using that new model going to hit contrl C going to make sure that I'm on chat gbt 01 preview going to paste in my prompt and then all I have to do is send it off what you're going to notice is the response is going to take a little bit longer than usual and that's because it's actually charting everything it's thinking of solutions it's evaluating its options it's actually thinking before it speaks with this internal Chain of Thought prompting technique that
it uses with reinforcement learning take a look at everything it's doing before it's actually giving me an answer these are all the steps it's taking in the background before it even gives me an answer with this large prompt I mean just take a look at this this is awesome okay it hasn't even given me a response yet it thought for 36 seconds but that wait time's okay because we're going to get an accurate wellth thought out response so it says understanding the problem as a professional product developer I tasked with creating a unique product that
leverages AR with llms with multimodality and it kind of reiterates what problem it's dealing with it brainstorms the first Solutions just as mentioned in step two gives probability evaluation and then it excludes the losers and isolates the winner that was just steps 1 through four but now we have this three-step 3x Loop that it has to go through so it says beginning brainstorming Loop iteration one of three and then it gives a probability evaluation so you can see solution one is a 75% solution two is a 65% solution 3 is a 70% now I encourage
you to pause the screen and read some of this um as I'm going up I don't have time to read this entire output in the video but what you're going to notice is it's well thought out and much more different than any large language model that's currently on the market so brainstorming is an awesome use case with this new 01 model especially when using my brainstorming Paradigm again I'm going to leave these all in a Google doc in the description so you can just have fun plug and play and use it to your advantage next
prompty use case that I use 01 for is drawing deeper connections between two different subjects when you have loads and loads of information two different subjects things can get messy so let's consolidate that information in all one so I have quite a simple prompt here compared to the last brainstorming Paradigm that we ran through it says based on the research provided tell me how X is so revolutionary when used with X so when you're using this prompt this isn't necessarily a set prompt you have to use there might be some different word play but think
about how you can draw connections between two different things because the parallels the 01 can see is unlike any other large language model so I'm going to be taking a tactical approach to this and uh making it geared towards this video so what I have here is this white paper on uh Chain of Thought prompting that was released January 10th of 2023 so they're finally just implementing chain up thought prompting into these models in order to reason with their responses before actually giving you an output which I think is just awesome more accuracy less hallucination
and more quality outputs you know I could sacrifice speed for that any day of the week but anyway let's get into this we have this white paper here what I can do is I can just copy all the information now when 01 supports PDFs and images that will be absolutely insane but for now we just have to copy and paste certain things into here so I'm just going to copy and paste these 10 pages all the way up to the conclusion into chat gpt1 I'm just going to enter down a couple of times and paste
it in and then in place of my X here I can say tell me how Chain of Thought prompting is so revolutionary when used with or instead of used with I can say used internally in large language models beautiful and so we have Chain of Thought prompting how is it so revolutionary when used internally with large language models that's kind of what this is going over and um you could even upload This research if you wanted to but I think it's kind of cool to let 01 run with the idea of chain up. prompting and
see how it's uh so powerful when used internally en larg language models then I can say without the user physically prompting beautiful I can send it off because again we have the 10 page PDF all pasted in here ready to go when again when A1 can support PDFs that's going to be absolutely insane but let's send it off as you can see this prompt is very very large a lot of different numbers equations information and things to take in as well as my question so it's got to map out my question it's got to do
a lot here and then this one actually thought pretty quick which is actually quite impressive it highlighted the advancements from that PDF information that I posted in there it adhered to my instructions discussed immediate reasoning leveraged chain up thought prompting funny enough enhances the reasoning and also the model comprehension as well so what it gave me was a very nice detailed answer on why it's revolutionary to use this in large language models it says it enables models to internally generate immediate reasoning steps without explicit prompting from the user I think that right there is pretty
important what it says is this mirrors human thought patterns very very cool and also very true allowing the model to tackle complex problems much more effectively beautiful and then it gives me why this approach is groundbreaking and just awesome well-laid out steps I recommend you pause the screen and read this once again so you can see everything and all the information that it provided me so again I recommend doing this to draw deeper connections between two different things o one thinks about the connection it doesn't just spit out an answer you could do it philosophically
or you could do it hypothetically in a situation that isn't actually real yet you could say how might this connect to this or how do these two things connect maybe they don't connect at all 0 is going to think about that and really try to draw a connection if there is any and that's the beautiful thing about this internal Chain of Thought prompting prompting use case number three is to ask a one to explain its reasoning on how it got its answer and to give you a sustainable understanding of why it actually works now you
might be thinking okay Drake I can do this in any other large language model but the problem with that is there's there is a completely different level to this because not only is 01 already thinking before it gives you a response now but if you ask it to explain it's going to think about how it thought about the previous prompt or the previous output that it gave you it's thinking about how it went into that it's thinking about its previous thought process so it's a completely different level and it's thinking of how it's also going
