I think we're we're good to go so uh hi everyone thank you for joining us um I'm zertalikai I'm an advisor solution consultant in Risk resilience and ESG here at servicenow uh based in London UK and my expertise is in financial services sector where I provide guidance and support to a diverse range of clients I've been conducting these um series of webinars focusing on integrated risk management GRC and this one um is actually the third session um throughout these uh webinar series we explore how servicenow platform goes beyond the traditional boundaries of GRC empowering organizations to transform their risk management and compliance processes today I have the the pleasure of being joined by my colleague Joe Montgomery so I'll have a hand over to him to introduce himself and then we can kick off yeah thanks Florida uh good to be here my name is Joe Montgomery I'm a solution architect here at servicenow and um like like blurta I focus on our financial services sector I support our financial services organizations here in the Midwest I'm based out of Kansas City and yeah looking forward to the conversation so thanks for having me Florida yeah so if we move on um before we we look at the agenda just some housekeeping rules during this session you will be automatically muted please use the Q and A feature to ask any questions that you may have uh we are recording this session and then we will be sharing this within the servicenow community forum and in the YouTube channel afterwards and following the sessions I think at the end of the session you'll be asked to complete a brief survey so we would highly appreciate uh your feedback in the end if we look at the um if we look at the agenda we are going to start with the significance of AI and ml in that with the organization working currently and how am I and ml can help streamline and optimize some of their processes some of the challenges and then I think specifically predictive intelligence can um can help on um Joe is going to talk about the Frameworks that we have as servicenow uh on predictive intelligence and basically go throughout those and explain like how you know take examples of like how every framework that we have will help the business Drive decisions we are we have prepared two uh different use cases to demonstrate uh where we will basically go in and explain you know some of these Frameworks that we uh we will discuss previously and then I think if time allows we will just unveil some actionable insights um from the platform data you you can basically have um and leverage uh considering the servicenow is a platform so I think we can all agree when I say AI is no longer just and I still have Edition because already has revolutionized our daily lives making tasks easier more efficient so whether it's just finding answers to complex questions resolving issues or just even automatically addressing employee inquiries it has become an invaluable tool and drastically transformed the way we search for information so with AI we can now obtain accurate and relevant answers quickly this saving assignment effort and the Ali algorithms have the capability to analyze vast amounts of data to provide us with the most comprehensive and up-to-date information available and as we navigate through the risk compliance and and resilience World AI continues to play a significant role as organization can effectively assess and manage risk identify vulnerabilities and can ensure compliance with regulatory requirements so basically AI powered analytics help analyze large risk and compliance data sets identify patterns and generate insights that Aid in decision making ultimately enhancing the overall resilience of the business so in today's business environment traditional GRC practices are inefficient and platforms need to embrace current trends if they want to allow customers to automate time-consuming activities and enhance risk management resilience and compliance programs so as we explore some of the critical challenges that organizations encounter in their GRC efforts and have an impact on operate patients and Innovations I name a few process optimization so manual actions and outdated processes often require systematic Improvement efforts to streamline operations so inefficient and time consuming tasks not only hinder productivity but also increase the risk of errors and inconsistencies so organizations needs to identify areas for process optimizations and embrace modern approaches that leverage technology and automation to drive efficiency and Effectiveness in their DRC processes for data duplication and reconciliations we know that one of the recurring issues in GRC is the creation of multiple records by users leading to data duplication this unintentional duplication creates confusion increases the risk of an inconsistent data and requires substantial efforts to reconcile the information so there is the need of robust mechanism to identify and prevent data duplications ensuring accurate and reliable information for decision making when it comes to limited visibility into historical data and Trends we know that without access to comprehensive historical data and Trends organization will struggle to identify patterns insights and opportunities for improvements so this limited visibility hampers the ability to proactively address risk or identify emerging Trends and even making data-driven decisions so to overcome this challenge Advanced analytics and Reporting capabilities are needed in order to provide a holistic view of historical data enabling leadership and risk and compliance seem to identify Improvement areas and have a chance to enhance their programs uh slow Innovation speed so uh Legacy tools and processes often impede the pace of innovation in GRC Converse the manual workflows outdated systems they slow down teams and delay the delivery of new initiatives so this limits the ability to adapt to changing regulatory requirements evolving risk and emerging Technologies hence there is a need to embrace modern GRC solutions that leverage automation Ai and collaborative platforms because these accelerate Innovations and these enable organizations to stay ahead in this rapidly evolving business landscape so embracing modern Technologies Automation and data-driven approaches will enable organizations