Bring the Agentic Enterprise to Life | Admin Keynote, Dreamforce 2026

Hello, and welcome to the admin keynote. It is so great to see so many of you in person and online on Salesforce+. Now, this is my 10th Dreamforce, and I've had the privilege of spending a lot of my career working alongside Salesforce admins. I've seen this role evolve and technology change, but there's one thing that hasn't changed, and that is the magic of this community. The people in this room and online who bring technology to life and make it real for your businesses and your companies. So before we get started, I just want to say thank you. Thank you for everything you do for each other. You answer questions, you share what works, and more importantly, what doesn't. You celebrate each other's wins, you welcome new admins into this community, and you help them find their way. And that is what makes this community what it is today. And this year, we celebrate 20 years of the Trailblazer Community. Yes, give it up for everything you've built for each other. And just look at how far admins have come. 137 admin groups, 120,000 members in 43 countries around the world. That is global impact. But behind these numbers, you're learning, you're growing, And you're discovering what's next. And now what's next is changing. Every company is becoming an agentic enterprise. And the agentic enterprise doesn't just change the technology you use, it changes the work itself. So there are a lot of new questions that every company is going to be asking. What work should agents take on? What access to data do they need? What permissions and guardrails need to be in place? And most importantly, when should they act? And when should a human step in? This is the new work of the agentic enterprise. And again, everyone is asking these questions, but who is going to answer them? Who is leading the agentic enterprise? I heard someone say it. I want you to all take a moment and look to your left and look to your right, because you are correct. The answer to this question is you. Every single person. Yes, give it up for yourselves. Because every single person person online and in this room has a role to play. Because the expertise the agentic enterprise needs isn't new. You already have it. You understand your users, who they are, what they need, how they work. You understand your data, what it means, who needs access to it, and the context behind it. You understand your processes, what can be automated and where humans need to be in the loop. And you understand the security that keeps it all trusted. Your expertise isn't new. What's new is how you apply it. And you've been here before. The admin role has evolved with every technology shift. AppExchange gave you new ways to extend Salesforce. App Builder gave you new ways to build experiences. With Flow, you could automate even more complex business processes than ever before. And Prompt Builder gave you new ways to bring AI into the flow of work. Now, if you're looking at this and you're like, Ella, this is not every change we've experienced, I know, I know there've been many more. But the point is, every time the technology changed, the role shifted, and you evolved with it and continued to lead the way. And today with AI Force, we're at another pivotal moment. What's possible with the platform is expanding, and your expertise powers every single level of what is possible with this platform. So we've all seen this slide. It's Thursday, so I know you've seen this slide a number of different times before, but let's dig into what this means for you. Your data that you manage, that provides that context for your agents. The apps and the automation that you build, that provides the ways that they can act. Those security, those permissions, the guardrails that you define help you determine what an agent can do and what it can't. And now with AI Force, you're taking everything to a whole new level because every single Salesforce experience that you're building within the Salesforce UI can be in any interface where work happens. Your expertise doesn't just sit at the foundation of this platform, it enables everything that's possible across it. And you're already doing this today. So I want to bring up someone who's doing just that. So please give an awesome admin welcome to Managing Director of Technology at Co-op Careers, Gordon Lee.

Thanks, Ellen.

Welcome.

I need one of these at home with my kids. This is pretty good.

I think we're going to take that away from you after this session. Thank you for joining us, Gordon. You have been a Salesforce admin for nearly 2 decades. You've been leading the way with technology and nonprofits. You've seen a lot of change over that time. What What admin skills have become most important?

Yeah, yeah. First, thanks for calling me out as a tech dinosaur for being here for 20 years.

That's great.

I think, oh, excuse me. I think honestly it's like the same 2 things that have, I've seen separate the good admins from the great admins. And it's always been the human skills, the people skills, right? And also the ability to know when to delegate. Usually when you start off in this ecosystem, you focus 90% on the technology and 10% on the people. And, and that's correct. This is a very technical ecosystem, so you have to get your confidence and your sea legs underneath you. But once you get that, you should flip the script and focus really 90% on the people and 10% on the technology because of the fact that AI is a tool. Tools are used by people, and at the end of the day, you have to understand people more to be of better service. AI has its place, definitely. It is the assistant that is amazing at technical work. And so I say let it do that technical work. admins have the experience and the judgment that simply cannot be replaced. And so I say delegate the work, but don't delegate the judgment.

