so yeah let's get started so hello everybody uh today is uh September 28th uh Digital Services Forum bi-weekly session uh today we're going to have we have a special guest uh today Brian Dunbar uh I'll let him introduce himself when he starts his piece I'm J lascola I'm a senior Enterprise architect with service now uh we're going to talk today about hyper automation service now capability ities in hyper automation uh and we're going to talk about uh RPA and automation engine as well so Brian's here to help us uh talk about those subjects and just a quick uh going forward um we're g just going to welcome new members to The Forum so today if it's your first time just go ahead and go in the chat and put your name company what you do and uh you can be welcome by other members so welcome to the new folks in all right uh so quick Vision about us Digital Services Forum so for those that are new I want to be uh one of the the vision is to be a leader in digital transformation uh enabling our members to drive their own digital Transformations um want to do we're going to share you know many different things here with discussions papers uh we do have a shared instance as well and then we have some videos obviously these get posted to YouTube um we like to have them very interactive and uh feel free to ask any questions uh that you want through the process here and I will be monitoring uh the chat so that we can make sure we get your questions answered you can also go to The Forum and put in the blog put uh your your questions in there and John myself or one of the other coers will uh uh will answer those questions offline as all right uh so today's agenda will be uh the automation with RPA workflow uh I think we'll talk a lot about hyper automation I think there's a lot of uh discussion around that and how it relates to things like uh Advanced AI and how we're doing you know intelligent automation across the platform so Brian uh with no further Ado here go ahead and uh introduce yourself let's see let's go forward with what you have here thank you awesome thanks Jason uh hello everybody super excited to be here to present uh today around automation integration RPA capabilities um I lead our uh solution Consulting Specialists as it pertains to integration and automation globally here at service now uh I've been with service now for eight years prior to that I was also a customer so I've been around the ecosystem for about 11 so I know where where some of the the bodies are buried if you will uh in the platform and I really want to share with you all today uh kind of an overview of these integration capabilities that can help customers like yourselves go on a a hyper automation journey to really leverage the full uh toolkit of capabilities to automate and digitize processes in the service now platform without having to string together a bunch of Point solutions to meet your end business goals um I'm going to talk through some kind of uh perspectives I'll share some use cases I'll share some customer references and then also I thought I'd make it real by kind of doing a demo and showing you what does a hyper automated workflow look like leveraging some of the tools that we're going to be talking about so so some of you may be asking like what's this hyper automation term that you're dropping and um at the end of the day hyper automation is a term made up by an analyst I Gartner um but how we want to be thinking about it and why we're leveraging and and really latching on to it from a strategy perspective within service now is about hyper automation is about linking Systems and Technology together transform your organization to achieve better experiences so you know really service now being a platform of platforms being sort of that Digital Services hub for your organization hyper automation plays into it very nicely and I think about hyper automation as a business imperative so much like digital transformation has been a focus for organizations over the last call it decade maybe decade plus hyper automation is kind of writing on the coattails of that digital transformation and we're taking all of the good work that we've done to digitize and and process but it's to really automate as much as possible uh in driving uh those outcomes now what are some of the challenges that we have within hyper automation is companies operate based on departments and departments like to put Moes around themselves right and so it's breaking down these silos in order to help drive endtoend workflows one of the other challenges that we see with hper automation is there is a lot of enabling Technologies so AI machine learning low code tools chat Bots RPA integration platforms process mining process automation tools orchestration tools and add in you know the the the topic dour of generative AI which you know is a a branch of AI uh within here but be remiss if we didn't call it out specifically and organizations have been really working to automate work but what we observe and as I talk to various different teams is they've taken a very tool Centric approach which is creating islands of automation organizations may have went out and bought an RPA solution and every automation opportunity is solved by a bot and next thing you know you got a hundred Bots tucked in all over your organization you don't know if it's continuing to generate value uh or you brought in in a new low code application development platform and then teams went off and built a bunch of low code applications or you know the the the OG of of uh low code applications like Lotus Notes or even SharePoint U and you have proliferation and so hyper automation is really