to overcome these challenges and effectively unlock the potential and navigate these complex uh landscape of governance risking compliance Joe how can service now resolve these challenges and what are the solutions we provide in the space yeah great that's that's awesome Florida so I will let's take a step back here and cover I want to talk a little bit about just kind of artificial intelligence and machine learning and then we'll talk through servicenows function which is predictive intelligence um on the platform to solve for some of those common common challenges and common areas that AI is is leveraged to improve business operations but where where I want to start is you know artificial intelligence and machine learning like we often use these these things in the same breath we speak about these as they're the same um when in reality machine learning is a subset of the larger artificial intelligence domain so if you think about you know voice recognition Alexa and uh Siri right that's part of the larger AI domain if you think about chat GPT and generative AI which is sweeping the world and changing the world at a pace we've never seen before um that's that's larger artificial intelligence domain and there was a question I saw come in in the Q a around do is service now starting to leverage generative Ai and what's the roadmap in addition to these current machine learning capabilities so there is a road map I believe Florida correct me if I'm wrong I believe there's another webinar coming up to cover specifically the generative AI capabilities that are coming out um do you happen to know that offhand yeah so like basically this uh webinar specifically uh focus in predictive intelligence the one I think around September will be basically focused on generative AI chat GPT and you know this uh this sort of uh how this this two sort of Notions can help one specifically on risk and compliance perfect so and today yeah so today we're going to be focusing on predictive intelligence which is servicenow's machine learning capability so when we think about predictive intelligence you can think about the servicenow solutions we have available for machine learning and predictive intelligence provides four Frameworks that you can use to create machine learning Solutions in your instance each one of these Frameworks delivers a different solution type for training the system to predict recommend and organize data so a trained solution can can be called to predict the outcomes and these four different Frameworks will talk through each framework here um briefly before we kind of jump into the platform but the classification framework of predictive intelligence allows us to set field values during record creation um to categorize and Route work based on the data that's in the system so based on past records the similarity framework allows us to identify similarities between new and existing records to recommend Solutions so the similarity framework is what we use in the out of the box machine learning capabilities for integrated risk management that we'll talk through so grouping of issues grouping risk events recommending remediation tasks those all leverage the similarity framework so that's where we use from an out of the box perspective and we'll see some of the configuration of those the clustering framework is grouping similar records into visual clusters to really identify patterns or reoccurring issues reoccurring events and then finally the regression framework is used it's more of a numeric output so it's used to really predict the time it takes to resolve an issue or the probability of of issue resolution but it's using that historic data um to predict the numeric outputs so those are the four primary predictive intelligence Frameworks that we have we'll dive a little bit deeper into each one here before jumping into the demo the classification framework is again this is used to set data to to set field values during record creation so think of this this example as an a user is reporting a new issue so whether this be through virtual agent or through the employee Center but as they're describing the issue and the name or the short description the language that they put into that field it's going to trigger a business rule which is going to call the prediction AI API which is ultimately gonna going to call the model that's been trained to look at the language that's in that issue so the prediction is going to say because because because you know customer data unauthorized access to customer data the prediction is going to be the issue type is data breach another prediction would be the assignment group or the classification would be compliance and then based on all of the data and the training jobs that have run the confidence based on the language that was provided the confidence is a 92 percent and the threshold that we had set in the model was 80 percent so the 92 percent confidence you know goes above the threshold so what we do there is we can automatically and during the record creation process we can say that the issue type is data breached we can classify it as as a compliance issue and we can you know even do things like make the issue rating very high right so what this allows us to do is because all of that data is set during issue issue creation or the initial input of the issue we have faster resource allocation so you know we can assign that to the compliance team automatically we have a quick time to Value because we you know we don't have the initial analysis of you know the human subjectivity analysis of what type of issue is this what you know how do I classify it what should my issue rating be we can do all of those through the machine learning capabilities reduce error rates for the same reason and of course the increased analyst ability to focus on what's the most meaningful task so because we can rate this issue as very high it's a data breached we can prioritize we can escalate and do different you know Downstream processes because of the information that the machine learning solution provide provided us so that's an example of the classification framework in action the similarity framework again this is what we use with the out of the box irm machine learning capabilities that we provide so this is recommended content