I absolutely love that. I feel like we should all take a note, delegate the work, but not the judgment. And I love that you called out those people skills too, because I think there's a real opportunity today for admins to really step up and lead. And what's your advice for someone in this room or maybe online who's not feeling quite ready yet? What's your advice to them?

Yeah. I think there's a couple of things here. One is that the key is don't think that you have to become an expert at every single technology that comes out, but you do need to keep cultivating that mindset of curiosity and of ownership. Everybody is trying to figure out the best way forward with their AI strategy. Everyone's Googling it. Everyone's trying to frantically figure it out. But very few people I see are the ones actually leading their organization throughout that chaos. And so if you are like Ted Lasso, you should be curious and not judgmental. And so I say start small, right? Do what the great admins do. You start small, you iterate, and then you expand. Solve a problem for yourself and then solve that problem for your teammate and then solve that problem for the colleagues that have been complaining about that same thing for months. If you learn how to empower yourself, you will then learn how to help empower others. Now on the ownership side, I think this is where my nonprofit colleagues have a step up on this, okay? Us nonprofit people understand that we face budgets that are sometimes tighter than some of the skinny jeans at the concert last night. Just saying. And so we understand that we have to own the process all the way through. The good thing about that is that we see all the good and bad things that happen throughout steps A through Z versus someone who is only really siloed and they don't see much. And so I say that you have to basically learn how to connect the dots so that you can connect with each other, and then you'll become more than what you thought possible, and your org will thank you for it.

That's incredible. That's incredible advice. Thank you, Gordon, for giving us kind of a blueprint or a path to follow to really continue to lead the way.

Thank you, Ella.

But we're not done yet. We also want to thank you for all of the contributions that you've made to the community over these last 2 decades. So it is our honor to present you with The golden hoodie.

Right. Look at.

Thank you, Ella. Welcome to Life with Goldie.

Thank you so much. Thank everyone.

Give it up for Gordon one more time. Now, Gordon's story is such an important reminder. This— the technology is going to continue to change, but those people skills, those admin skills, that is what's going to help you lead the way. So today, we're going to bring the agentic enterprise to life and show how admins can do that with a ton of new technology. First, we're going to build trusted AI, taking you from a business need to a trusted solution faster. Then we'll deliver trusted AI everywhere. We're going to take that solution and bring it into every interface where work happens. And finally, everyone in this room knows the importance of security and making sure these solutions are secure as they scale. So we'll finally govern trusted AI. So let's start with build trusted AI. And to do that, please welcome Senior Director of Product Management, Tiaan Kruger.

Thank you, Ella. Hello, awesome admins. Before starting at Salesforce as a product manager, I spent over 10 years as an admin, developer, and architect building enterprise solutions for the enterprise on the Salesforce platform. As my team and I work really hard to evolve our platform into the agentic era, I lean on that experience to make sure what we build helps you get more done. So let's talk about AI. AI is not just changing what you can build, but it's changing how you build. At Salesforce, we're investing in a brand-new set of agentic tools for admins that help you get from business need to trusted solution faster than ever, leveraging AI in the process— build process itself. Like Ella said, you know your business, you know your users, you know your data, and you know the Salesforce platform. The right tool should amplify that expertise. And to help with that, we are launching 2 amazing new products today. First, Quick Pages. This is the ability for you and your most trusted users to vibe React, single-page React apps right in your org. On top of your existing data and metadata. This means you can unlock or unblock your power users to solve their own workflows and streamline those workflows with no risk to your existing metadata. Pages are going beta in October. It's amazing. But this next product— awesome. This next product I'm really excited about. You've been asking for it. Builder Central. This is a brand-new way to build apps and agents on the Salesforce platform. You start with natural language, you refine, you troubleshoot, you deploy, all from one experience. And the best part is, it's already compatible with your data, metadata, security, and governance. Builder Central is shipping as a beta in just a week. Just a week. So in a couple of days, you can start building. Now, I know your next question. It's the question you all ask. How much? Well, I am absolutely delighted to let you know that with reasonable usage limits, Builder Central will be free in every org.