bringing together all these capabilities putting it under a banner much like we did with digital transformation to really focus on things uh you know more holistically and you know these problems of multi departments multiple tools have led to vendor sprawl I got 20 different tools that I can use automations are breaking we don't have good insight to when they happen and then you know as I talk to Executives like and I asked you know what has been your Roi and your your RPA and intelligent automation investment and some are unable to really answer that uh more holistically and so we're going to talk about capabilities and solutions within service now that are going to help address this and another kind of statement of gardener of what we're seeing and why again we're investing in these capabilities and hoping customers like yourself can take advantage of them to drive forward this is organizations are transitioning from a a Loosely set of automation Technologies to a more connected automation strategy right automation platforms and our first step in how we guide organizations in this hyper automation journey is connecting that Enterprise through automation engine capability and you might be asking yourself what is automation engine so automation engine consists of these six capabilities that I'll be walking through um in a little more detail and then also demonstrating them and the first one is RPA Hub so RPA Hub is a in platform RPA solution service now acquired a comp called intellibot about two years ago and as we do with all of our Acquisitions we rep platformed and made this a available uh now four releases ago uh in platform and continue to make investments and really this supports attended and unattended automations but it's also native to the platform we ship with our you know connectors and templates and solutions and it's really allows for you to be able to automate workflows where apis may not exist where we need to interact with a UI we may need to interact with a third party that doesn't publish an API or won't let us or we may not have the the right security rights to do so but we can interact with the UI and you know I while I position and help sell you know RPA within our customers um one of you know RPA remains uh an automation of Last Resort in my opinion right it is fragile it can break especially when you're doing screen scraping UI interactions those sorts of things but it does have a time in a place and it could be a good Band-Aid solution it could be a good quick fix to really help you automate a process end to end while you may seek further investment to fully digitize a process for example uh integration Hub is our low code integration framework or kind of think of it as a integration platform as a service on top of the service now platform and it ships with 180 spokes these are pre-built connectors to Common Enterprise systems along with a custom action Builder where you can build your own actions to maybe your own uh in in-house uh integration so again where an API exists we should be using integration Hub to connect your workflows to those third-party systems and then uh document intelligence so this is a relatively newer capability uh it's been around three releases now and this is an intelligent document processing solution which enables you to connect images and PDFs and other files to extract data off of them intelligently using AI to drive workflows forward so as you've just seen with these three capabilities if we don't have an API and we need to interact with the UI we can throw a bot at it if it has an API and it's robust we can use integration Hub if we're interacting with PDFs and documents we can use document intelligence to be able to do that our next uh core integration capability is stream connect and so this is a capability that has been in platform for two releases now and this allows for you to connect your service now instance to other uh Enterprise systems via Kafka so Apache Kafka is an open source project which is really uh an evolution in and and helps Drive event-driven architectures uh so if you've been around for a while there's you know older middleware technology uh like uh tipco or or mq Services right that were were there to kind of get data between various different systems cofco is kind of that next Evolution that we you know almost every organization Under the Sun is using it somewhere uh stream connect allows for you to connect your instance into that ecosystem to drive event-driven uh architectures and workflows uh leveraging the service now platform and most recently added to the automation engine portfolio is process mining for external data so service now has had process mining for in platform so you could look at your incident process and run process mining against it uh your HR case management processes those sorts of things that's been in platform for quite a long time we've recently added um the ability to extract and process M external data so we can ingest uh audit logs and log data from an external system uh maybe like a a you know a procurement system or an Erp system and model those processes so that they could be rebuilt in service now either using our packaged applications or uh app engine to build out bespoke applications and um provide provide that insight into that mining operation and then last but not least is automation Center which I'll be going into a little more detail which is our vendor agnostic approach to bring together all of your automation Solutions so this is where hyper automation comes together on the service now platform we can we have task management for managing our new automation initiatives right so a