[Music] [Ben Robbins] All right. Thank you everybody. Thank you so much for joining us today. We really appreciate it. My name is Ben Robbins, and this is-- [Jason Baron] Jason Baron. So we really appreciate you coming in today. What we're going to talk about is how you can actually use AI and what we're building with inside of Workfront to simplify how everyone in your organization can use Workfront to get some of your everyday things done and probably solve a lot of challenges that have been very painful for you for a long time, where you've just said, I wish someone would come to our rescue. So we hope that when you leave this session that you say, yes, you came to my rescue and we're very happy. Now, before we get into all the details, what we're going to show you is confidential. Some of this stuff is very new and it's just coming into the alpha stages. So you're some of the first people who are going to get to see and hear about it. So we're excited to share it. Just remember, try to keep it on the wraps. So as far as our vision, Workfront looked really hard at what the market was doing with AI, and we took a strong look at our competitors. And one of the things that we recognized that our competitors were doing that, we didn't like, is they were just building a bunch of point solutions, just a bunch of single solutions that they could put on a public press release, and say that they had some AI in their product, but it wasn't really solving customer problems. We didn't feel like that was good enough. So we wanted to build a solution for our customers that helped everybody within their organization. In order to do that, we established a North Star. And our North Star was just to take AI to make Workfront more approachable, easier to use, and more powerful. When I say approachable, that means we want people who aren't just power users or project managers to be able to have certain capabilities inside of Workfront. We want everyone, the marketer to creative that feel comfortable in that home when they're in Workfront and not feel confused. We want to make it easier to use, so that you can take more of your capabilities and you don't have to navigate across Workfront to find information, information is coming directly to you. And we're making it more powerful. When I say powerful, I don't just mean new workflows that you couldn't achieve before. I mean, finally making it easier so that more information's coming in the Workfront. So not only can you do more with it? But you have more data and information. So you're reporting and how you get campaigns and projects that work out the door, that's power too. So that's really what our North Star is, when we envision how we solve for AI with Workfront.

So this is a pretty big slide. I'll step through it very quickly. It's just how we look at a day in the life for our customers. Everything starts with an idea. Someone says, we need to do a campaign to solve x or to reach a particular market. And the idea starts very simple. The words are very clear. People understand what the objective is. But then as you really start diving in on what you want to do, the task becomes harder and harder. What started with just a simple campaign with a few hero assets, all of a sudden starts meeting the realities of the real world for how campaigns go out. When campaigns go out, they hit multiple segments, multiple channels, and you have to have assets that are spun up for each one of these different variations for what you want to accomplish. So it's not just one asset for a campaign. The number of assets that you have to deal with grows pretty quickly. And not only do you want to send them out to multiple areas, but they also need to be responsive for more form factors. So you have to then augment and size and scale the assets for that. Sometimes you want to send them out to more regions, so it grows and grows. And then you want to measure off of those assets. So when you think about the campaigns and the things that you're doing, we noticed our customers had lots of pain points when it involved their assets, and we decided that was going to be where we wanted to anchor, how we establish a lot of our AI philosophy. So how we can really help our customers because let's be honest, asset proliferation is real. Managing those assets, finding those assets, making sure that they reach the right people, making sure that they go through the right branding, right compliance, that's several people's full-time job. And the more that we could help there, the more that we could help an entire organization. As I said, the very beginning, going back to that customer-oriented vision.

So we started by creating value streams, ways that we approached work for each persona, multiple people within an organization, up and downstream from how they work with the assets. You heard a lot about the content supply chain earlier today during the keynote. So we wanted to make sure that our value streams reflected how people use a content supply chain. That was the way that we knew that we were going to do better than our competition, to make sure that we were addressing multiple people and that we were speeding up work across an organization. Because if you speed up for work for just one person and you're not speeding it up downstream, then you've just created yourself a bottleneck. So we wanted to make sure that from beginning to end, we put AI in place to help solve problems for our customers.