All right.

Who wants to see what we can build with Builder Central? All right. So for our demo today, we're going to be following a fictitious company called Radiant Homes as their admin is building some solutions for them. This company does things like locks, lights, cameras, those kind of things. All right? And to help bring our demos to life, I want to introduce you to our amazing demo drivers, Kate and Heba. We ready to build? All right. So let's help you understand what Radiant Homes is trying to do. They have 2 problems they're trying to solve. First problem they're trying to solve is they're looking to build a customer site or a customer portal. Customers need to be able to get help. They need to be able to manage their devices, all those kind of things. The second thing they want to do is they want to build an employee app where their employees can manage these requests coming in from customers via the Data 360 feed from our partners where they can approve, modify, and generate device bundles for these customers. Now, to get started building, the admin, since we went to our sandbox, click the link, Builder Central fired up, no setup. It's just there. There's permission set for it. But once you have that, you just fire it up. Don't let that blank box scare you. You just start describing what you're trying to do. The agent will ask you questions and actually help you get to that right thing. But we're going to just ask you to build this employee app, this customer portal for Radiant Homes, and we're going to attach a requirements doc. Now, I recommend you do the same. You can give it a bunch more context, right? This requirements doc's got some big-ticket items in it, like our IT department asked us, can we please build this site with React on Experience Cloud? It's a thing now. React on Experience Cloud. Come on, that's amazing. All right. It also has some smaller details in it, like new homebuyers can actually get a $100 credit for installation. Note, this will come back later. And other details, what's in scope, out of scope, what's our folks, the people that we're building for. So we're going to go ahead and get it going. And the first thing the build agent is going to do, it's going to read all the document, the document and the prompt we gave it. And it's going to generate this brief. Now, what's a brief? The brief is here to help you see what did the agent understand? What is it going to build? There's some problem statements, solutions, some core features. And this is not just a one-way document. You can edit it via conversation like Kate's going to do here, or you can actually edit it directly and send it to the agent to update it. So if it didn't quite understand something correct, you can actually fix it here. before it actually starts building. So let's go ahead and give it the OK to start building. This is where the magic starts happening. So now it's actually starting to look at our org. And as it's building, it's looking to see what can it reuse, what does it need to build new. It's giving us some recommended things to reuse. We're going to just accept those defaults. And as soon as we do that, it starts being added to the project and the Explorer. You're not used to building with projects today. This is new. So give yourself a chance to get used to it. So in the Explorer, what's the Explorer? The Explorer lets you see all the stuff the agent's doing. Agents can generate a lot of metadata really quickly. So seeing what is generated for you is very important. Now, this ran really fast. And you're like, wow, that's amazing performance. No, it takes about 20 minutes, give or take, to build a real app. But we wanted to show you the whole flow today. So we sped it up a little bit. But let's look at some of these previews and visualizers we have. So let's open up the website, and we're going to look at that React portal. There's a live preview. You know, see, you want to see something even cooler? Click that edit button and pick one of the elements on the page. You have direct editing of that React page. We call this hybrid editing. You can edit directly. You can also edit via conversation. It's amazing. Let's look at some of the other visualizers. Data model, where you can see the objects got created and how they're related. Now, remember I said this is a project? This stuff's draft. It's not in your org yet. I'm showing you stuff that's not in your org yet. You can modify it before deploying it for the first time, publishing it. All right, let's look at the agent. We have an agent visualizer in there. We can generate some awesome agents with this, by the way. There's your agent. You can dig into the details. And let's look at the permission set, an admin favorite, right? Now that you can actually modify it with an agent. There you go. Great. Let's end on the flow.

Okay.