request form that comes in hey I want to automate this process in finance we can discover prioritize it then we can go build it and we could build it using service now tooling as I just showed like an RPA bot or integration Hub solution or if you've already made investments in other thirdparty automation Solutions you can go build it there but then we bring automations into the service now uh cmdb and we track them and look at their operational health so that we can manage that overall ecosystem and then lastly we track goals so every time you know if I if I wrote a thirdparty bot using automation anywhere every time that bot runs it sends an execution log to service now we take credit for the the the automation savings for that and we bring it all together so this is a multivendor holistic solution sitting on top of itsm is super powerful um and I'll show that through uh the demo as we get into that but um I'll pause there I went through kind of the the six main capabilities in automation engine and see if there are any questions before we kind of wrap up on on some other uh key capabilities yeah I have a question question yes yeah Lou from San Diego so uip paath is clearly the leader in the RPA business I don't even see service now on the chart plus integration Hub is definitely a separate subscription and you it basically if you're not going to get professional or Enterprise game over I I'm summarizing I mean can you use it RPA without integration Hub that's just a recommendation and why would we benefit from using RPA as part of uh what what is it um well why would be benefit using RPA Hub versus uip path if you already have uip path and you're using it and can they somehow be synergies between the two three two of those yeah good question and absolutely correct right so uipath is a market leader in RPA um and so uh I I am not going to Advocate anybody to go throw away their UI path Bots obviously you've made the investment to go automate those tasks within your organization um and the uh from a from a service now perspective um one of the things that we're aiming to do and I I won't get it you know you're correct we the way I think of it is from a licensing perspective we have two two tiers right integration only you buy integration Hub or or you buy an automation engine skew which is the super skew that contains all of this stuff right and so if you're all in on looking at from service now from a toolkit perspective you just go to automation engine Enterprise um and and as opposed to just buying the the integration portion um and and fully noted that you know you may may have a use case where you want to just buy RPA um and I can I can provide that feedback to to product management but um within in integration Hub we do have uh uipath spoke we have an automation anywhere spoke we have a blue prison spoke fully acknowledging that those are Market leaders in that particular Point solution area um so you can invoke a uipath bot from a service catalog request right so I got a service catalog request I have a bot that goes off and helps in maybe procuring laptops uh that could be triggered from integration Hub um one thing that I've observed uh and we did a webinar on this a little while back is we asked uh uipath customers how many of them are logging their UI path Bots into their service now cmdb and about 20% were and one of the challenges that I hear from Executives is um when automations break it can be very difficult to react appropriately to them um and so bringing everything into the service anal cmdb and also uh doing Incident Management and change management so if I have a UI path bot that is interacting with a UI say an sap interface and I am going to be upgrading sap and I log a change record um if I don't have that in the cmdb and I don't have that correlation uh within the cmdb to tie that bot to that business application I don't have line of sight to understand what my my change impact analysis might be and as I talk to application owners um there are a lot of application owners that have Bots running against their application that they don't even know about because there's an automation team off doing a bunch of stuff and so we're really trying to solve that problem I I'm not going to get into a knife fight of when I whether I should use a a service now bot versus a uip path bot but um service now RPA Bots or at least our our licensing is um as we've done Benchmark analysis is about uh for anywhere from 50 to 60% of the cost of a UI path bot and I also see a convergence um in the the market occurring where organizations are moving towards their core platform so it is um they're building Bots using power automate around their Microsoft ecosystem they're using you know service now Bots around their service now ecosystem they're using mu soft Bots around their service you know Salesforce ecosystem and um instead of kind of having uip path be be kind of in the middle there and and that's just a trend that I'm seeing um you know it doesn't mean that anything's going away but um if you look at where UI pass path is going um you know it's trying to get into workflow automation low code application development it's trying to shore up areas where service now already has strengths as well so that's just my you know highly biased opinion on it all I don't know if there's followup questions or or or feedback on that front or if that answered your question feedback follow the diagram is great it tells the story of why you go at service now other other options because you're holistically dealing with it in one platform and it's handing everything and all the synergies with the cmdp ETC so not that