to explain that to you so it goes much deeper than any other large language model it thinks before it answers recommend you give this one a shot if you want a more sustainable understanding into the solution that 01 gives you so we're going to get into that right now let's take a look at an example real quick let me just create a quick content calendar for a social media account that way we can get an understanding of to why 01 gave us that content calendar in that specific format so here's the initial prompt I have
typed out that I'm going to have 01 explain its reasoning to I say create me a Content calendar for the month of October 2024 for my artificial intelligence YouTube channel here are some recent data and stats for my past 10 videos and then I just put in some data here for my past 10 videos uh I got this from my YouTube analytics copy and pasted it in here real quick and this is an example of a good prompt already but this isn't the example right we have a goal here we have some context and we
also give it an action and then it gives me video ideas it thought for 24 seconds beautiful this is our initial prompt we code o we go to 01 for things like this too because it can draw connections between views and multiple different things compared to 40 gives me video One video two video three video 4 and video five but this is where the prompt use case number three comes in to have it explain its reasoning okay so I'm going to paste in this prompt right here that I have this just is very basic it
says explain to me how you came to this conclusion says give me a sustainable understanding of why this works so you're getting the actual result you're getting the answer but why does this work make it think about it again okay make it go through it's thought process that it thought about for 24 seconds I'm going to send this off and this prompt is super powerful with this new 01 model because it's thinking about how it's thinking and it's amazing actually it's absolutely insane in mindboggling how much better this model is with content creation when you
give it some data compared to the other models it says analyzing past video performance it looks at the high performing videos okay it's giving me what it ran through so it looked at these high performing videos says key observations automation is a hit full guides perform well new features and updates those are three key observations that I had based on my content and what works for my channel and so it says identifying successful content themes after it had all these observations it kind of gave the themes of my channel AI automation tutorials in-depth guides new
model or features tips for beginners really exactly what my channel was about with giving it very little information and then it went into how it crafted the content calendar scheduled on Thursdays regular posting same day each week helps build an audience expectation and increases return viewership then it gives me the why doesn't just give me the answer it gives me the why it works beautiful response content creation unlike anything I've ever seen but more specifically the insights you get from asking it to explain its reasoning is unlike any other model the next prompting use case
is for those of you who like developing plans actionable plans in order to accomplish big goals small goals mediumsized goals this new 01 model is amazing at doing something called backc casting thinking from your end goal to where you need to be today I like telling the large language model to think backwards in order to help me achieve a goal because it reframes the mind of the Chain of Thought prompting or the mind of the large language model if you want to call it a mind so again I have a big prompt similar to the
brainstorming prompt that I want to provide for you in order to get some action plans created with this new large language model excited for this one let's take a look now the goal of this prompt is to have the large language model think backwards in order to create a strategic plan for achieving your desired future state so step number one we're defining our desired future State doesn't really matter what it's for do you have a big goal do you have a big run that you want to complete maybe you're doing a triathlon maybe you want
to accomplish a certain Financial goal a certain weight goal whatever your goal is you can use large language models especially this 01 preview in order to uh help give you a strategic plan and in this case we're using backcasting so it thinks backwards instead of front to back it's thinking back to front it's a good way to reframe the shanath thought prompting in order to provide you a very good strategic plan so here you're just going to insert your desired future State and goal and then we're going to do a current state analysis so we're
going to assess the current state of your situation and then you're just going to describe your current situation here based on the goal you want to achieve like where are you currently at if you're trying to lose weight for example what's your current stats what's your height what's your weight what's your body fat percentage what's your BMI what's your visceral fat what are all these things that go into losing weight all of your metrics providing context to a large language model is super important and this is what step number two is for so those are
the only two things you need to do this is what the large language model is going to do it's going to give you Milestone development it's going to identify key milestones for you so you have certain goals goals that you can hit maybe they're 90 days out you can mention how long you want these Milestones out for and then it's going to work backwards to create Pathways I'm telling it your job is to work backwards from future state to present before developing a plan it's going to think okay where does Drake want to be in
his future desired State how are we going to get him there let's work from Back to Front then I sell it to run through this before giving me an answer so it's doing all of this in the background underneath the hood with the new chain up thought prompting maybe I have a member goal for my AI foundation's community that I want to hit in a certain amount of time okay so I'm going to give all the information I say I want my community on the platform school called AI foundations to get 3,000 members paying $49
per month this community teaches people everything about artificial intelligence large language models generative ai ai sound AI video Etc and then I give some stats of what we do three to four live calls per week and then I get into the the granulars what I really want this community to be about I want this community to be more than just AI I don't want people just get their content and leave I want this to be a network of friends who help each other Thrive partner with each other join the live calls interact I want it