So there's a little sneaky edit button on every one of these visualizers. This means you have hybrid editing in every one of these. We're going to show you how this works. So remember that $100 credit I was talking about earlier? New homebuyers, they can't be a customer, To get that $100. And so we looked at this flow and we were like, it looks as a new home buyer, but it's not checking to see if they're an existing customer. So let's use that in edit. So then we picked the node and we're going to ask it to change just that node. All of the details get sent to the agent. And just like that, boom, flow updated.

All right.

Now, you are all awesome admins. I can see it. You know, you guys are all, like, paying attention. You're not going to just ship this to production, right? You're going to continue to revise this. You're going to publish it. You're going to test it. But let's say you've done all that. It's ready to go. You may even have used our incredible new preview orgs feature. We're not going to get into it here. It's too complicated. But come to our session at 1:30 in the theaters. I'd love to tell you more. It is awesome.

Amazing.

All right. So we are going to push this into our DevOps pipeline right from here. We're just going to ask the agent if it can push it to our next org. Now, this is not just pushing it straight to production. This is pushing it into your DevOps pipeline. So you have all of your testing running, your validation, making sure just the perfect stuff lands in production. All right. So just like that. Now let's jump forward a little bit and let's go take a look in production what we built. First thing we're going to look at is the beautiful portal. We have help. We have the ability to log in to Magic devices. You'll see that later. There's even a very cutely named agent named Ray that can help you answer all those magic questions. Let's jump into the employee site. This is a Lightning app. It'll build Lightning apps. And there you go. That's a beautiful Lightning app, right? Who knew Lightning can be pretty? There you go. All right. So beautiful Lightning app. And you can see that move-in concierge queue as it's coming in with these requests. Kate, can we dive into one of these records? Let's take a look there. All right. So fantastic. There we go. The Rhee family looks like they're moving. They're actually signed up for our concierge move-in service. Let's use the concierge agent to actually come up with a set of devices for them. So this concierge agent is going to look at the size of the house, what the customers have provided for us, and come up with a recommended set of bundle of devices that make sense for them. This is a beautiful example of an employee agent. The agent on the portal is a great agent for the customer service type. All right. There we go. Great. The employee can now accept those recommended devices, adjust it as is necessary, approve it, and send it to the customer. Pretty darn amazing. So let's recap. We started with a real business need. We then went to Builder Central. We used natural language to build a customer portal, employee app, even an agent. And then we actually went and modified a flow with natural language. And then we deployed to production all in one place. I would say that's how admins build trusted AI. Back to you, Ella.

Thank you so much, Tiaan. Now, I know so many of us in the room cannot wait to get our hands on Builder Central, but building is just the beginning. Now you want to take those experiences and deliver them everywhere where work happens. Let's deliver trusted AI everywhere. And to show us how, Please welcome Vice President of Product Management at Salesforce, David Greene.