my question was against doing it I just wanted to compare and contrast and you did a great job and I appreciate the answer thanks yeah yeah you bet Y and and that's where I don't like I said I'm not we're you service now was a uip path shop and we actually refactor we we ripped out all of our UI path Bots and built them back with service now RPA once we got the same product so but that's that's not a probably a prudent step for most organizations but as you're thinking about building new digital processes with automation like if you can do it all in one toolkit is that it reduces handoff to various different tool teams it can lead to efficiencies and in the overall build but you know that's going to be you know Eye of the Beholder if you will right so cool thanks for the question hey Brian there's a couple other questions real quick in the chat one was from Allen uh what's the integration engine towards other vendors going outward and what is the allowed integration coming in I I it's kind of a general question maybe Allan if you're on you can unmute and clarify but so we can address that question yeah I can I can take a stab at it so you know integration hub um is um it is our solution for um integrating with vendors outward right so um we we ship with 180 spokes um to various different systems so for example I'll just use a very it example um within Incident Management I may have an issue with u a piece of software that our developers wrote and our developers use jira um and I want to basically escalate that issue to the jira team and they they do their issue management through jira so we have a Spoke um that drives a workflow to be able to create um a an issue record in jira related to the incident in service now that's all pre-built kind of TurnKey and we're we're continuing to make that investment such that if you were to have to kind of build that on your own all the all the tools are there in the platform to do it but you know if I were to go to one of our partners and ask how what's an S so to go build that you're probably talking you know a couple hundred hours if you think about build test you know deploy to prod all that kind of stuff whereas we could literally build that in about we could get that to production in an afternoon um using integration Hub so that's really our solution and this also is handles inbound as well so let's say the opposite when that issue gets updated by that developer and we want to update the incident record with those comments it's all pre-built and ready to go so some of those systems like Jiro that's a modern system we have a web hook integration where we can do kind of real time updates on both sides through flow designer workflow driven um but then there's other systems like uh an Erp like an sap Hana um we can do OK and RFC spoke Integrations that we have pre-built on that now sap isn't as fun to work or easy to work with as like a Jura and a service now platform being both SAS based applications um and so there's there's uh some a little bit more there but that's you know we have a robust set of capabilities whether it's you know rest web services soap um if we need to execute Powershell commands or if we need to do SSH steps like all of those those uh utilities are available in the platform to connect to anything and then again if we can't do it through Modern web services means we got RPA if we need to so hopefully that helps yeah Alan thanks you said your mic wasn't working you can put if that answers the question in the chat thank you let the jurus spoke the jurus spoke is just called the jurus spoke um so there there are a few different ones um out there for different use cases so I brought up kind of issue management uh there are also use cases around uh kind of project management so we have our own Agile development capability within service now so if I wanted to get a a development step to to a team using jira uh we have we have that ability as well um so yeah U we can provide documentation on those 180 spokes um it is uh yes I think Kevin said it's like a Lego pieces so I I like to think of it as we we have a bunch of ingredients and a bunch of recipes for for integrating with other systems and you're always going to have you you might like a little more salt on your recipe you're empowered it's not a black box right you can uh you can add on uh but it really just helps accelerate time to Value to connecting to other systems yeah there was another question Brian real quick around the um any kind of documentation around actual implementation of of the solutions uh maybe a holistic view looks like a step-by-step guidance on the implementation uh hyper automation Solutions that's that question was from Ankit yeah I think we can provide that after the fact so um if you aren't familiar there's a a now cre cre website uh which has all of our implementation documentation so you know as we went through all these capabilities like um you know implementing an RPA bot is going to be different than uh an API based integration or document intelligence solution we have documentation on each one of these capabilities um and how you uh a proven process that service now has to to go forward and Implement them um I am going to show a demo where we kind of stitch together of a bunch of these capabilities to really automate out one workflow and we can kind of talk in a little detail about what that might look like but um yeah okay great yeah we'll keep moving forward and then yeah as questions come up he feel free we'll we'll uh address them in the chat try to cap them in the middle so automation engine fits into a more holistic uh appr approach and and this is