So we immediately came up with five core ways that we were going to be able to help our customers solve what they needed and what was most important to them, all backed by the power of AI. The first two are generative requests and our brief AI assistant, that's where we start to reduce work immediately from the very first time you start building a campaign inside of Workfront, we're making your life easier. What we found is our customers oftentimes take unstructured documents and word, PowerPoint, PDF, they spent a lot of time already collaborating, building the perfect campaign, going back and forth on the information they need. But then when they want to take it in the Workfront as a system of record, they have to take bits and chunks out and put it in piece by piece, essentially creating some redundant work before they can even get their campaign off the ground. No one's excited about doing that, especially when you've put a lot of time and effort into it. So the way that we're solving this problem for our customer is we are making sure, they can use these documents that they've already created and then just drag and drop them into Workfront to get the initial project kicked off to get you where you need to be a lot faster. So how does that work? Well, it starts with generative requests.

Everything starts with an intake when we're working and trying to get work into a system. So we take the campaign brief that you've dropped into Workfront, we extract information that we recognize is important, we create a clean, crisp title around the information. We put information into the summary, into the other fields that you have, and make sure that everyone understands what the request is about, and the most important information is there without missing any key detail. But to make it even better, we can use the information that's inside of that brief to also search through your DAM. So if you're using AEM as a native connector inside of Workfront, search through that DAM and find assets that we can now recommend for reuse for you. So you can finally start to reuse some of your assets instead of creating things from scratch. And these are going to be attached to the request at the very beginning. So it's not going to be a problem that someone has to look for and hunt and solve midway through establishing the campaign. It's going to be right there. But what if you don't have something that fits the bill right in your DAM? Or what if you don't use AEM? That's where our Firefly APIs come into play. We can create some assets for you that are generated based on some of the information that you have in your brief and based on your brand, and then bring those assets, so you can attach those to the request as well. Again, making sure that you have a clean starting point, so that everyone doesn't have to start from the bottom floor.

We leverage a lot of those same capabilities with the drag and drop to establish our brief assistant. So we take the information from your creative briefs, we can start putting that information into the detailed forms you have inside of Workfront or recommend forms for you to use. And let's say for example, if you don't have the exact right form, let's say typically you don't do an international campaign, but now you are. We find the form that's very close, then we can even recommend additional fields for you to add because we can say, we've noticed that this campaign hits a particular Geo, you don't have a field for this. Would you like a field for the Geo? And then once you establish that field, we can also put in the new information, the summary into that body of text as well. Again, from the very beginning, taking your intake process, making that smoother, and then filling out the forms for you. So you can now start getting the project moving without doing a lot of that overhead work.

Once we get it into the brief, you want to get this work assigned to people. Today inside of Workfront, we have a flat list of all the people with the Workfront license. That's great, but it doesn't, it takes people a long time to find who they need sometimes. Especially, and the larger your organization is, you're more likely to have people who have multiple names, which makes it even harder to identify who is the right person for this particular task. So we wanted to solve this problem for our customers as well. What we were able to do is we're able to take contextual information about a user, whether that be as simple as their job role. So for instance, if you have two individuals and one's in legal and one's in creative, but they have the same name, you don't want to recommend or accidentally send an asset to the person legal for them to work on it. You want to make sure it goes to the creative. So we can bring in context like the person's role. We can see if they've worked on an asset before. We can see if they've worked on similar campaigns for similar markets, information such as that. And then give you a stat rank list of three to five people who are the most best case for you to send this particular project too. Again, relying on people's expertise and their talent, so not only this work start more quickly in Workfront, but you can get it to the people who are going to be best poised to do the job, getting the work out the door more quickly as well.