really to help analysts solve issues faster um so the out of the box solutions that we provide with machine learning is identification of similar issues compliance cases and risk events so the ability to you know create a parent-child relationship among issues or group issues together to remediate and mass we can propose new issues or link to related existing ones and then we have the ability I think this is the most powerful example of what we provide out of the box is using predictions to help analysts investigate and remediate problems faster so based on and we'll go through this during the demonstration portion of the webinar but based on the language in an issue the name you know who's assigned to the short description or any data points that you define we can we can recommend remediate tasks remediation tasks to resolve that issue so based on similar issues and how they had been resolved previously we can recommend actions that need to take place in order to is that issue in a more expedited fashion okay the clustering framework is going to be you know grouping similar records together really to optimize efficiency so identifying patterns with with unsupervised machine learning not going to go too deep into this today but some of the Frameworks um use unsupervised machine learning and some of the Frameworks you supervise machine learning so it's really about is it labeled data or not um and so this allows us to really you know uncover automation opportunities fast it's going to provide a visual clustering of similar records so that we can take action quickly and prioritize different actions so that's the clustering framework and then finally the regression framework that we have is again it's going to be more about numeric inputs and ranges so think about being able to predict the time needed to solve a risk event or an issue so based on similar issues or risk events in the past and the time it took to remediate and ultimately close out those issues we can have an estimated time so that we can have you know a good understanding of what this means to the organization the resources it might take to resolve the issue and again another example would be predicting the probability of remediation success so of similar issues you know what what was the problem what is the probability that this will be closed out within the certain time frames or slas that we've identified now this is sort of a uh an honorary framework I would say so analytics Hub and we we have blurta had done a webinar specifically on performance analytics but because of some of the capabilities that that analytics Hub provides us things like uh being able to forecast feature data see Trends across data um any report that we have built with performance analytics when we go to the analytics Hub it's going to give us a really nice visual be able to compare quarter over quarter see Trends predictions breakdowns all of that so we include that as an honorary member of the of the predictive intelligence framework and we'll talk about that um when we get into the demo as well so yeah and like you know we wanted to give you like a sort of understanding of like what are those sort of algorithms or what we do in space of predictive intelligence before we drill into a demonstration this is where we can now show like some of these examples or some of these algorithms and how they apply actually in your you know your day-to-day activities or like um specific areas that you you work on so now if I go uh and I bring the system instead I should be um I should be able to show exactly like you know how you can identify like you know in this example we do have um a risk risk event so uh specifically this is a a risk event view it's designed for business uh professionals that are familiar with risk processes they are responsible to manage risk events let's say within their business unit so um the information that they can see on the overview tab is basically a visual and representation by the input that we have provided so far on this risk event uh conceptually with this this view provides analysts the ones that are currently analyzing this risk event as well as the second line and later the approvers with a Consolidated uh perspective on the on the event it highlights what has what has happened what a cure the current stage of the event the actual loss and you can see that's basically both on quantitative and qualitative terms which then also encompasses the non-financial impacts as well so but what's important here if we look at the um right side in here we do have this uh similar risk given feature so this is the recommendation engine that provides insights into turning risk into them that have you know those risk incidents that have uh happening that are trending or that have you know currently or have been they you know it also looks historical data not just the current one that's currently currently happening this feature is you know is invaluable for analysts especially for example when it's needed to understand uh similar risk incidents how they were managed how they were contained uh you know what were the tasks what were the root causes action and treatment plans because I can just go and click and I can see information and or like I can straightly go to this specific risk events and understand what has happened and how this you know this risk event that looks similar as mine uh maybe has been uh has been uh remediated so um this feature like you know the one that you see for recommending actions here additionally allows me to to determine whether this you know risk incident that I'm currently analyzing is the duplication of an existing case within the system so you can see those two look exactly the same so then I can identify and I can you know reject this one and link to the related case so that this way I can avoid unnecessary duplication uh of efforts because this already has been entered in the system already been analyzed so like I can just uh you know remove duplication I can assign this to a an already case that uh is being currently analyzed in the system so it not only validates the data but also helps me and my team to to cope with a substantial amount of information so I can analyze manual work to just go in and see like if similar events have been entered before I go and analyze