really service now trying to provide a proven approach and I know you guys are all practitioners and want to learn like how we think about this and so I affectionately call this our hyper automation wheel and really what this shows is I talked about automation engine which allows you to kind of connect to anything and automate anything but then we need to think about how do we make workflows intelligent and how do we make sure that they're future proof and proven and continuing to generate value and so if we start off at the top this is our approach to hyper automation is first you digitize the process you can do that through you know no code workflow right so app engine build your and digitize your your process you could buy a packaged application like itsm or HR Service delivery for those sorts of HR processes it processes customer support processes and so once we have it digitized then we need to start to automate tasks and connect to other systems so that's where automation engine comes into play RPA API based integration document intelligence and chat Bots and assistance for interacting with those end users and then once we have our processes digitized we've automate we start automating tasks we can start to enhance decision- making through Ai and machine learning now why this is at this step is we need to have data to be able to do that right so if we're thinking about uh triage of customer support cases using a capability called task intelligence within service now in analytics to be able to help use intelligence and providing insight and automating steps using Ai and then lastly is automating once we you know kind of have digitized automated and and embedded intelligence it's about making sure that that process continues to deliver value and so you can run process Mining and task Discovery within the service now platform so you can identify new bottlenecks new automation opportunities to make sure that that digital process continues to be healthy you know processes are going to change like regulations around those processes are going to change and I could go pay a consultant to come in and we could do a two Monon whiteboarding Workshop about how we're going to make this process better or I could just run process mining against my current state of my process and get insights into where my bottl X are which are more empirical than you know the old whiteboard activities that we used to do in the past and then automation Center really helps bring this holistically together along with continual service improvement and you'll notice in the center generative AI is across all of this is service you know whether it's digitizing the process using text to code or you know we're we're looking at uh things like autog GPT right where um I don't have to necessarily know to tell an automation to run I'm just going to ask it to do something for me and it's going to say hey I found these five automations that might work so that is going to be applied through throughout and we continually embed AI not just generative AI into how we help you deliver on this holistic uh platform approach so a little bit about use cases successes that we have um and these are all going to vary right they're not going to be using every tool in the toolbox for everything um but really like USDA um is one really great success for us um they have streamlined their process for people to apply for uh various different permits by leveraging automation um and RPA and document intelligence to ingest those documents extract data for those workflows and then running RPA Bots to interact they got a lot of Legacy systems as you can imagine in the federal government um RPA is a huge tool for them to be able to bring all that together and they find huge value by just having one platform not a handoff but between a bunch of different tools to to to complete that that end goal um you know one uh software company saved a million and a half just in automating access management so getting people access to systems and yeah we we should like some organizations may have that solved but doing access management to like the Erp can be challenging at times like sap doesn't have great apis to provision new user access and so they built a a RPA bot that goes out and Provisions that access does password resets you know little things like like that so just thinking about how you can help drive those outcomes for your your your customers as well now I want to make this real for you all um and how do we how do we get into so I'm going to do a quick little demo and then we'll have some time for some questions and I I wanted to do a a payroll processing service right so just pick on a a maybe a easy little service and today if I wanted to update or set up direct deposit you know I might have to complete a request and then that gets handed off to the payroll analysts and they have to go fill in some forms in our payroll system and then we need to validate it make sure we got all the the the banking information correctly and it it's it's a lot of handoffs and so how we think about automating this process and this is the demo that I'm going to show is we're going to show that end user interacting with a chatbot to get the original set up then we're going to use dock inel to extract data off of this the form that gets submitted we'll use integration Hub to interact with a third party to validate that banking information is correct and then lastly an RPA bot to go interact with our Legacy payroll system to update that payroll information a zero touch process but we can also use Ai and machine learning for exception so if we need to hand it off to a human right we had an