Speaker 1: Let's check out a trending GBT called ScholarGBT in today's video. Let's learn everything we need to know about this GBT, see if it's any good and see if you would actually want to use this in your daily workflow. Let's jump into today's video. The GBT store has been out for a couple of months now. We're going to go ahead and start testing out some of these GBTs that are, you know, kind of standing out from the crowd. The first one we're going to check out here is going to be ScholarGBT. This is kind of reminds me of ScholarAI, which should probably still be here. Let me see. Yeah, right there. So that was like one of the first ones that came out as a plugin, but now we've got ScholarGBT. So I'm going to click it and begin. In order to leverage GBTs, you need to have a ChadGBT Plus account. So keep that in mind, 20 USD a month. Let's go and try a question here. Let's try a question of find me the latest research about AI and see how it structures its outputs. First major plus I'm noticing here is that the data that it's providing is up to date and is relevant. A past issue we've had when it comes to this kind of tech was it'd be like you're in 2023 and it's referencing stuff in like 2021 because it was only trained up to that point. But it seems like this does have access to the internet and this does have the ability to give us relevant information within like the status quo. Let's go in and see what does want to click. Okay, perfect. So it actually takes us directly to the website. That's solid. Let's go and gut check this a little bit further here. So we got an output of around five here. It's going to look at this example real quick, which is role of artificial intelligence in the field of renewable energy transition and decarbonization. Let's go ahead and check out together, whether this can actually read it and find relevant information and give us a quote within it all at once. If it can, then man, I would have totally used this in college. Honestly, probably too much. Well, let's see if we can even do that. Let's go to reference the article by just hitting right click here. I want to say copy. Okay. In this article, let's make colon paste article name. Give me two quotes that would support the thesis that AI is positive for energy transition. Hit enter. Let's see if we can do it. Y'all. Ah, I can't do it. Oh no. Okay. It seems I encountered an issue for the detail. So I mean, a workaround for this would be basically downloading the PDF here and then uploading it to GPT and see if we can do it to showcase that. Let's go ahead and check out another article is referencing, which has a review paper on a real time defake of voices. Let's go ahead and see if we can get an article here by simply uploading the PDF itself. Went ahead and downloaded that PDF. I'm going to go ahead and add it as an attachment. And this one seems to be more the topic of deep fakes. So I want to say give me two quotes on the negative impacts of deep fakes. This may be a positive article, so it might not exist, but let's see if we can even do that. There we go. By attaching it, we do get a ability to read it, but we're actually going to see if this is true or not. Because sometimes you can be a little confident in its answers and be entirely incorrect. So we got to, we got it from 1456 of to avoid the technology being used in evil objectives. All right. Let's see. Come over to this 1456. Okay. 1456 command F. Ooh. Okay. It's actually correct. No, that was actually good. 1456 and this entire quote was the one, this right here, this sentence was number one. Let's check number two. When creating AI systems, ethical considerations, command C command V. There we go. Now, to be fair, the quotes are basically on top of each other, but you could probably differentiate and choose the one that is more relevant to maybe the paper you're writing. Interesting. So we got one thing. I'll provide a research paper link, analyze it. We already have a research paper. So let's upload a PDF. Maybe we don't even need to upload a PDF here. Maybe we can actually just paste it in like this. This would be interesting. See that is what I was thinking is going to happen where some sites are just restricted and the data you can grab from a link. So we're going to go ahead and provide a PDF, same PDF here. Let's see what it comes up with. Okay. It was able to understand the information and here is a solid output overview, objective, key findings and concepts, voice cloning, ethical concerns, transfer learning. Pretty solid here. I'm assuming that within this GPT, they kind of proctored it in order to give this specific type of output when it comes to headers and bullet points. Now to see if they actually did proctor it to give that kind of output of header bullet points. Let's just ask a regular GPT the same question, same question, PDF, hit enter. Slightly different, slightly different, same thing of headers, more paragraph oriented. But when we internalize data or read information, typically we like bullet points because it's fast and we can kind of skim through it. Overall, I think this GPT is pretty useful, but more in the context of if you're just trying to find research papers, maybe not so much internalized and actually summarize or read the underlying research papers. So this seems like a great tool to look through a bunch of directories to find said research papers. But then maybe once you get the research paper, you would internalize and reformat the data in another form of GPT or in another service. Make sure you leave a like, it's completely free. If you want to learn more about GPTs and how to leverage it in your business or personal life, check out the playlist at the end here. I'll see you in the next video. That's the playlist I was referring to when leveraging chat at GPT, more specifically the GPT store, everything above the board there, even making your own GPTs for very special use cases. That's a random video could be good, could be bad. That's my face. I'll see you in the next video.