now I hate it when people over promise and under deliver but I genuinely think that this that I'm about to show you is an absolute game changer for research phds and anyone doing higher level specialized knowledge jobs it's chat gpt's latest release it's a beta feature called code interpreter I've been playing with it for a few days and I am just amazed by what it can do let's check it out now the first thing you need to do is sign up to chat GPT plus it's about 20 to 25 I think a month and uh it is in the past been questionable whether or not you really need to use it I think you definitely need to use it if you want to use it for the higher sort of uh understanding higher creativity language models but now I think we're code interpreter it is well worth the money every month essentially you're getting a data scientist sat next to you through all of your research at all times once you're signed up to chat GPT plus the first thing you need to do is head down to your settings and you'll get this pop-up then you go to Beta features and down here you can turn on code interpreter and then you should be able to see up here you get code interpreter which is right here the question is what can you do with it essentially any data scientist thing you can think of code interpreter will give it a good hard go so here I've got some data about the plasmonic responses from UV Vis the one thing I love about this first of all is that it takes the data and it says hey this data set contains all of this stuff and it just sort of like knows immediately what's inside it sometimes it needs to tidy up the data but it just seems to try things until it gets an idea about what's going on so here we can see that it's suggesting for initial analysis we can do certain things we can start generating descriptive statistics and it's just sort of like doing all of the heavy lifting for you and all I've done is provided the raw data that came straight out of the machine and now it's saying hey let's visualize the absorption Spectrum for each sample so it's giving me some descriptive statistics about each sample which is brilliant but now it's working on giving me a plot now this would have taken me hours this would take taken me ages to go through this think about it and as you can see it's now given me absorption Spectrum for each sample and all I have to do is say save image as and then it never actually includes the the file ending so I need to go dot jpeg but then it's saved and I can use that in a report it's brilliant because here we are it's still going here are the correlation coefficients between wavelength and absorption for each sample it's doing the sophisticated analysis all on its own without even prompting remember I've only given it one prompt and it's doing all of this I think this is just incredible the last thing I love is right down here these results suggest that the absorption of Light by the sample generally decreases as the wavelength increases and this might indicate the interaction between the silver nanoids and carbon nanotubes influences the absorption properties of the mixture very interesting so it's giving me not only the results but it's also interpreting the results so that I can then just sort of like start thinking about the next steps putting in raw data to this thing is just incredible and I really feel like it's just changed the game for PhD students you can put in other people's data you can put in your own data sets and then then you let this program this code interpreter just do all of this data science stuff for you I really cannot overstate the effect that this tool would have had on productivity during my phds and postdocs it's just incredible let's add some more data and see what happens I've now uploaded some solar cell efficiencies that I got from using different types of transparent electrodes all I've asked it for is a visual summary of this data so first of all it will go in and have a look to see what kind of data we're working with it opens it up and now it knows that we've got this short circuit current open circuit current fill Factor series resistance shunt resistance and efficiency so now before performing any visual summary they're going to clean the data this is stuff that you do manually as a researcher but now it's just going on in front of you something would have taken me half an hour to an hour is now getting done in like minutes now this is crazy right it's come across a problem and it's asking me okay apologies for the confusion it seems that blah blah blah could not execute then it's now talking to me like it's an actual kind of like supervisor or collaborator it's asking could you clarify and provide more information about the structure of this data which is brilliant it's now asking and prompting me am I other AI or is that the AI oh I know what it's saying I've looked at my data and it's now confused about this sort of like thing here is that there's nothing in here but it's expecting stuff to be in here so I just have to say here that the row each row is a separate device measurement narrowly use that clarification to actually provide with a visual summary of the work and here we are this is what we've ended up with we've got the different devices with the short circuit current the open circuit voltage fill factor and now I've got a visual representation of what happened to each of the elements that I was measuring for each type of device and a transparent electrode that I was testing now this is just incredible I cannot believe they're at a stage where something like this that would have taken me ages can now be done in a matter of minutes incredible if you are on the fence about getting chat GPT plus I really think this code interpreter is the thing that will push most people over that fence to actually purchasing it now I think that anyone in a PhD should start using AI tools like code interpreter for their work not only does it improve your efficiency and productivity but also it can give you actual kind of like uh conclusions from your work that maybe you hadn't thought about you can talk to it like a collaborator it can replace a load of different grunt work that would take you so much time and I think now this has really shown its power for the future of high-level knowledge work incredible stuff from open Ai and it's really sort of like a nod to the future of what will become possible and I don't know if I'm like some sort of Grandad at this point where I'm like oh my God I wish I had this in my day but really it is just something that is just going to be so valuable going forward now if you love this video go check out this one because here I talk about the three AI tools that are going to actually be useful that you'll use every day I know I use them so go check it out there and really Empower chat GPT to work for you in the most efficient way possible so there we have it that's how I think chat GPT will be changing the future for PhD and researchers I cannot overestimate how important this tool will become in the future going forward whether or not it's chat GPT or another tool like this that's released by another company the fact that you've got a data scientist in your pocket with you at all times is just super valuable and in my opinion well worth the money that you spend on chat GPT plus now let me know in the comments what you would add have you tried it let me know what visualizations you've created I'd love to see it but also remember there are more ways that you can engage with me the first way is to sign up to my newsletter head over to Andrew stapton.com 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 podcast I've been on how to write the perfect abstract and more is exclusive content only available for free so go sign up now and also go check out academiainsider.com that's my project where I've got my ebooks the ultimate academic writing toolkit as well as the PHD Survival Guide I've also got my resource pack for applying for grad school there I've got blogs and forums and it's all there to make sure that your PhD and Academia works for you all right then I'll see you in the next video