error in the dock inel extraction or if uh somebody submitted this uh direct deposit request from a state that we don't have within our regulations yet we can hand that off right so so this is thinking about how we're going to use some of these Technologies to bring it all together and then lastly once we build this we have analytics automation Center and process mining to make sure that we keep this process healthy uh throughout the duration so with that I will hand off uh start up a little uh demo so let me just hit Refresh on my old browser since I haven't been in here for a minute um and I am logged in as Amelia um so she's an end user that is looking to update her direct deposit setup and so she's gonna engage with the virtual agent and I need to start a new session all right and so for her she's just going to type in direct deep is it and the virtual agent um is going to be intelligent enough to look at all of the catalog items that we have out here and it says hey what what would you like to set it up for and we're going to set it up with our for our checking account and then it's asking for me to upload an image of a voided check and so I have a voided check here and I'm going to open up and attach to the file and our virtual agent has submitted our requests on our behalf it gave me my task number and is there anything else that I need nope I'm good to go right so we keep this uh super simple for our end user uh they don't have a lot of complexity they don't have to fill out a lot of forms they didn't have to go search through a knowledge based article to go do this they just interacted with the chatbot to do it so that's you know kind of one uh approach that we have so let's see how we've automated this on the back end so I'm going to impersonate Jonathan Hendrick and again this is zero touch so technically Jonathan doesn't need to be back here but we're we're through uh we're going to show how we have this all built out so um we have a workspace set up for setting up direct deposit you can see this task can in it's sitting in an open State when I click into this task it's going to open up the uh the workspace for this particular uh task and we have a Playbook so the Playbook is going to be the first is going to extract the data we're then going to validate the the banking information and then we're going to update the payroll system again nobody no human would need to be involved with this I'm you know it'd be a pretty lame demo if I'm just like uh everything just magically happen so we're kind of of showing you what's happening in the background here um within this so the first step that we have is we need to extract the data so if we look at this banking record you'll see the account holder the account number Bank name are blank but we have the check that we got from Amelia and what we're you're seeing here in the workspace is document intelligence and what I'm showing is call kind of human in the loop and so we can see here that the um that the uh the uh uh AI has done the extraction of the particular data elements of the account number the account holder um and it is going to give us suggestions and so you can see the AI has a 96% confidence that this is the account number and just again for demonstration purposes I'm going to hit submit so this would be called human in the loop processing where human is actually validating uh the extraction uh with that high of a confidence at 96% we would do what's called straight through processing just let the AI bring all that together so you'll see our bank record now has the data extracted from that check and so we can mark this uh step complete and mark the extraction complete and we're going to move on to our next step so that was using document intelligence took the check information we extracted it intelligently and now we're going to validate the data so one of the things that you'll see here is the routing number validation was successful so we had a workflow running on the back end that took that extracted data and one thing we wanted to do is just make sure that we extracted the routing number correctly and so there's a an available open API it's called routing numbers.in where I can just send a bank routing number and it's going to validate that the bank is set up correctly so what you're going to see here is uh the operational view of that flow designer flow which is using integration Hub so this is our API based integration um where we can interact with thirdparty systems in this case it's a a thirdparty RPA uh third party API and we can uh we set up a validate bank so you can see it came back with JP Morgan Chase um so this gives us all of the data it's all workflow driven uh triggered by this new bank request or updated to that that current record we also can look at um the report view so again this is the instantiation of a particular run of that um but this allows for you to be able to apply workflow to any process that you have whether it's an API integration or a dock inel solution or an RPA bot so let's come back to our task we've marked this complete again I wouldn't have to be clicking these buttons this is all going to be automatically happening and the last thing we need to do is update the payroll system and this is done through an RPA bot now um because this runs on a payroll system I don't have access to I'm just GNA kind of show a video of the the bot running in the background and so with that here is our RPA bot it opens up the payday application it takes that data extracted inputs it into that system and our RPA bot ran successfully right so this is just regular old RPA information um nothing too magical there uh but