All right. Well, thank you, Ella, and hello, admins. It is so great to see everybody here. During my career at Salesforce, I've had the opportunity to work across a lot of different areas of the platform. And don't get me wrong, I like pro-code, but I think the magic is in the world of low-code to no-code, which is this audience right here. So there is no other place that I would rather be for the next few minutes than with all of you, because like Ella and Gordon mentioned, everyone here in this room watching us live is uniquely positioned to bring the agentic enterprise to life. Okay, admins. Now, for years, you have been making sure that the right users get the right experience at the right time. The agentic enterprise That responsibility is actually expanding. Work no longer happens in one application. It no longer happens on one interface. Your users need work and help wherever they are, but that doesn't mean that all of you should have to rebuild that foundation every single time to support that. Now, the question is, how are we going to make that happen? As Marc shared with all of us on Tuesday, AI Force is how we bring the power of Salesforce across all of the different surfaces wherever work happens for both humans and agents, right? So, while the interface may change, the same trusted Salesforce capabilities behind the scenes that you're already using today, they'll stay the same. Now, the next few minutes, I am very excited to share with all of you what we are bringing from a product perspective to light up your admin toolkit So we can go support all of those users across every interface. The same data, automation, and business logic you have already built can power these new interfaces, right? While the interface may change, it's the same access, security, and governance that you're also already managing behind the scenes. So wherever your users work, they can rely on one trusted foundation from all of you, our admins, whether that's across Claude, Slack, and as Tian pointed out, Lightning too. So with all of that in mind, Kate, let's cover Radiant Homes. Let's bring this chapter to life. Now, to set the stage, our Radiant Homes concierge team lead, Maria, has just asked us for her own custom view of the upcoming sales pipeline for all the home closing events. I can just feel the grimace inside of the room when you hear that, right? There's a lot of ways we might have solved this in the past, but the next few minutes, we're going to show you how with Agentforce Coworker, we can actually give Maria her own autonomous AI teammate that knows her business from day one and can even act on her behalf. So here we are. We're over in setup. We're simply going to enable Coworker. After that, we'll skip a few steps, but we'll do things like configure some of our data sources from Data360. We're going to assign Coworker the right access to our users like Maria. And like that, our job is done. Seems a little too easy, though. So Kate, let's jump back over to Lightning as Maria. There it is. Top of the screen, Maria now has access to Coworker. And when she wants that custom pipeline view, she simply asks. Now, behind the scenes, Coworker is orchestrating across your data, your automation, your business logic. Coworker is even orchestrating across different agents to pull all the right information together at the right time for Maria. So as we start to see a response come back for Maria, instead of Navigating through custom pages, list views, objects, and filters, she's able to get that information she's looking for right there in real time through a conversation. So, I'm going to pause here for just a minute, like, let that sink in. What did we just see? We saw how all of you in this room with Agentforce Coworker can go supercharge every user in your organization with their own autonomous AI teammate. It's pretty cool, right? Okay, now, Kate, I know the concierge team loves Lightning, but they also love Slack. So, to speed the demo up a little bit, we've already gone ahead and configured an event-triggered flow that's going to look for real-time notifications in that Partner Data 360 feed, And bounces back into Slack. Now, the concierge team loves to work in Slack because of that multiplayer capability, right? So, as we see, there's a notification coming in about the Rhee family's upcoming home closing event, and while the concierge team is starting to all collaborate on what we think the right recommendations are, Slackbot's joined the conversation too. Not only can Slackbot help us out from a collaboration perspective, but Slackbot can also take action. So, here, Slackbot has gone ahead, it's actually executing one of the custom skills that our Radian Home Admins team built, and as we start to see this pop back in, we now have a rich and interactive response back from Slackbot that's showing us all of that information. What you're seeing on the screen right now, this is the headless experience layer. Now, think of the headless experience layer as the way that all of you, all of our admins, can start to go build rich, deterministic, and interactive UI that you're able to deploy across any interface wherever your users are working. So, as we think about what it means to kind of really supercharge productivity through agentic conversations, HXL and Slackbot, Not a bad way to go. Okay, Kate, we're going to go ahead and add a bundle to that quote. Behind the scenes, Slackbot's going to go ahead and execute our business flow, and as it goes through that, we'll say that for the demo's sake, we're done now. Okay, pause here for just a minute, because we've seen Lightning, we've seen Slack, but Kate, Let's keep this multi-interface journey going. Let's cut back over to that beautiful portal that Tiaan just showed us how to build with Builder Central, right? Okay, homepage still looks great. I didn't have a chance to break that yet, which is good. But let's go ahead and log in as our customer, Daniel Rhee. Now, as Daniel's logging in, he's greeted immediately with information about the upcoming installation that we've scheduled. He's seeing information about the existing smart home devices that they have. He's even getting personalized recommendations back on things the family might do to save energy. But this portal is not just about pretty pixels. This is always about business productivity, right? So, in the upper right-hand corner, we can see that that $100 credit we just authorized back in Slack, Is now available for the Reeve family to interact with and claim directly from the Reactbase portal. So, like everybody else, the Reeve family would love $100 off of their installation costs. So, we're going to go ahead and redeem that.

Great.

