you're listening to accelerate a service Now podcast series about achieving your business outcomes where we talk about how to get the most value from the now platform hey everyone this is Crystal Clevenger and this is Dri we're your hosts and excited to talk to you about service now value acceleration solutions that include expert Services training and impact hi everyone and gives me great pleasure to welcome two guests today Rob pastor and mo fagar I'm so happy to have you on the show today Mo and Rob I love it if you could tell me a little bit about yourselves your backgrounds and what you do at service now so let's start with mo hi Crystal it's a pleasure to connect with you as well my name is mofa gear I'm a technical consultant uh at the technical accelerators at scale team here in impact at service now been here for around two years and I work my field of expertise as AI I've done my um back my background is also a bachelor's in Ai and been in the field for around seven years now thank you Rob what about yourself so I am a technical portfolio manager for that same Taz or technical accelerators at scale team uh I have been at service now for over 12 years um but I have joined the Taz team about two and a half years ago and as a technical portfolio manager I am responsible for creating the accelerators um that are that we use in Impact um and I specialize in the AI pod we call it the AI accelerators that has offers wonderful well if it's not obvious by now we are here because we want to talk about Ai and gen AI in general right so um let me ask Mo this question why has there been so much Buzz around Ai and generative AI lately that's a great one and indeed um Ai and generative AI it's been a huge transformation and we've seen in the past um two years there with generative AI now and not the ability to just use AI to predict on based on historical data we now have generative AI that can generate new content images videos and many more um forms thereof data so that has been a significant transformation and what has enabled that is we've seen deep learning and significant Transformations there with generative AI models also like Jan models which are generative advis serial networks that's a type of algorithm that that has also supported um this transformation and also the computational power has been a significant um advancement with computational power there following M's law as well so that is another transformation that has also enabled um this capabilities and we have a lot of data now these large data sets that we can go ahead and consume to feed into these generative Ai and large language models that has also enabled us to go ahead and train and uh generate start generating new content so has been a real Buzz all around and we have seen that as well uh in service now yeah also so yeah speaking of service now right um so what has service now been working on you know related to AI because clearly that's your specialty so tell us a little bit more about that I've been in working on this subject of AI for a long time so for instance uh We've started with parlo um even before that but as an example parlo in 2018 was an nlu company we um service now have acquired since that has enabled service now to provide the the ability to have conversation like um experiences relying on unstructured data and we've also seen a lot of um Acquisitions that we have done in the AI space over the past few years only looking into just in the past year there it ultimate sweet task mining company G2K and internet of things or iot company hit skills intelligence um capability and also in just the past two weeks there atret which is another AI um SAS in the Telco space there been a huge number of um Acquisitions and also focus in the space only Acquisitions but we've also been um working on these capabilities for instance element AI which um has also provided us with AI researchers and those research scientists also worked in some of the significant transformations in the AI and geni space and the the realm there so if we look at just generative AI for instance just in before even the transformation we had with chat gbt 2022 there was a paper and a patent that was filed by a researcher from service now about text to flow how we can go ahead and generate um uh flows there and complete processes just using natural language and text and over the year we have also had significant Partnerships with Nvidia where we have already released um generative AI Solutions on the service now platform from the Vancouver release we've had a number of iterations and releases since we're continuously expanding the Gen space as well with our capabilities here we have other Partnerships with hugging face with um Deo centure ey in the generative AI space that also enable us to to become the one of the leading um gen players in in the enterprise software platform realm so how is service now adding all these AI capabilities that we've acquired to uh to the products that we sell to our customers as I mentioned we've been working with um many players so for example Nvidia Nvidia has enabled um us also and and hugging phase we've released a large language Model Star coder which is one of the world's leading strongest performing large language models in text to code as well as one of the world's most responsibly developed LMS there so essentially looking at only open source um data as well as service now data to train these llms star coder has already been in use for several of the capabilities in the service now platform for example to go ahead and get a summary of a knowledge article for an end user or if I'm an agent and trying to just get a summary of an incident rather to than going through a long history there or task history or even try just trying to generate a resolution note so even in not only this to solve this incident but even in future cases agents can go ahead and rely on those to try incidents all those use cases we have U are these large language models that we have worked with Partners we have these U powering these specific use cases today in the service now platform and many of our customers are are getting real value productivity gains for their end users for their agents as well as saving time you know in these processes as well valuable time so their users can go ahead and focus on on more valuable tasks as well right so we we probably as users of the service now platform even internally right we've probably been using AI already we just didn't necessarily know like we didn't label it as AI to your point about uh summarizing a knowledge article and probably gets done automatically for us but we we're not necessarily seeing it that way we just don't think about the fact that it's AI generated is that correct exactly it's as simple as a click of a button there for for an agent or a user there for an agent if I just get a incident or case there I can just click on a summarize button can already generate understand the context of what is the issue what are some of the actions that were taken in the history of this and summarize that and get get an agent