Once the work's been assigned, then it's going to move over to a particular individual who's going to do the work. What we've seen is that, once a creative finishes a task, they oftentimes move on to the next task because they don't want to sit idle, and they have plenty of work in their queue and then work that asset they created moves into a feedback and approval system. But sometimes a knowledge worker who receives it recognizes that there's something basic about the asset that's missing, something that always has to be there. To use an example, we have a lot of people who are in regulated industries. And as part of being in a regulated industry, they have to have clear bodies of texts at the very bottom of all their assets to make sure that they're legally protected. It's a compliance risk. But if that body of text isn't in there when the creative finishes it, they have to go back, tell that creative that they now need to come back to it, which usually results in the creative only coming back to it when they finish the new task that they've picked up. So you had a campaign, an asset that was this close to being done, and now it has to sit and wait. We don't want that to happen for our customers. So with our asset QC assistant, you can establish a set of rules and other types of information inside of a schema, and that's going to set the boundaries for the things that you absolutely need to have. Jason's going to tell you a lot more about that in a moment. But what you're going to see is that there are going to be key elements, whether it's your brand, trademark logos, bodies of text, et cetera, even key parts that have to be in a picture that you can identify as critical. And if those are missing, it lets the creative know right away that which items are missing so they don't have the context switch. They can see that feedback immediately, and then they can go ahead and correct the image themselves. And after they've corrected it, it then goes through the approver again. And now that it's passed, only then does it move to a knowledge worker. So no longer is a knowledge worker wasting their time looking at things and now having to wait for a creative to comeback to them. It really streamlines that flow and makes it more beneficial for everybody. Also, reducing frustration across those teams. So once we've-- Now have an asset that needs compliance and other types of standards, someone started to review it, we then move into our next innovation, which is with our firefly proofing. So we have a concept where, if you're-- Today in proof, you can lasso a particular item. You can put comments around it and say that this is an item that needs to be touched up, but people have always said it, a picture's worth a thousand words. And today, when people type that up, the language can be vague or misleading and allow people to be confused about what is the final goal. So we've decided what if we wanted to use our own capabilities with Firefly, it's generative response and instead of just lassoing it and putting words assigned to the things that need to be edited, we now actually take gender fill and allow the person who is sending it back for a quick change to be able to insert what it is that they are seeing or imagining as a need for that particular asset that needs to be reviewed or revisited. Then that way when it goes back to the creative, they know exactly what it is they need to do. Again, resulting in fewer frustrations, fewer back and forth cycles through emails, and allowing people to get work done right away. So when we think about these five items that we can do for our customers. Again, creating a value stream, that's going to help multiple people. At each stage of the content supply chain, we know we can start to helping you deliver content more quickly, more accurately, and with a lot less risk.

So I did mention earlier, we do have some screenshots, we do have some things we're going to show, and this is just a quick screenshot of how you can be in Workfront, and you can see the import from file to be able to start to bring in that brief. And once you click on import file, you can bring in the brief and start running through that intake process, I mentioned with the agenda of requests, and also to start being able to do some of that brief generation that we wanted to do. But now to really value, I would like to introduce you to my good friend and colleague, Jason Baron.

Thanks, Ben. And thank you for all of you being here. I know it's 4 o'clock, you've had a long day, and you're here at the top floor, the furthest end of the building you can get to. So thank you for being here. This is such a cool topic. Workfront is doing some amazing things with AI. It's a constant discussion. I'm in constant discussions, Ben is on how we leverage AI for Workfront, and there's just so many great areas that we can focus on. And so we're just, we're just thrilled to be able to show some of those to you today and also to gather your feedback as time goes on. On other ways we can optimize your flows and processes to make Workfront really truly work for you through AI. So today I'm going to be talking specifically about brand and in a previous life, I managed a team that included 10 creative directors for a large organization, and that, that team, they were responsible to make sure that there was alignment within the organization around brand itself. And so, as you can imagine, and I saw firsthand, alignment with brand can be a big challenge in an organization. And there's a lot of detail specifically around brand. So a lot of you probably have something similar where you have some kind of Brand Guide with the organization. And most likely it's digital, but you also have physical copies as well that are referenced. So this house is all your information about the guidelines associated with your logos and fonts, and color schemes, and application of certain design elements, as well as maybe your image guidelines associated with your brand. And so here we're showing the weekend brand. This is a fictitious brand where it's a company that is focused on the outdoors and products for the outdoors. And you probably have something that looks like this where-- If you zoom in here, there's image guidelines. So in this case, there needs to be natural lighting to utilize for authenticity and beauty as well as just natural beauty, leveraging the outdoors for your natural beauty and then showing active exploration. So rather than passive leisure, showing actual active exploration as people are in these images, actually demonstrating the use of these products when they're hiking and climbing and other activities, as well as diversity and authenticity. So featuring a range of explorers. And also inspiration, focusing on those visuals that will inspire action for people. So these are all very nuanced guides, guidelines, just for imagery itself.