So, again, behind the scenes, credit redeemed, back into CRM, the Reeve family is set for their installation. Now, we've gone through Lightning, we've gone through Slack, now we've gone through a custom portal. But remember, in the agentic enterprise, work happens everywhere, and that includes Claude, ChatGPT, and even Teams. So for Radiant Homes to pull this one off, it's going to start with 3 little letters, MCP. For those of you who aren't familiar with Model Context Protocol, The way I like to think about it is this. This allows you to take everything that you've built and customized in the Salesforce platform and expose it out to any AI, agent, and app. Here in Setup, we can see that Salesforce actually provides several of our standard hosted MCP servers out of the box, so you can do things like access S-Objects. Pretty cool, right? We also have a couple of exciting betas that are out there right now that I would highly encourage everyone to go take a look at. However, in this case, Radiant Homes has built their own custom hosted MCP server for an escalate case flow. So Kate, let's go ahead and jump into that. Here, we see some of the basic details about our custom MCP server. The key thing being the server URL, you want to remember that one for a little later. But let's dig into the tools. This is a powerful screen. Our Radiant Home admin with clicks, not code, was able to just expose the custom auto-launch flow in their org, escalate case through their custom MCP server. So let that sink in for a minute. That means that any AI app agent across any interface in a secure and governed way, of course, is now able to interact directly with our own custom flow from the org. So, I encourage everyone here to think about what that means for you as an admin in this new Agentic Enterprise, and what we want to do to make it as easy as possible for you to bring your organization to life. All right, before we cut from the MCP screen, there's one little plug I have to put in, and that is for Agentic Identity. This is a pilot that we just launched this week here at Dreamforce. And I love agentic identity because for an admin, think about this notion of agents running around and taking action, right? Like, it can be a little interesting at times. Agentic identity is going to allow you to actually have more granular control over what agents can do within the system through things like scoped permissions and make every action an agent takes fully traceable. Just like the same level of traceability you have for your users taking action in the system. So whether you're just starting your MCP journey right now or you're already on your way, definitely go check that pilot out. Okay, our admin team's getting pretty tired, but we have one last step. So we're going to jump over into Claude. Now, in Claude, if we go into the customize section, We can see that the Radiant Homes team, they've already grabbed and configured and added some of our Salesforce-hosted MCP servers, and they've added those in. We, of course, can connect with a lot of other options that are out there as well. The key thing is, when we jump over into the personalized section, we can see that we have gone ahead and added the Radiant Homes custom MCP server. So that connects our case escalation flow, With Claude. Okay, to bring it home, concierge team has just heard about an issue with the Reeve family and their installation. So we're going to go escalate a case directly from Claude. Behind the scenes, Claude has discovered our tool. It's actually running against our auto-launch flow that we exposed. In this case, Claude's going to ask if we actually want to go ahead and execute that flow. For the demo, Why not? Okay, so just like that, we're done. I'm going to pause here for just a second, because like every great demo, you have to end with the data. So Kate, let's jump over into Lightning one last time. As the concierge team in the support case, we will go in, and we can see that right there at the top, Actually, now in the middle, our case has been escalated, and it even captured a reason for that escalation that we were able to provide from Claude. This is how Claude works with AI Force and how all of that works with what everything you have already built in Salesforce today. Now, I'm going to ask for a little bit of a big round of applause because, Kate, that was a pretty killer demo. All right, so to recap, we saw trusted AI move where our work moved, where our users moved across Lightning, Slack, Claude, and even a custom React-based portal. So this is how admins deliver trusted AI everywhere. And with that, bring it back up to Ella.

Thank you, David. So we've built our trusted AI, we delivered it everywhere, but every admin in this room knows how important it is that we need to make sure every solution that we ship is secure and trusted as it scales. So it's time to govern trusted AI. And to show us how, please welcome Director, Field CTO, Security Solutions, Rachel Beard.