this uh this risk event because I already have this information I already see that the platform is leveraging existing systems so it helps me streamline my analysis and overall I think it helps improving the end-to-end risk brisk event process so this is one of the features in addition to this feature the platform is also able to provide valuable guidance and what I mean by that you can see we have this recommended actions that's part of the in in the side panel in here what this does provides me recommended actions right within the same view so um ensuring that I have access to the relevant information without disrupt accepting my workflow so usually like you know when you're working on something you have to jump on different multiple tests to find out like you know how you should approach this type of risk event or how what what sort of the next steps that you should do in case you've confirmed something so like basically I'm able to see this information I can seamlessly refer to the guidance that's in here without you know stopping what I'm currently doing and I can continue with my analysis of this risk event while looking at this recommended actions that have been provided to me so um the platform enables me to include this recommended actions enables me advise and suggest and suggest specific actions that uh uh should be performed based on the analysis that we are currently doing on this brisk event so uh you can see you can see um helpful prompts that guide me towards what's the most appropriate course of action that I should address to manage this risk event effectively so I can see a guidance like I can go through videos I can check what what is this I can look that this is basically an actual loss I should go to risk event entries and I should you know provide you know what what where the losses you know based on the click Behavior if I go and I say this is a compliance breach for example what the system will do like what this recommendation action will do you can see now it's recommending that I have a compliance breach confirmed within this risk event so I should go in and I should complete the compliance breach section so it's analyzing my click behavior and it's recommending um actions that I I should do so um concluding like you know as a summary of like what predictive intelligence can do here that only enable me uh look at the similar risk event but also providing me on-demand guidance and offering me recommendation action based on the my click behavior in here so like it while I'm looking at this screen um while analyzing this risk event I can make informed decision I can go I can look register the losses I can go I can complete the compliance Bridge section so it guy it it it it's guiding me to make those informed decisions and take appropriate actions while maintaining a seamless workflow in here so this is uh in I think we already mentioned now intelligence it's a platform feature and it extends beyond the the rescue been so like this is just an example about the state to breach if I look at another example you can see it's another type of risk event that I can see all those similar risk events I can see because currently we are in the new analyze state so if I go to the recommendation action I will see what you know sort of recommendation actions of the the system is is suggesting to me so all this sort of stuff it's built within a risk event out of the box and it's built with our now intelligence or predictive intelligence platform features that we have so everything that you saw here that apply to risk event it can apply to every risk and compliance stable within servicenow platform and then if we go in I can show you the ease of predictive you know like how you can manage those so Joe earlier went through and gave you like an overview of like what each of these algorithms can do so like um I think he mentioned that like you know one of the things that we have out of the boxes a lot of a lot of you know solutions for similarity and you can see this is one that says similar risk events when I can go in and I can configure so you can configure similar Solutions this one just to show you the ease of configurations and how simple is to tailor the platform to the various use cases that you may have it's basically you with creating a record you were defining what would be the word Corpus for you know that specific table or specific process that you have you can see I have selected here risk event you if I just click on here I will basically have access to every table in servicenow where I can you know start building this predictive Intelligence on top of like you know what the uh the platform is allowing me to do and then I can set training frequency I can update training frequency and I can do all of that like while doing configure configuring it's nothing complex than just doing configuring are looking like a word Corpus looking at the table looking What fields do you want to look like how you want to test it if you have any sort of filtering and then once you do you know you update this this is where you start uh training the solution and get those you know recommendations that we we saw earlier and one thing just add their blur to the the word Corpus I I've learned this over time but that's really the vocabulary of the model right so any any of the language you can Define within the word Corpus and it will create a default one for you as well so you know if you go in there and add words you can choose which fields to select that would be included in the vocabulary and that's what the model actually trains on so the word Corpus is a is a big piece of it but not everyone may know what that is and yeah yeah and that's where I wanted to say that actually all this stuff is again a configuration that you would be uh you would be doing because you already have the the platform you already have the predictive intelligence algorithms out of the box it's a matter of you picking which one would be the right for you based you know on the definition that we went for earlier and then you can set up this you know creating a new record um looking at what table do you want to apply this and what sort of uh criteria do you want to make uh as part of this you know um algorithm that you're trying to do for similarity in this case for risk