this is you know triggered an RPA bot from that flow now that we had that routing number confirmed we can safely feel like we could go update that that system of record which is our payroll system with that bot and so that uh allows for our overall process to be completed now Amelia is going to get an email or the chat back it says hey Amelia you're all good to go the data you submitted is updated in the payroll system human didn't have to touch anything we showed dock inel we showed integration Hub we showed RPA Hub but the last thing for Jonathan Hendrick is to look at automation Center and so automation Center which I mentioned is kind of bringing together all of these automations um into one place where we can look at their overview the execution health and the value that it's providing in addition we can do things like automation uh requests so uh the ability to do kind of a mini project management around building out new automations and we can look at um them tied into business applications future automations we also use Spotlight uh capability to prioritize for our customers as new automations come in and with geni we're actually going to be building a recommendation engine to say this sort of automation would be best met with RPA integration Hub uh those sorts of things to make it super simple uh for you to automate more you'll see here on the executions tab this is where being on an itsm platform is huge I can look at my automation incidents I can look at changes that may impact automations that I've built whether it be a bot or whether it be uh an integration with an ipas a flow designer flow um all of that and we can look at operational health and get insights into them we can slice and dice this by you know department so if I'm automating stuff for HR I can have a conversation about that I can look at various different automations or various different sources right so uip path or blue prism you know again we're not trying to to conquer the whole world but what we are trying to do is provide a holistic view of your overall Auto automation landscape and then lastly it's about expressing value uh back to the organization so for every automation that's run how much are we saving what are by automation sources what are the cost savings what you know what should I be investing in more um and those sorts of things really bringing everything together into one place so that I can manage it holistically for my overall hyper automation Journey so hopefully uh you uh found that demo a little insightful uh kind of hopefully made this a little more real for you all and uh that is all the content I had and I'm happy to kind of go through I see some questions coming into the chat and we can go through those yeah hey thanks Brian appreciate it that was awesome uh yeah there's a couple questions in there I don't know if you you're able to see I I've been addressing some of them but I think that there's one from Sarah here how does data validation occur yeah yeah yeah go ahead yeah so the data validation on Doc inel is there's a setting so as you saw when I did that drop down it said it had a 96% confidence that it found the account number correctly there's a setting within your dock inel solution so they're called use cases actually so every use case so that we would have had a a a check extraction use case you can set a threshold either at 80% um or you know 95% depending on your process and if it's below that it will do an exception and send it to a hum to a human in the loop right so there's a it's called a straight through processing threshold if it's above that threshold it'll do stra straight through processing if the confidence of the AI is below that it will send it to a human to validate and train further um the Havey found the cyber security group reacting to put unended B activity into service now um we have not observed any I I would say additional scrutiny from uh cyber teams in my experience um most organizations have RPA Solutions deployed um the methods in which service now RPA works is the same as how mid server technology works so most organizations that are leveraging the now platform that need to connect to on premise systems have deployed mid servers we use the same architecture for putting unattended Bots that may be communicating back to the platform and so that that conversation like that that sort of uh security conversation uh has been pretty easy in my opinion just because of it's already a deployed known approach uh to interacting with the the the platform Let's see we got a couple other ones I realized that I wonder how the practice of marking the oh I think you're asking like on the check if I had a huge void stamp over it or there's some other opusc of that data like that you know the extraction engine's probably going to be pretty challeng and that you're you're correct um I think the you know OCR technology and intelligent OCR technology is coming a long way if you saw the latest on chat GPT you can now do image recognition through chat GPT and things like that um we are still there's still challenges around handwriting uh there are still challenges around having waterm marks over things um and other sorts of of of challenges so I think there is you want to have prudent use cases I think if you think about the sort of workflows that we're driving in service now um you know it it's um you just need to be mindful of those potential limitation y uh can we utilize a threshold to run an automated reject validation so uh you can so so the question is around extracting the data and then uh validating it after so what you can do is the VA the data would be validated and then what