All right. Thanks, Ella. And hello, awesome admins. I am genuinely so so excited to be here with you this year. I started my admin career 20 years ago, and it has evolved over time to where I am now as a field CTO, where I get to have security conversations with our customers every single day. And I've noticed that naturally these conversations are changing as AI becomes more and more a part of your business processes. Until recently, admins focused on securing your human users, but now things are changing and you have to think about how you secure your agents as well. And that's important because admins can— because agents can access data. Admins can too. Agents can access data and they can take action and they can operate across your business. So you really do need to understand what your agents can see and control what they can do, and it's important to spot risks early. You set the boundaries. So today, I'm excited to share with you 3 ways that you can stay ahead of risk, and that's important now and also as AI capabilities continue to scale. So let's talk about Security Center Essentials, which is helping you understand your security health and the overall configuration of your org. We're going to be talking about product— excuse me, Privacy Center, which is going to help you understand compliance in your environment and identify risks in your data. And then we're going to talk as well about transaction security policy accelerators, which help you stop risk and activities in your org that could be detrimental. So we'll talk about all of that. We're gonna start with Security Center Essentials. So I'm really excited about this one because security is a very complex demand that our admins have to manage. You have broad responsibilities, and a lot of times it's hard to get a full picture of what's going on in your environment. Security Center Essentials is a centralized and no-cost dashboard that gives you insight into your critical security metrics. So you can see a whole bunch of details about the security health of your org. No more clicking around between setup menus, trying to piece together that story and figure out where you're at. You can even get right to your health check from this dashboard. That's gonna save you time and give you additional visibility that you need. Okay. Really excited for that, especially the no-cost part. I know that's always a win. But of course, security isn't your only responsibility. You have to consider data privacy as well. Very—

Why not?

Did I walk outside the microphone boundary? I don't know. Maybe I'll stay put. What I was trying to say is that data privacy is critically important, and data privacy policies have really evolved over the years since GDPR and CCPA and other data privacy policies went into effect. So now you have these broad compliance policies with different types of demands, and it's important to understand those as well. What you're seeing here is the compliance agent And this is giving you an automated way to identify risks in your org and be more proactive about your security and governance. So this is really cool. The compliance agent has already run, and it's showing us some of the issues that it's flagged in our environment. So here, we can not just understand what issues it found, but we can see the steps that we can take to resolve these issues, our proposed solutions. We have a suggestion here to add data classification to our org. This is important because your orgs contain sensitive information, personally identifiable information, or PII for short. We want to be able to mark that data as being PII so we can layer on the data privacy protections that we need. So let's apply it and click continue. Great, we can confirm it. And success, it worked. That's great. So yes, please. I love seeing the ways that we can work agentic capabilities into data privacy to make a kind of boring and monotonous job a little bit easier for you to manage. Really helpful. All right, data privacy, like I mentioned, it's a very broad topic. It's not just identifying compliance risks. You also have these policies That are very different by region and by industry, and you need to be able to address your data subjects' needs to access or correct or delete their data. And you do this through DSARs. It's a data subject access request. We're making this easier as well. First of all, before we even go to policy building, let's check out this new UI. We've really cleaned it up. It looks really good. It's been refreshed. Looks good, right? All right. But it's not just a simple cosmetic update. We've also changed some of the functionality behind this. We're going to build a policy live, and as we do that, this is different too. You can see here that you're not just building data privacy policies based on your CRM data, but you're also including other types of data sources, actually, Sorry, forward-looking statements. This is not here today. This is coming soon. But you'll be able to add in Data Cloud and also Data360 and also backup and archive data sources as well. So you can bring together one consolidated output that includes that data subject information across all these different places where your data's information, your data subject's information is stored. So let's build that policy. We're going to start by bringing in the contact object from our list, and we already know our contacts are going to have PII in them. So let's make sure that we link those PII-related fields. And here we can even see suggested related objects and bring in the data from those as well.

All right.