events and this this is just a use case uh Joe now I think he I believe you will be going through a different use case uh on on predictive intelligence when you can show the um like similarity on issues like suggesting remediation actions based on what the issue is and then I think drill some spend some time on analytics as well and how you can drive decision based on the data that you have in the platform so let me share my screen here sorry about that yeah yeah and while you do that I think we do have some uh questions that I can definitely answer on on this when it says what should I do to enable predict the intelligence in servicenow instance that's predict the intelligence work using apis so basically this is the platform features you don't you don't need to do anything like if you search with predicted intelligence you should be able to find you know those algorithms that we I think Joe will go through as well and um it should be part of this servicenow platform it's a platform feature and then we have is is this part of the advanced risk capability in Utah or currently in AIML is predictive intelligence out of the box in Utah for irf so predictive intelligence it's a platform feature again it's applicable to the to the to the platform and then it applies to everything risk related we bring out of the box use cases every release that we have within rest and compliance for example you know the one that with risk events that you saw it was released in Tokyo this one that Joe is going uh going to go through I think it was released even before uh before Tokyo I don't really exactly remember the um uh the the release but what's important to maybe mention in here is that every release that we make like we have twice releases a year we make sure we do have out of the box predicting intelligence use cases embedded within those those releases in in all GRC Solutions and uh included on the on the road maps uh yeah and I think the last questions before you jump in joystick is there any now learning courses available I believe there are there definitely are I was just looking um the other day and there's there's quite a few there's a there's a fundamentals there's an implementer there's quite a bit of of content out there around predictive intelligence so okay so let's jump into just another quick use case just to show you another flavor of how we leverage predictive intelligence so what I'll go through here these around issues management this is all part of the out of the box uh machine learning capability we have with irm so again all leveraging the similarity framework of machine learning so what I'll do I'm just going to go ahead and let's create a new issue here um so I'm going to go ahead and just give it a name and a description so we'll say you know a new issue all right resulting well slowly error okay so we'll just give it kind of a generic name here we can give that a description as well and I'm going to go ahead and just save that record so the idea is that you know somebody may be reporting a new issue and maybe coming in from multiple Avenues right virtual agent employee Center um however however the organization organization set up to bring in an issue it might be you know from a failed control or a risk assessment right issues can come from a handful of places but based on just the basic language that we provided so the idea is that the more data that you have the better the the machine will predict the outcome so that's part of the training job so just based on some some basic information we have today here I can go up to my assigned to and I can click my suggestions and we can see who my potential assignment for this issue can be based on the confidence score so again you can always see the confidence score of what the model how confident the model is that that's the right selection we'll go ahead and select myself as the assigned to so that's one of the components is is being able to assign issues in an intelligent way we also have the ability to go over here and look at my parent issue so when we think about issue grouping you know for resolution purposes or creating yeah you know parent-child relationship to inherit remediation tasks and things of that nature I can go ahead and again click on the suggest light bulb and we can see what other issues may be may be associated with this issue that we're creating right now um to be able to group these together right so let's just say I need to go ahead and you know I want to select this issue and I want to group those together that gives me the ability to either create a new group or group with any other issues that we have and provide some additional information assigned out in a way that is Meaningful so what the model is doing is basically just looking at the data giving us giving us a confidence score and allowing the analysts to you know rather than going and searching we can just see you know are there any similar issues that we want to associate with this issue again to really really limit the number of duplication or duplicative issues or risk events that we have in the system to you know free up the analyst time to work on things that are actually meaningful okay so those are two of the primary components from an issued Management on the predictive intelligence side the last one that I want to highlight here and as we go through and let's just say you know we're going through the analysis of this issue um and we need to scroll down and we actually want to look at this record here sign this out so once we have the issue signed out and we're actually going through you know we're working the issue we've decided we're going to remediate the issue but we want to see you know what based on all the information we have what are some ideas for remediation tasks so I can come down here to my remediation test and I can select suggested remediation tasks and again we can see where we may have remediation tasks that were associated with the resolution of a similar issue so we're looking at all the issue data any closed issue that had specific remediation tests associated with it and just you know Show an example again this is It's demo data the more data you have on the system the better the remediate the the predictive intelligence or the remediation is going to the more accurate it's