How I Use ChatGPT-5 to Analyze and Trade Stocks (New & Best Method)

0:00 OpenAI just came out with their latest AI model called GPT5, and it's the best one yet for stock analysis and trading. So, in this video, I'm going to show you how to prompt this new AI model to find winning stocks. Then, I'll cover how to actually use GPT5 to analyze charts so that you can find patterns and get price targets on your stocks, helping you become a more profitable trader. Now, before I dive in, let me quickly explain why GPT5 is going to be a gamecher for both stock research and trading. So yes, it is the most intelligent model that OpenAI has released. But the real magic lies inside the hallucination rates. So there's going to be a significant reduction in madeup numbers, fake analyses, and just blatantly wrong statements. So diving right in, let's first prompt GPT5 with a hard list of guidelines to find either high growth stocks or undervalued stocks. When crafting out the prompt, we want to be as detailed as possible so that we end up with a curated list that follows all of our criteria. and I'll go ahead and copy this entire prompt into the description of the video so you can go ahead and paste it into your own GBT model and then swap out some key terms that you want to filter or search for.

1:00 Now, the first thing we need to determine is if you want to focus on either growth or value stocks. So, this means either stocks with a high potential for continued growth or stocks that are currently discounted in value. But in my case here, I've just gone ahead and said growth technology stocks that fit the following criteria. We want to make sure to use the latest quarterly and annual reports as a main reference. So this way they're going to have access to the latest financials for the companies in question. And we also want to site footnotes in the reasoning. And this way we can go ahead and kind of check the AI's work to make sure that they are referencing the latest figures and they are all accurate as well. And this is great because obviously we don't have to read through hundreds of financial statements and reports to figure out either key financials or quantitative measures. Then in our second step, we can tailor our searches even more with key ratios since we do have access to all of the company's financials. Now, this is obviously dependent on the type of investing strategy that you want to personally use, but if you want, for example, financially sound companies, you would typically want a liquidity ratio above one. And then from there, we can focus on more important metrics like strong profit margins, positive and growing free cash flow over the last 3 years, as well as either a high PE ratio compared to the market average or to competitors.

2:09 And then moving on for our last kind of criteria, we have a qualitative screening. So this is where we can highlight strength of leadership teams if there are any. So the AI model will basically tell us whether the leadership teams are good for the company or not. Then they can summarize general news sentiments. So this can be all recent news articles or news articles perhaps within the last few months to a year to see how the current sentiment of the company is shifting and whether the public kind of agrees with the direction they're going or not. And again, this is more qualitative. So we'll be able to see things like risks, regulatory issues, and even major partnerships with other potential companies. And then obviously you can make the search as wide or narrow as you want. But in my case here, all we want is a list of at least five stocks that fit this profile.

2:50 So basically with AI, you don't have to seek out or research all of this information yourself. You can kind of delegate the due diligence to the specific model of your choice. And by prompting correctly here, you can basically piggyback off the intelligence of the AI model. And then just as a final note here, we'll want to make sure that we're using GPT5 thinking. Now this is the most powerful model that comes with the free or plus plan. You don't want to use GPT fast because it doesn't have as much compute to provide you an answer. You obviously want something that's doing the proper research, analyzing tons of sources in depth and with thinking it's going to take a little bit longer, but you're going to get a higher quality output from the AI model. And here we got our final result list of five companies that are growth technology stocks that have all of the different quantitative and qualitative measures that we were looking for. And again, this is being calculated all through the latest financials. So we didn't have to go to like the SEC site, get our own 10Ks, 10 Q's, analyze them.

3:44 Basically, we're just delegating all the research to the AI model. And here we can see the first recommendation obviously is Nvidia, probably one of the most popular growth tech stocks in the market. Revenue growth. We can see the data center revenue was up 427% year-over-year in the last quarter. And we have super high gross margins, a good liquidity ratio, as well as growing free cash flow year-over-year. And then like I mentioned earlier, the magic here is that everything is footnoted. So if we have any doubts on any of the figures, we can literally click into the source and we'll be able to see where they're getting all their information from. As we can see here, they're just getting it straight from the 10Q. Now, obviously, everyone knows Nvidia has pretty sound financials and great leadership. So, let's just move on and see what other stocks we've chosen. So, next we have Data Dog here. And this again fits our criteria based on the latest financials.

4:35 So revenue growth, margins, liquidity, and free cash flow. And then interestingly enough, we can see they're trading at a premium to other peers inside the space. And it's even factoring in the current sentiment on products and platform offerings that they currently have within their company. Then we have Service Now, which once again has all the quantitative measures here that we're looking for. We have an experienced large cap software leadership with a disciplined operating model. Then here kind of an interesting pick. We have CrowdStrike, which is a cyber security company. And once again, it fits all of our financial ratios and metrics that we were looking for. And we can see that compared to other cyber security companies, it's actually trading at a premium. And then finally, as our last chosen company here, we have Broadcom. Even though it trades below Nvidia, obviously Nvidia is the top dog.