that does is it puts it into a table called extracted values and then those values are available in flow designer and so then in flow designer you would write your Rex to do your data validation and then if um your your Rex checks don't come back positive you would then in your flow trigger somebody to go into the dock inel solution to hit submit so yes you could do that it would be kind of the what I didn't show about dock Andel is it's all flow designer driven so when that task uh hit that attachment that check attached to that task there's actually a flow designer flow that is sending that to Doc inel and then once you have those extracted values you trigger the next flow to do whatever you might want it to do in that case I just had it putting the data in those data fields you can do post processing of those as well in that uh one thing I've seen is like date formats between Europe and the United States are different and so you may want to extract the dates and so they're all consistent with either European format or the US format and so you can do that postprocessing through F designer as well uh this is a straight API con connect to Payday um the payday payday application is a completely fictitious application that I made up of of a payroll application so um we what I wanted to show is RPA so that was there was no API involved in that we were actually just doing screen scra you know so we were doing UI interaction so we had the bot open up the application and actually type data into the form mimicking what a payroll analyst would do um you if if our fictitious payroll application had an API I'd probably do that rather than do the RPA bot it's it was mainly for demonstration P purposes um how RPA Bots interact with service now assuming other um so I think the question there is around automation Center um so automation Center is really meant to be uh a a oversight and governance panel for an automation Coe it is not meant to be an operational interface for uh triggering Bots with third party it it can do that I'm just saying its best use case is um you're going to use your centralized tools for automation anywhere blue prism to do all your Bots what they're going to do is they're going to report how they're how they're doing right and it's going to tie back into service now so I kind of have this 360 degree operational awareness rather than having to log into the blue prism console to see this and then the you know my my pegga bots over here or whatever it might be so it's it's really bringing them into the cmdb every time they run it's going to give a success or failure status that may trigger an incident um and and also can provide some insights like if they ran out of memory and then we're taking credit for every time they run of the value so we can Express that back to the business or other leadership um so from uh the the question on pii um and validating with AI tools um this depends so our document intelligence does take that that image and it puts it into a centralized AI infrastructure within service now so if you're using our predictive intelligence capability or you know we we're launching our generative AI stuff today actually um and uh all all this does go to a shared infrastructure we run it through the the AI training we obviously then delete that and we provide the results back into the customer instance um and so I I can't answer your your own company's perspective on pii um service now what I observed is a lot of organizations that have strategic relationships with service now trust in our security controls and our capabilities where they are doing workloads that have pii and Phi within them um but I you know that's always an eye of the beholder we certainly have customers that are processing pii Phi via some of the AI tools within service now it's just it's it's always going to kind of vary uh based on organization so I can't it it can do it there are also masking techniques if you don't want to do it um and and other sorts of things um but it that one's kind of a uh it depends answer unfortunately let's see is that the last question I think you I think so all right good let me uh screen share again and and we'll just end the uh conversation keep asking questions as we go we have a few more minutes so I want to make sure uh if you have any questions we'll go ahead and um let me share my screen [Music] again [Music] see okay so yeah go ahead and keep monitoring the chat but yeah just wanted to uh end with um thank you Bri Brian again it's very good uh we can you can reach out to us through the Forum so if there's any additional questions uh for Content as well we will post the video of this uh to YouTube and then we also post the content in our community so you can see that uh and be able to uh pull down the data and you can also there's a Blog there you can also put comments in there as well uh we encourage you to sign up for future sessions as well uh you can also you also through the community have access to all our previous videos there's a bunch of a ton of content out there and then please you know feel free to provide feedback and again thank you for your attendance today and time and uh we'll kind of open it up for any kind of last comments or questions if not uh yeah we we'll adjourn today yeah thanks everyone for the questions and yeah like Jason said if there's any any followup happy to help if uh you want to learn more um we can get you in contact with my team and we can come out and have a conversation about this so thank you y thanks everyone have a great day appreciate it look forward to seeing you on the next Digital Services Forum bi-weekly session thank you