going to be right so let's just say I find one that I want to copy in we can just select copy and that's gonna it's gonna create a copy of that associated with the existing issue that we have and then we can you know we can say that assign it out and and complete that work in that way so those are the three kind of key um areas that we have predictive intelligence again those are all out of the box models that we have built or solutions that we have built um and I think the what I'll do the kind of the important thing I want to highlight here is let me go to the predictive intelligence homepage so we're just showed this but the key thing is we do provide predictive intelligence Solutions out of the box with risk events issues compliance cases right but the idea is that we provide the platform has a solution so you can build Solutions or models to to enhance and optimize the business operations where it fits your organization so rather than just saying here's where you need to use predictive intelligence we provide the mechanism so that you can Define where you want to use it and you don't have to be a data scientist to be able to build some of these models so if I just go down to and learn to show this a little bit but let's just find our solution definition for assignment to the issue right so I can open up this model and again this is within our similarity framework um we can see we have a word Corpus again a system generated word Corpus which that's that's the dictionary that we want the machine to understand when we're recommending assignment or similar issues or you know recommended remediation tasks we Define the the table so the target table you know where we're actually going to be providing the recommendations to which is going to be issue and in this case we're looking at issue to actually learn as well so the test Fields we're looking at name and description and those are also what we're going to be recommending against do we need to have any filter against that data so we don't want the model to learn on any issue that's in a new status we only want the model to learn on issues that are in an analysis respond review or close complete status and you may even for things like remediation tasks you're probably going to want to filter down to only your closed complete issues because you want those remediation tasks to be valid and ensure that they were sufficient tasks to close the issue so you can really Target the model to the data that's relevant to what your you know what the objective is or what the solution is um so again you can Define these any table that you want to Target you can build these models you can build the filtering of the data um and you can Define the training frequency as well okay and then you can see you know when when was that training job last last run if you have the you know appropriate roles and privileges you can come up here you can update retrain that job but typically I think the default training frequency is going to be every 30 days so that gives you a good understanding of just you know how it's it is a front-end configuration you don't need to be a data scientist to to Define these algorithms and really understand what what's going on here um and I think that this is a really strong solution to provide flexibility and where we introduce predictive intelligence to you know the operational efficiency of your organization yeah and to mention I I'm glad you said that you don't need to be a data scientist to to do to do this because basically the algorithms are built in there so you just need to set up the criteria the data and like where this algorithm should have a look to generate this results they already built in as part of the productive intelligence featuring servicenow okay I see a couple questions um predictive intelligence use cases for irm documented yes they are so there is a I believe Florida Correction from wrong I believe there's an a predictive intelligence for irm plug-in that would bring in the different um machine learning solutions that we have specifically with irm all of the pre-built Solutions but those are also documented on our on our doc site I think maybe we can um we can publish those out if we're if we're sending slides out or making things available but there's a lot of information about what we provide out of the box out there publicly available and then how can we sorry how can we rely on it without further investigation reviews um with the confidence score of 60s yeah so that's where the confidence score threshold comes in so you can Define especially if you're automatically with classification framework you're automatically setting field values and that's where the confidence threshold comes in because you don't want to automatically set field values on a low confidence prediction but when you're talking about similarities where we're not automatically setting the field values we're recommend we're basically providing the findings of the model with the confidence score so you can determine where you you know where it's relevant and where it's not relevant so it's not going to group the issue at the ways to find out of the box it's not going to group the issue without um without the additional step of creating that relationship and then is the machine learning solution used for only a prod environment um that's a good question I know you can you can build them in in non-prods um but I I'm not sure Lord I don't know if you haven't yeah I think yeah exactly like you know the uh the nine I think ninety percent of uh accuracy on your um let's say predictive intelligence algorithms comes from the data if you know that in your test environment you have data that maybe you could do informed decision that I think you can leverage it's a platform feature you can enable it on your test environment but um usually you would have those data and production environment and if you would want to take those important decisions based on you know the data you have that's where you would enable it if you know that you know you may have a copy of this data and test and you want to look uh try and run in test environment I I don't think there's something that would stop you to to do that so yeah and then can we have all four Frameworks at the same time um or only one framework yeah the the four