this video is going to be a little bit different if you're a regular here you'll know that I do my little intro a bit like this and then I'll skip off into service now and I'll do the techie based stuff but hold on because things are about to get wild that's right I've decided to take a holiday or vacation if you're from the states so I'll be stepping aside for a couple of weeks so who do I call to save the day to fill that Gap to take over my channel while I'm being so selfish oh only the coolest most knowledgeable content creating machine I know that's just in Meadows he's a senior solution consultant all the way from North Carolina he shares my passion of running and spreading knowledge and he's a genuinely awesome dude and he's churning content out like a boss putting everyone else to shame you won't want to miss this so make sure you stick around for the whole episode and if you want to keep in the loot with what this content King has got going on make sure you go and check out his YouTube channel he's got some fantastic content on there make sure you subscribe head over onto LinkedIn get connected as well there and you'll get notified when he uploads new content which he does every single day and now without further ado let's roll the tape hey everyone my name is Justin thank you very much to the service nerd for letting me do this video on this channel nice to meet you all what I want to talk to you about today is artificial intelligence you've probably been seeing this on different channels different blog posts even in the mainstream media people are talking about artificial intelligence GPT chat GPT you might even heard things like DaVinci and QB and stuff like that there's stability AI there's all this stuff going around and I've even been doing it on my channel I did one where we have a the AI creating pictures and poetry I did another one worth generating draft knowledge articles there's another one where I'm fixing the resolution of incidents and customer service cases to be more user friendly more readable and I did another one with the CMA a certified Master architect where he had chat DPT write its own integration with gpt3 and put it on the service portal so the service now service portal so incredible stuff going on and what I want to talk to you about in this video is what is going on what is all this stuff what do all these new words mean and what have I learned by playing around with this now servicener did a great podcast where he went in and explored chat GPT so chat GPT is this one I was showing you here you can see that kind of highlighted in the upper left and essentially what chat GPT is is it's a chat bot so it's a bot that is built off of several different language models that can respond and have a natural conversation with you and if you haven't tried it you should go do it it kind of blows your mind but then there's the other models and that's what you're seeing in the integration videos from myself and from a lot of other servicenow community contributors Jace Benson comes to mind with his scribe monster where it helps you with coding but let's take a look at some of these models in particular I'm looking at open AI That's where I've done my Integrations using servicenows API capabilities and open ai's capabilities I want to start by talking about the overview of these models so we've got gpt3 which is a set of different models codex which is what Jace is using for understanding and generating code and then a Content filter which I haven't played around with at all but it's used to detect whether text may be sensitive or unsafe and so if we scroll down to gpt3 which is what I've used for the majority of my Integrations and the other thing I've used for the majority of my Integrations is text DaVinci 003 now notice there's these other ones here curio one Babbage 001 and ada01 that's obviously a naming Convention as these models are being released and matured so text davinci03 why did I use that in my Integrations well that's what's helping me do things like generate knowledge articles generate resolution so you pass it some text and say do this with it you pass it some uh you ask the question and it can come back with a response kind of like chat GPT but it's not a chat bot it's basically you generate 10 responses or five or just one response stuff like that and you can incorporate that into your workflows and service now and if that wasn't enough there's this whole image generation capability which they call doll E2 so Dolly too is what I use to create those images for typing anything you want and you can get them back in certain sizes 256 or 512 or 10 000 or 1024 pixels and again you're using the API to pass this information and receive it back so you can see there also on the left hand side is some different guides around text completion code completion and image generation there is some fine tuning stuff moderation rate limits this is all API documentation but the text completion is where we're actually doing sorry I'm trying to clear my cursor there text completion is where I'm doing that text DaVinci 003 so here you can see write a tagline for an ice cream shop we serve up Smiles with every scooped uh code completion is the Codex model we looked at earlier where you can have this artificial intelligence Journal code for you and the image generation is what I've done on my particular channel to have some fun and servicenow in the request catalog and creating images so I'll leave you with this there's a lot you can do with this technology chat CPT gpt3 text DaVinci 003 Dolly 2 and I'm just excited about the number of things in a number of ways we can bring this into service now and provide a better experience and start leveraging this technology what I want to leave you with is no matter what you feel about it what you think of it I recommend that you go start playing with this this is the next thing for those of us that are in the technology world I don't think it's going to be a thing that replaces our jobs anytime soon but as you know when new stuff comes out the best thing we can do is go Tinker with it and do that responsibly so you shouldn't be taking any of this code or these things and doing it in production without doing the standard vetting and testing processes that you've come to know and love and of course I saw a recent article uh recently a recent article recently about keeping in mind the Privacy idea so by using these artificial intelligence capabilities keep in mind what you're feeding into them and anything that may be private or you may not want to be out there on these servers in these data models that they're building up based on this particular content yes they're using publicly available stuff to kind of train these models but when you go and make an API call or you go and talk to a chat bot you're also training those models too last thing I know I thought that was last thing but the last last thing I'll leave you with is some of this does cost money as well so they've got to monetize this they've got to make money off of it and you could monetize it too with a really well built servicenow app if you wanted to so that's my Dialogue on artificial intelligence and service now I really appreciate again service nerd for letting me take the time to make a video for the channel I hope you get out there and explore and play with these applications and maybe even something to decides open Ai and until next time and I see you again I want to see you again don't forget to always be learning