to be sought after because the result of this community is making people's lives more efficient with AI this is my current state analysis and then the rest of the prompt is history right now that we have our future desired State our current state we can have it run through this process before giving us an answer it's going to think okay Drake wants 3,000 members in its group beautiful 6 months then it's going to slowly work backwards till it can give me a strategic plan from the front to the finish I'm going to copy this head
over to chat GPT and then I'm going to paste it in I like doing it in a Google doc it's a little bit easier for me to see but yeah other than that we have my current state we have my future State we have the backcasting analysis in here and let's send this off so we can develop a strategic plan with this new 01 preview model I'm going to send this off look at this beautiful long prompt it's going to think through this before it answers as you can see it already says defining the Future
Okay cultivating the community beautiful I love how it does this and how it follows your instructions it actually can run through loops and paradigms in order to get you where you want it's very very tailored doesn't really feel like canned responses that you usually get it thought for 24 seconds before even giving me an answer and then it gave me a vision statement it says Define the desired future state it gives me a nice vision statement here it goes through my current state analysis my Milestone development gives me month-by-month goals that I need to hit
and tasks that I need to complete in each month in order to maintain and keep my goal it even gives me the backwards path that it created in ordered for me to get to my goals which I think is very cool and it goes backwards month 5 to 6 5 to 4 and so on down the list it identified necessary conditions now I've already read through this I think the results are pretty solid and just really tailored to my situation I'm going through and then it gives me a strategic action plan again this was just
like the groundwork up here everything that I just showed you it didn't even give me my action plan yet but then it gives me Immediate actions it gives me short-term action and it really goes through my action plan it says analyze churn Reasons by collecting feedback review onboarding process so that new members feel welcomed and it's very strategic I really like how it broke it down like this in midterm short-term immediate long-term ongoing and the depth of this output that it gave me is astronomically more in depth than what any other model would have given
me so using that backc casting protocol is just amazing in order to create strategic plans with 01 and chat gbt it's kind of revolutionary I haven't really been able to see this good of results with a large language model when it comes to strategic planning with internal operations prompting use case number five is to enhance your writing with this new model now this is going to help because not only are you going to combine prompting use case number three where you go back and have it look at its answer in order to explain to you
why it gave you the answer it did but you're going to tell it to look back and review its previous answer come back to you with a better answer and then you're going to tell it to ask why it changed the things it did this is going to be an amazing way to enhance existing writing or to enhance the output that it just gave you so what we're going to do is we're going to go back to that strategic plan that we just created and we're going to enhance this plan and then we're going to
have it explain to us why it enhanced the writing so a very simple prompt similar to number three I say review your last response and search for areas of improvement tell me everything you've changed the reasoning behind changing what you've changed and rewrite the response I'm going to have it go back and analyze its previous response and look for areas of improvement so I'm just going to send off this very nice easy prompt it's going to think it's going to look back at its previous response evaluating and refining finding errors enhancing Clarity and then it's
considering tweaks to Milestones and adjustments to the action plan so very cool thought for 8 seconds again it's a small prompt but it's re-evaluating and look at this I identified several areas for improvement to provide a more actionable and detailed strategic plan refined Milestones detailed reduction strategies expanded marketing strategies it's not only refining things and changing things but it's expanding on things this is an amazing follow-up prompt to use in order to get the most bang for your buck with A1 so then it just rewrites my plan and again you can pause the screen and
read through some of this if you want it looks like it changes some of the things and puts it in bold so it changed six different things just after telling it to go back and rewrite the response very very powerful and very good to use especially if you have something that you want to actually follow or a strategic plan you're doing or a goal setting initiative whatever it is this prompt is amazing when it comes to enhancing writing having it check over its response and really dive deep into your issue so you can have a
perfect tailored solution using this new 01 model that's all I have for this video If you enjoyed please drop a like And subscribe helps the channel grow helps get these prompts out there so that other people can see the benefits and also if you want to be with us you know you want to develop more than just a little bit of knowledge in AI but real human connections uh where we do Zoom calls a network of people that you can work with bounce ideas off of then I recommend the AI foundation's Community now this is
so much more than just learning about AI we're developing real human Connections in here with all of the members every Friday we do a weekly rewards call and we also do introductions so all the new members we have hop on with all the existing members we introduce ourselves we get to know each other on a more personal level relationships are transcending far above AI so if you really want that and you want to surround yourself with a group of like-minded people who are using AI every single day in the AI news and all that fun
stuff then I highly recommend you join this community uh this is the best community for artificial intelligence it's actually your final stop for AI so that being said that's enough promo I hope you enjoyed this AI video going over the new 01 model and how you can prompt with it again I'll leave these prompts in the description but with that being said I hope you enjoyed and I will see you in the next video
