Speaker 1: The first tool you should know about is Sourcely. Sourcely is an AI power tool for finding, summarizing, and formatting sources for your academic papers. And the great thing is you can check out its power for free. So click on free and you get sent down to this box here. And you can type in your essay or essay title, but they have given me Sourcely Pro. So let's go check it out here. Let's just give it like the minimum amount of information to see how well this really does. Let's say that I wanna find out about new materials for transparent electrode materials. That is two materials. Anyway, let's go for it. Let's say find sources. So here it says, please wait, this may take a minute. So we're just gonna wait. And that was surprisingly quick. I didn't have to wait very long at all. Then down here, you can see that we've got a load of publications. You can download the publication if it's got a PDF. You can get a summary, which is really great. Getting a summary just means that you can cite this and know what the main points are. But obviously this is no substitute for actually going and reading the paper, you've been warned. So here you can see it gives a nice summary all on the app. Now let's just see what's special about this because there are other things that you can use to find references. But if we go to pricing, this is where you kind of really find out what you're getting. The first thing I like is that it's not very expensive. Only $7 a month. You can see here that you get pasting of your entire essay. So if you've got an essay and you want to find sources to increase its credibility, you can put that in there, brilliant. And you've got assured credibility of sources. Now I'm not quite sure what that means, but I'm pleased that there's some level of quality control going on behind the scenes before they give you the sources. The one thing I love about the sources that it brings up is that they are from very reputable sources. They are things that I would have used in the past. And you can go visit the paper just to make sure you can download it to your device or put it in your reference manager. This is a really great tool that I think every single researcher should know about because finding reliable sources is one of the hardest things about doing any sort of early stage research. Now of course, any AI tool like this needs to be used in collaboration with your own knowledge. You can't just follow these blindly. But the fact that you've got another way to find sources and you can do it largely for free or at a very low cost means that it just makes it so much easier. This in combination with something like elicit.org I think would be incredibly powerful for really making sure that you've got everything you need to know about a particular subject. The second tool that I think you need to know about is MirrorThink. Now, here we are. This says general AI for science. That is a big claim. And so here you can see welcome to MirrorThink and it's got loads of very sort of like sciency-based suggestions which is not something that ChatGPT has. Now this is a very similar layout to ChatGPT so it should be easy for anyone that's coming from that service to this one. You can see you've got literature reviews, facts checking, Wolfram Alpha for mathematical accuracy, technology scouting, science funding and experimental protocols. Let's just click on some of these to see what's happening. Oh, and also down here you've got find papers, find patents and then ask Wolfram. So if I click in on find papers, you can see that it puts this little dialog box and now let's find papers again for transparent electrodes. Oh, we are afraid you're no longer eligible to use MirrorThink for free. Well, that's a shame. All right, let's upgrade. Pay what you want. All right, pay what you want. That's not so bad. Let's say covers costs. All right, let's go with that. So now that I've actually paid what I wanted, you can see that I've got a few more options. Down here I've got web research and literature reviews. So with the extra bit of money I've invested, this is really appealing to the scientists and researchers in me because this is all of the language that I want, all of the information that I would use AI for in one place. So it gives me a lot of confidence. I'm asking it to write a literature review on organic photovoltaic devices and here it's coming up with a plan. So this is more like an agent rather than something like a chat GPT strictly because it seems to be knowing what it wants to come up with. So there's multiple AI agents working together to produce a literature review. The one thing I love about this is it's grabbing loads of different types of papers in different sort of like sub sections of that field. So here it's reading paper one of five. You can see that it's going through it and it's actually reading the paper and producing a little summary, which is great. I wonder what it's going to do at the end. Is it going to combine it all or is it just going to give me these kind of like paper summaries? This is actually quite exciting to me. I've never seen anything like this before. I actually really like it and here it's composing an answer for me. Let's see what it's like. All right, so it does that same thing where it's sort of like just run out of memory I think and so it just stops and I've just said continue and it is continuing. The one thing I love about this is it's giving me long form content. It's giving me references where it's referenced in the paragraph and also it just keeps on giving an awesome amount of information. I think that this is actually one of the best AI tools I've seen specifically made for science and research. They said general AI for science and they're starting to deliver on it, no doubt. Every part of science is covered from literature review to finding fundings to looking for new technology, finding