But it's really important, as you all know, to be consistent with our brand. And the brand is strengthened through consistency and really the gap in that is where there might be human error along the way as the brand is evaluated.

So you saw the demo this morning or some information about it, about the new Gen Studio standalone application, which is super powerful. I'm going to show here, this is the weekend brand loaded into Gen Studio. And you can see there's information about the brand, the logos associated with the brand, the color scheme, some fonts. I'm going to focus today though, specifically on these sections right here around the images and then the tone of voice.

So again as I mentioned, when we looked at the image guidelines on the previous screen, those guidelines, that information can be extracted and then put into Gen Studio. And I will say Gen Studio is a manifestation of the brand guidelines right here. But the Adobe strategy is to have this information centralized, so that way it could be leveraged across other applications as well. So it won't just be within Gen Studio, it will be housed centrally. So that way you can access your brand guidelines not only within Gen Studio, but Adobe Express and other areas. And so you can see here though the manifestation of the brand, these guidelines. So again, the natural lighting, the color, the background, the composition, also some restrictions. So you want to avoid those staged photos or anything that has studio lighting...

and also with the copy guidelines, the tone of voice. What do we want the tone of voice to be, to be consistent? Also, what are our core values for our brand? So now when we think in light of those guidelines that you just saw housed centrally, how can we access those and leverage those guidelines within Workfront to save time and improve efficiency for you as our customers? So here we are. We're on the homepage of Workfront. You can see I have some approvals waiting for me. But up in the top right when I click that in the setup area, we're introducing a new area within the setup area.

On the bottom left-hand side, you can see under the Review and Approval tab, you can see there's the AI approval Assistants area. And you can see a list of AI Approval Assistants that I had already created previously associated with my brand.

And what I'm going to do is I'm going to add a new AI Assistant. But before I do that, also something you can see is you could have either active or inactive AI assistants. So as you're, maybe working on an AI Assistant, you don't want everyone be using it yet. So you can be building it out. Well, you can just turn off the activity status and when you're ready, you just flip the switch and turn that on so it can be leveraged.

So here I click Add New. And now I can give this AI Assistant a name. So in this case, I'm just going to call it Weekend Brand Standards. And right below that you can see I can add Brand DNA Guidelines. So I'm going to go ahead and do a type ahead weekend image you can see. And then you can see the weekend imagery standards, guidelines that are being pulled from that centralized location that can be brought into Workfront for evaluation.

So I'm going to select that one. But actually before I do that, you can also see there's product standards, color palettes, video standards. You could also have, for example, copy standards for voice and tone, also exploring, having the text be able to check for spelling errors or grammatical errors. So this could potentially be something where you have a one-to-many relationship. So you can see, maybe I've added some additional criteria that I want this specialized AI approval assistant to be trained on.

So now with this AI approval assistant, it's going to be checking for those imagery standards, for their logo guidelines, maybe to make sure that any photos that are used are licensed appropriately and so forth.

So now that it's been added, you can see it on the list.

I'm going to go into an actual project in the documents list and here I'm going to be able to see the document within the document summary. And as you're all familiar, I can add an approval to this actual asset. So I'm going to click the approval area here, and now I can give it a due date. I can associate a template to it if I'd like. But in this case, I'm just going to add the AI Approver that I just created. So here's the type ahead. You can see I have a series of AI approvers from that initial list. One of those is that Weekend Brand Imagery Standards AI Assistant that I had created. So I'm just going to select them just like I would any other individual that in my organization to do a review and approval.