Frameworks are all in there and you can use those four Frameworks as it's relevant to whatever the business outcome your you know you're searching for so you don't you don't pick one it's just when you're building a model you would identify what you know what's the outcome do we want do we what's the best outcome and does that outcome is that outcome provided with the classification model or the similarity model or the clustering uh model or framework so um but yeah you can you can have different solutions built with the different Frameworks as you need to there's another question so I'm asking about the volume threshold for implementing this AI algorithms you can basically you can put the thresholds while you know I think we saw you the back end of like how you set up these you can set up like what would be the minimum data you need or as a threshold like as a limit for the algorithm to work the key of this thing is like as more data you have it's better from for for the prediction to work so like there is you can decide on like what the limit is but uh uh it should the vast amount of data more data you have it will it will have a better uh accuracy results or you know information results at the end so last thing just to show here real quick Martez on with the analytics sub which again is is not exactly a pi framework but one of the capabilities that we have and we have a lot of reports within irm built with with performance analytics but just to show you what the analytics Hub looks like uh when we open up a specific report we can see you know the trending over time and all the all the you know Associated data if we want to expand this out and look at a larger subset of the data we can do that um you can compare you know quarter over quarter and this issue is you know or this report is new issues created so there's a lot of capabilities in here again it's not a predictive intelligence framework but it does give us the ability to kind of play with the data if we want to you know add a trend line to the report here or do any kind of forecasting based on the data that we have in the system we can do all of those those things within the analytics Hub so again it's a powerful tool to be able to see and kind of understand what you know what's happened and what might happen based on that data that we have in the system I guess like before we just went on and conclude this there are a few other questions that maybe we can quick quickly um answer I think one of the questions is can we enable this um intelligence in the PDI uh with demo data I think uh uh no I'm not sure I never had like a PDI instance but I I I I I I'd assume it's basically it has all capabilities that uh uh an instance would have with you know uh so like I when you install the irm plugins or the GRC plugins you should be able to see the the um GRC intelligence plugin as Joe mentioned as well and then I think you you would be able to to enable uh enable this and um I think one of the questions is regarding the privacy of my data how is my data used in training the AI model where are they stored is there a documentation so basically it's your platform it's you know you'd be doing this on your platform whatever you you uh you service now as a cloud platform so it's like uh will be uh stored on on cloud and I think um and the contractual terms that you had the service now you already have decided on on the Privacy data and all those things it basically will be your data and you would be managing your data and use the platform will use your data to leverage and um predictive intelligence for informed decisions so I think uh I uh we did have one last slide uh to uh to show like maybe I can uh crap screen from you uh Joe to uh basically maybe just conclude this um conclude this uh this uh webinar just wanted to make sure that you know the examples that we provided were useful so you could see like actually how it could basically help you know different roles within the organization on their day-to-day work activities so I'm not going to spend time to go for each of how this would work but uh um I guess the predictive intelligence can transform the potential um for various stakeholders within the organization the chief risk officer can get better inside and better data-driven analytics for proactive decision making so I think that's what that's something that Joe showed on that predict uh that analytics um uh dashboard the issue to our nurse can benefit from Automation and the efficiency on the issue resolution again that's something that we showed Joe show like how you can you know look uh uh bands and decide like you know what sort of remediation plans are being suggested to to solve that issue uh with Risk Managers you can streamline processes you can have decision making by you know providing uh recommendation actions to them and then I think essentially the end users you know having those guidance having recommendation actions it it go it gives them a a way to and enables them to take informed decisions um on not just in issues and risk events but in all GRC objects that they may be dealing dealing in their day-to-day um activities so I think in conclusion like we want what we wanted to show here is how you can you know use what the platform has in predictive intelligence for risk and how this can transform like you know a various roles within the organizations how it can help them take those informed decisions and then we we aimed I guess to to show you the art of what's possible uh on servicenow platform leveraging predictive intelligence features and there is so much more that uh you can you can explore you can achieve with predictive intelligence so I would advise you to reach out to us if you want to know uh more about this we would be happy to take those you know uh questions from your side and and show you how you can you know leverage not just you know predictive intelligence in this case but all those platform features to enhance your risk and compliance programs um I think we have all those questions answered as well um so thank you everyone for your time today and uh I think this uh webinar will be um will be uploaded to YouTube will be uploaded to uh on our servicenow community so like if you have questions afterwards please don't hesitate to reach out thank you thank you