experimental protocols. It just all seems to be there and it seems to be doing a pretty good job at it. So Mirrorthink is certainly something you should be checking out. If you like that tool, remember to stay around because there's tools that really surprise me coming up from companies that haven't even released their main product yet and that gets me excited. The third tool you should know about is Julius. This is an AI powered data analysis tool. Here you can see you've got chat, you've got your own personal AI data analyst and you can upload your files. Now I get asked a lot on this channel about the security of your data once you upload it to servers. Here they say all of these files are for your current session. They will remain available for an hour after your last use after which all the files will be permanently removed from our servers. I hope that's enough to convince people that this is a safe service for them. Use cases for Julius, you can use marketing, healthcare, academia. Yeah, that's what I want to know about. So let's have a look at the full Julius conversation. Go to Julius and check out this conversation to see what can really be done with this tool. So here you can see that they just inspected the file. The data has been successfully loaded. Here's a snapshot. So then you can prompt it by saying stuff like, this data is from the 2022 General Social Survey. We are going to try to analyze which features contribute most to happiness. Now you can do something like this in ChatGPT, Code Interpreter, only if you're paying for it. I've done all this for free at the moment. Go check it out for yourself. The fourth tool that I think you should know about is HeyScience. HeyScience is your personal AI research assistant that can read millions of scientific papers for you. Now here's the thing, is that it's not currently available. I've asked for early access, but I've not got it yet. But one thing that excited me and the one thing I want to report on to you today is this AI reviewer. This AI reviewer is essentially a review from your supervisor or peers without the out of office responses, which we love. So this is a way of getting your paper checked before it even touches your supervisor's desk. I think this could be an absolute game changer for those with supervisors who are less than enthusiastic about reading your work. So let's upload one of my old papers and see what it does. Here you can see it's analyzing my manuscript. So on the next page, you can actually enter your journal where you want it to go. If you need a journal recommendation, they'll also give it to you. And that's great if you're an early stage researcher, you're not sure where different papers go. And so you can see here that it says polymers, solar energy materials and solar cells is actually where it ended up, and nature communications, oh I wish. So it's giving me actually pretty good information. It's got title suggestions, it's got abstracts and keywords. I've got knees revision, oh no. You should have those keywords in your abstract and other important parts of your paper. So it's done a pretty good job at sort of like finding those. It gives me information about why it's not novel and why it's 77% or maybe 77 something similar to another paper. And so once it's all finished, we've got a review of the paper strengths and weaknesses. So this is just like a reviewer would give you after peer review. These are all things that you can at least be aware of that may be brought up by other potential researchers and peer reviewers in the future once you submit it. So being on the front foot with these may be a really fantastic way to ensuring that you can answer those questions when they pop up. And you could even do this for something like just before a talk. So you know where the strengths and weaknesses are of your talk so that then you can sort of like prepare slides to answer those weaknesses directly. A really great tool and something that makes me pretty excited for what Hay Science has in store. And the fifth tool that I want you to know about is Vercell.ai. Essentially, this is a way to compare multiple language models with each other in real time. So you can see here we've got OpenAI, which is there, and then over here we've got MetaLlama. And we can type in the same prompt to each chat. So here I want to say, give me a summary of organic photovoltaic devices. So it goes off, it puts it in two different large language models, and it essentially allows you to compare them directly. Why would you want to do this? Well, it's because different language models give you different strengths with different questions. And it can be a little bit tricky to work out which one you should use. This gives you a perfect comparison of different types of large language models, and you may find one that's better for your research and your use case than ChatGPT. So give it a go and let me know what you think. If you like this video, remember to go check out this one where I talk about the six academic AI tools that you need to know about. So there we have it. There's everything you need to know about the AI tools that you need to know about for research. Let me know in the comments what you would add. And also there are more ways that you can engage with me. First of all, go check out my newsletter at andrewstatham.com.au forward slash newsletter. The link is in the description. And when you sign up, you'll get five emails over about two weeks. Everything from the tools I've used, the podcasts I've been on, how to write the perfect abstract and more. It's exclusive content available for free. So go check it out now. Also go check out academiainsider.com. It's my project where I've got my eBooks, I've got my resource packs, I've got a blog, I've got a forum and everything's over there to make sure that academia works for you. All right then, I'll see you in the next video.