can quickly get a at a glance overview what's of what's going on so really enhances the the speed of an agent right and as you mentioned without having to know that gen is powering all of this an agent can just go ahead and trigger a button likewise with an end user just enter anything in natural language and we can the llm can understand or geni can understand that context and provide a personalized concise summary to that user so without having to understand all the the nuances with generative AI users and agents can these can be infused into their current workflows gen and accordingly they can still leverage that to to realize the benefits that I just mentioned yeah that makes sense so let's talk a little bit about um impact now and how how service now impact can help customers get started with AI and and Rob I'm going to look to you to answer that question absolutely so uh within impact uh we have what's called accelerators right these are fixed scope offering that will give you an expert uh that'll quickly provide value where you need it for and they're they're very specific for what uh per topic um the teams that Mo and I work for we deal with just one uh one of the three types of accelerators and those are the technical accelerators okay so the the difference in technical accelerators is uh what we do is we will take a copy of the customer's instance and create a temporary instance where we will actually be able to demo the work that we're going to show and we'll be able to show you how for example Genai a virtual agent will work in your environment so it's it's very valuable we've got two types of technical accelerators there's most there they most fall into either a jump start or a tuneup so jump starts are for when you're not familiar haven't seen haven't started using a product and we can come in turn it on and show you how it's going to work uh a tuneup would be something for something you're already using and might want help improving it modifying it um you know helping with performance and and guide along the path on what you've already done and where you need to go out of these accelerators U there are probably some that are related to generative AI uh including one that's probably called jump start your generative AI right there sure are yeah we have we have uh 133 accelerators just for uh AI Alone um so we we'll speak about the one for jump starting your generative AI but there's there's one for for jump start your AI Journey which looks at a much higher level and helps you road map um we have for predictive intelligence we have for task intelligence we have jump starts and tuneups for AI search and then there's a whole Suite just for virtual agent um where which will focus in different different areas of virtual agent where you can run jump starts and tuneups on those as well but let's talk geni because that's what everybody everybody wants wants to know about that yes exactly so in jump start your Jer of AI we take that temporary instance we basically turn on uh using service now's large language model the llm were able to demonstrate how it will work from from now assist from the from the the Creator which is allowing the text to code it's going to allow your developers to actually just type in prompts and have ai generate the code it works in flow designer where now you can just sort of describe how your process works and and the AI will will create the flow for you um it's it's actually some some pretty pretty amazing stuff yeah and then we can also show how all the AI cap ilities you know how they work together so one of the examples we might show in the in the in your demo for Jumpstart your gen let's say it we'll start with the demonstration will start with the user in Virtual agent talking you know asking questions and maybe that will use AI search to to return results answers that it's going to pull from your own knowledge articles right that we'll be able to show that this in in the in these examples then we could also show the pth where the end user has an issue and they want to talk to a live agent we transfer to a live agent and now ai is going to summarize that virtual agent conversation the end user had so that the live agent can now just read a single paragraph and not have to scroll through three pages of prompts and answers to know what this user was looking for and now all of a sudden they can quickly read that and and start interacting with the end user and if they create an incident now we can demonstrate uh some of the stuff Mo was talking about the incident summarization right if you have an incident with a lot of back and forth again a new a new fulfiller comes on board they want to know what was done and where we're at scroll through four you know four pages of of back and forth emails and talk or hit the summarize button and now get a quick paragraph telling them you know the important parts of what's been what's been done on the ticket and the questions and answers that have gone back and forth so it's it's really amazing and then we could also resolve the incident or the case and show the resolution notes being generated by by AI looking what the an you know at the answers that were provided in the work notes or the comments and and create resolution notes and resolution codes for us well I hope every company will use that feature because there's nothing more frustrating than you know you call for customer service right and you talk to one person and they can't help you and they pass you on to the next person and you have to repeat yourself and you do this two or three times until you can get to the right person so it sounds like you know with this functionality you know gen can summarize the ticket for for me for the agent and and help everyone save time exactly so I think that's the great point about this uh accelerator and this impact accelerator as well as we can as Rob mentioned with the temperary instance take a clone of the the custom res instance there we enable generative AI so you can see the value of how generative AI is is going to add to you get a preview of that right in in a 90 minute session there that's great um one last question if I may if someone who has service now impact is interested in getting um this particular accelerator what what's a quick thing they can do to activate that accelerator two ways they can reach out to their customer success manager that will be um one direct way they can do that or they can just go into the impact digital experience and from there they will be able to request that accelerator as we noted the jump start your generative AI accelerator great well thank you so much both Rob and mo for your participation really enjoyed our conversation and I hope our listeners did as well thank you very much Crystal been a pleasure awesome thank you for the time I'm Crystal Clevenger and I'm D adri and this is the service now accelerate podcast series you can find us in other service Now podcast in the service now community at community. servicenow.com be sure to follow us in your favorite podcast app so you don't miss our next [Music] episode