5:21 It's still at a premium to other diversified semis. So this can be a great alternative. And recent coverage highlights in AI pipeline and VMware Cloud Foundation traction. Now again, this is just an example of what I searched for. But obviously, you can craft the prompt to be whatever types of criteria that you're looking for instead. And if you don't know what type of companies you want to focus on, you can simply have the AI essentially answer those for you as well. So if you want them to focus on potentially undervalued companies like Palunteer before COVID, you can add that as a criteria. You don't have to be exact with the ratios as I was either. Again, this is just all tailored to you and what you're looking for, but I find that the more detail you provide, the better the outputs and the results will be. And there are some additional factors that you can add in on top of my prompt. So, things like if you want to look for either short-term or long-term holds.

6:09 And you can even provide your current portfolio, say what type of risks you want to take on, and what you should potentially add or cut from your portfolio based on perhaps like the latest financials, sentiments, and other qualitative measures. the possibilities really are endless in terms of what information and suggestions you can really get out of an AI model. Plus, GPT5 is not only good for this kind of like more research focused tasks. It's also great at analyzing charts. So, if we had a stock that we liked, we can literally paste in an image of the chart and have the AI model provide either chart patterns, buy targets, sell targets, and just any other information we would need to become a better trader.

6:45 So, let me go ahead and show you how you can actually do this. Now, let's go ahead and use chat GBT for more technical analysis where we can analyze charts, detect patterns, as well as entries and exits on our specific positions. And just like the first section, I've gone ahead and pasted in this exact prompt in the description. So, feel free to copy it yourself or just slightly tweak it based on what use case you want to use this for. But what I've written here is that you're just a professional trading assistant and technical analyst. What we want this AI to do is to analyze charts, detect price patterns, and identify possible trade entries, exits, and targets. And in terms of data sources, what we're going to provide them is first a screenshot of the chart, right? Because they'd obviously want to see what the chart looks like. But what we also want to provide is the raw candlestick data. So what I've uploaded is a raw CSV file.

7:30 This includes all the candlestick data. So open, high, low, close, and volume for every single 15-minute candle. And you can simply get this by exporting chart data from Trading View. Now, the reason why we do this is just simply for more information, right? Because if we only provide an image of the chart, what the AI is going to do is basically use OCR to recognize what the image looks like. They're going to try to analyze the image and then provide us back an answer. But this isn't failp proof, right? It's not fully accurate. So this is why we're going to go ahead and just provide raw data to the AI as well. So this will supplement the information from the chart with the actual raw candlestick data to make sure that all the insights are a bit more accurate and a bit more thorough as well. This just ensures that chat GBT has as much information as needed to perform this task. And for the task at hand, what we want them to do is basically read the CSV data and then align it with the chart screenshot. So again, this is basically combining the two data sources. Then we'll want them to detect any chart patterns. So things like head and shoulders, double bottoms, etc. You can also have it more tailored to your own trading strategy if you prefer certain indicators. Then we'll want them to highlight the most likely pattern and explain why it fits. So price structure, symmetry, volume behavior. Then from there we can provide entry and exit levels. So this is where we can find like breakouts, confirmation levels, stop losses or profit targets. Now all the information is great, but we'd obviously want to see it neatly packaged back to us. So finally, what we want to do is output a visualized version of the chart with annotations. So what this means is that they're actually going to provide the chart back to us and kind of mark it up and show what different types of levels they want us to focus on and what their analysis is. So, we're going to be able to see the patterns being drawn, entry, stop, and target levels clearly marked on the chart. And then finally, this is optional, but I find it helpful to have clear trader friendly language as to what's happening, what's the trade idea, and what are the key risks or invalidations. And then to neatly package everything with a bow.

9:24 Basically, we're going to want to format our answer as first the pattern detected, why it fits, entry, stop, target levels, as well as the annotated chart, and then a summary in plain English of what the analysis actually is. Now, again, this is more of a general use case, but if you have specific positions you want to consider or are open, then I'll just tailor this prompt a bit to just make sure that it fits your own use case. So, now this is our super quick analysis on Nvidia's 15-minute chart. We have a clear head and shoulders pattern and then we can see that this is bearish according to the AI analysis. We can see why it fits.

9:56 The price formed three peaks. A higher middle peak around 184 and two lower shoulders near 182. Then it actually gives us entry, stop and target levels if we want to open or close a position. And then finally the summary in plain English basically says that Nvidia just completed a head and shoulders breakdown suggesting weakness. Basically just a super bearish movement. And we can add additional analyses. Again, I would just tailor this with your own trading strategy if you have one. And if you don't, just go ahead and continue with the AI's refinements as well as suggestions to really figure out what you should best do in this specific type of situation. And now, obviously, I'm not a financial adviser. AI is not going to give you the perfect setups. But AI is a super powerful tool and when combined with the right prompts as well as just, you know, context needed, you can really get some great insights as well as just confirm some of your own beliefs or provide different viewpoints on the positions that you're either considering or have open. Now, if you found this video helpful, make sure to like, comment, and subscribe as it greatly helps out the channel. And if you're interested in learning how to actually master AI for investments, so things spanning from stocks, crypto, and even real estate, then make sure to click our website in the description.