So then once I'm good, I'm just going to go ahead and submit that request. And then what comes back is the response within Frame.io, where it shows the actual image that I had uploaded, as well as the response from the AI Approval Assistant. So here it provides actual context of the decision that it made. You can see on the top right saying it needs work and the why behind it needs work. So in this case, for this ad, I need to choose a different image because this doesn't really show active exploration. This shows more passive relaxation. So in that case, I have very specific feedback from the AI approvals assistant. I have a decision that the AI approvals assistant had made. And now as a creative, I can go back and change that actual image and then make sure that it aligns with the standards and then submit. So here I've uploaded a new image, and you can see this one did indeed pass the approval, as well as you can see the context of the decision that this meets the brand standards and is approved for use in this campaign.

So just a super powerful way to leverage AI with Workfront in Frame.io. And I'll also just put in a plug for a presentation tomorrow at 02:30 around Frame.io with the integration, with the new review and approval process and experience. I definitely encourage you all to attend that as well. It's called Revolutionizing the Content Production by Etienne Bosch and Caroline Ossmann.

But here I am, I have that. And now I can compare the two versions together just to kind of see what the actual approved asset was with the one that was rejected and make that comparison. And as we look at the approval process today, you have a request that's sent, the reviewers are notified, they sign in to review, they send their feedback, the feedback gets reviewed, and then the decision is made and the asset is ready to go or not ready to go if it's marked as needs work. On average across all customers, we see about an average of three days to make an approval in that case. But with the new AI approvals assistant, that's just done in seconds with more accuracy than you would get with an individual trying to review the asset in keeping in mind all of the different nuance guidelines that exist within your organization. And so we see this as a huge value add to organizations to be able to really unlock the power of the content supply chain and improve the efficiency that takes place within your organization through AI. So we're super excited about this.

I'm sure a question is going to be, when will this be released? We're focused on the second half of this year having this released and for, set of customers to gather feedback and then with a broader production release. And we're super excited about the progress we're making. It's moving along really rapidly, and we'd love to partner with you and gather your feedback as we go. So that's the AI Approvals Assistant within Workfront, unlocking the power of the content supply chain. - Back to you, Ben. - Awesome. Thank you so much, Jason. Yeah.

Really cool, right? It's not often that you can take something that used to take days and just make that just a matter of seconds and also increase your accuracy, right? So we're shortening time and increasing accuracy, helping protect you from compliance risk and other things. It's just really fascinating what we're able to do. And I'm really excited to be part of Workfront and partner with my friend Jason here to be able to see this come to fruition.

So when we talk about AI, we spoke about our value stream and how we built things up and down to mimic our own content supply chain. So we have another layer to it as well. I'm going to introduce in a second our AI assistant, but I want to talk about our philosophy of how we're going to continue to help empower people. And again, that power part of our North Star statement and also making work easier. We look at work as AI is an orchestra, and the person, the knowledge worker is the conductor. You can have the most beautiful instruments in the world. And if they start playing their own song together, they're going to sound very, very poor together. You need a conductor to be there to instrument, to tell things how to fit together and that's how we imagine employees of our customers using our services to get their work done, to make something beautiful happen out of what would otherwise just be noise and miss.