Now, this is great. We just built the policy. Kate and Heba just did that so fast. But we need to test it. So let's do that. And this is also really simple. We just add in a contact and hit Run Test. This is now allowing you to create a simple policy that links all that data together and brings it onto one screen. The other thing that we've updated here is the ability to see the output. All that data consolidated on one screen, And this is giving you not just a view of what's being sent to your data subject when they've made that request, but you also get the linkage to what in there is PII. So really helpful for data privacy purposes. Okay, I know I just said a lot about privacy, which is maybe not everybody's favorite topic, but I do wanna say one more thing. So coming very soon, we're gonna be adding a new way that you can work headlessly with those DSAR policies. So I know you're getting, I see some nods in the audience, I know you're getting more of these as AI continues to grow. You have people who are updating their consent or how you can work with their data all the time. So soon you're going to be able to access those DSARs through Agentforce or API or APEX, and it will continue to work with that auto-detect PII function. It will let you manage policies, and you will also be able to execute those policies. on demand. Okay, now we can go away from privacy. Let's come back to security for a minute because one topic that I hear over and over again in my conversations is that there's a real risk of data being exported out of Salesforce. Data exfiltration has become a real problem, whether it's through APIs or report access. And many of our customers who have Salesforce Shield are not fully using transaction security policies to prevent those activities. So, I'm really excited to share with you an update to Salesforce Shield that's going to make that much easier for you. So, let's bring it back to Kate to show us, as an awesome admin who's building these policies every day, why is this new feature so exciting?

Thanks, Rachel. I would love to tell you about the Transaction Security Policy Accelerator because this is a really new, update that we have for Shield, and it is live right now on the Agent Exchange. You can see right here. So this is a Salesforce Labs app, Transaction Security Policy Accelerators, and it is taking Shield's event monitoring and allowing you to actually create your transaction security policies from ready-made templates. So this is— yeah, that's awesome. So, With transaction security policies right now, if you've leveraged these in the past, it has been a situation where you have to create these policies from scratch. It means you are either writing complex Apex code, or you're configuring multi-step in setup, and making sure that those come to life. But now, we've taken the most requested policies and pre-built them for you. And you can see that there are tons of them. I feel like I'm scrolling forever. It's really hard to do the scroll and talk. But it's fast. It's easy, fast. It's going to save you tons of time. In fact, it's so fast that I think we can actually do one together right now.

Yep.

All right. We've got just enough time. So I am going to head over to this detect high-risk data in report export. And I think that every single admin in this room would love this policy in place in their org because it is going to monitor your report export activity, and it's actually going to block anything with a field that has high-risk sensitivity. So, you know, when we just had the compliance agent up and we were classifying our data as PII, this is our next step for Radiant Homes that we can actually block that PII from ever being downloaded. So when we get to this screen, we have our description, we get a little information about the triggering event, We get to review the prefilled fields and fine-tune anything before we deploy. And then all we have to do to finish this up and take it home is enable this from setup. So, you can see that the policy is right here. Live demo. We hit enable, and we are live. How fast was that? All right, no matter how fast that was and how easy it was and how tempting it is to just go and push that button to deploy to production, I don't have to tell every awesome admin in this room that we follow a sandbox-first model. We always want to test in the sandbox before we rush to deploy to production. Thank you so much for letting me show you my new favorite feature.

Thanks so much, Kate. That was awesome. I love to see it. If you have Salesforce Shield or event monitoring today, please go download that. That's already available to you, and it's going to help you get started really quickly with putting protection into action with these transaction security policies. They run in real time. All right, we just covered so much great stuff. We talked about data privacy and how Compliance Agent helps you identify risk and take action in your org. We talked about both preventing data loss in your security as well as understanding your security posture as things change in your environment and in the world. So a lot of things happening here. I'm really excited that we were able to share that with you. And that is how you govern trusted AI. Back to you, Ella.

Thank you so much, Rachel and Kate. And give it up one more time for every one of our amazing presenters today. They did a fantastic job. So you saw how to build trusted AI, deliver trusted AI everywhere, and govern trusted AI to bring the agentic enterprise to life. And we showed you a lot of new innovations today, and we know how important it is that you can connect with the product managers that are building these features. So make sure to join us for our next True to the Core deep dive. This is a livestream series. You'll be able to join and hear more about Builder Central, and Tiaan and his team will be there. They'll share more about the vision. Yes, they'll share more about the product vision, and more importantly, they will take all of your questions. So please make sure to scan the QR code if you haven't already to join us for that session. And with that, thank you so much, and have a great rest of your Dreamforce.

Thank you.