So we have multiple capabilities that we're introducing with our AI assistant. I've got some screenshots I'll show you in a moment. But we've looked at some key use cases that have just come back from multiple areas of feedback that our customers continue to give us. We have lots of customers who are either external agencies or they operate as their own internal agency within their organization. And they'll come to us, and they'll say, Ben, we love Workfront, but the people that we put Workfront in the hands of, they don't know how to use it, or they don't know what they want when they're trying to build a campaign. They don't know what they want when they're trying to put the request together. So our first request comes in and it's noise, and we have to have a lot of meetings to suss out what do they really need, what do they want, what goals are they trying to achieve? How can you help us? One of those methodologies that we're going to choose to help our customers solve that problem. Again, whether you're an internal or external agency or just someone who works with clients in general, is to allow our AI assistant to work with people who are trying to go through intake, to ask them questions about what they need. Looking at the forms and information that you have inside of Workfront is your system of record to generate questions to prompt to a user. So when that user is trying to fill out a request or create a project that they don't know exactly what they want, Workfront will ask them the right questions. So you start getting the right information in, so you don't have noise coming in. And again, you reduce painful cycles and frustrations. Going through meetings, just trying to figure out what do people actually want? What is their goal? You can start moving on to the more strategic stuff about how do you achieve that work? How do you achieve that goal? Next is making Workfront easier to setup and scale for users. Sometimes people want to set Workfront up for the first time. Sometimes they want to expand it to different parts of their organization. Sometimes they just want to open up for a new use case within the same department or same group, but they have to create a new structure inside of Workfront. We have a solution for that in Workfront today. It's called Blueprints. There are out of the box templates that we've built based on thousands of consulting hours where you can piece together and find a use case and skip Workfront set up to about 80% of where you need to be, 80% is pretty good. We save a lot of time for our customers. A lot of people are very happy with Blueprints. They deploy on the move Workfront rapidly, that's great, but it wasn't good enough for us. What we found is we can take that 80% and turn the knot dial closer to 95% by using AI to converse with an individual to ask what you need. So if you're trying to set up a dashboard, if you're trying to set up some sort of reporting infrastructure or just a new process inside of Workfront, you can use essentially the basis that we would've had within a blueprint. We can still get you there. But then engage and ask more questions. Is there anything else you need? Do you use a different terminology for your projects, tasks, and issues? How do you refer the work? How many structures do you have? Et cetera We can ask these questions and help people dial it in, so they have something that's a lot more appropriate and better mimics what they want. Again, reducing the amount of time that people have to spend manual hours to get something done, and ultimately work needs to happen, not work to get work done.

We also want to have conversational report generation. What we found is a lot of people use Workfront, yes, to bring in work in to track it. But as I mentioned about making Workfront more powerful, people like to report on it too. They use Workfront to answer lots of questions, get visibility in the progress, understand how things are flowing from point A to point B. Sometimes they answer that information with reports. They build reports for two reasons. One, they have a simple question, and they want a simple answer. Two, they have a stakeholder, and they want to be able to show this information to others. We're going to use conversational AI to help solve both those problems and make it easier. Let's start with the first use case, where you're just an individual. You have a very basic question, which are, what five projects are the most overdue? You might build a report today to get that done. You're going to leave what you were doing in Workfront, build a report to answer that question and take a lot of time out of your day. Wouldn't it be nicer if you could just ask an AI Assistant and it's running in the background for a minute or two to give you that answer, while you finished doing the work that you already had in front of you, while you stayed in context and present with your work? Instead of having to build a report just for a simple answer to a question, that's what we're going to do with our conversational AI. You'll be able to ask these questions within the assistants and get an answer back. So you can have some of those simple answers. But let's go to the other use case. You do need to build a report because you have stakeholders or other people who want to see it. That's fantastic. You can type into the assistant saying that you need to build a report that shows the following capabilities or following types of information, and then it will start building that report for you in the background. Again, you're staying present with the work that you have because you've typed it into the assistant. You can continue to edit some of your projects or your tasks, move work around, participate in comments, and then when this report, this foundation of report has been created for you, the assistant's going to output a link for you to be able to click on and then get to that report, so that you can now go and either share it immediately that has all the information you need or make whatever tweaks you need to make. But either way, it's saving you time and making your life easier.

We also want to make AI front and center for people who haven't historically had the best time in Workfront. Specifically, that's marketers and creatives, who get in Workfront, they're not a planner, they don't need all the complicated areas of Workfront. They need something that's easy to use, and today they have a lot of overhead because of the way that Workfront's structured. They have to navigate to many different places just to get all the proper context that they need to get their job done or to start a particular task. We're creating a special place inside of Workfront, that's going to be a landing page for marketers and creatives, and then we're going to make AI front and center to help them get the information they need to help move through their tasks or identify their tasks more readily and more easily. Again, that also goes to the philosophy that we had earlier about making Workfront more accessible for people and also making sure it's more powerful. Because the more you have accessible for people, the more inclined they are going to use it. The fewer change management items you're going to have to go through. You're going to start getting cleaner and more data in, changes and tasks will be done appropriately, and work will be more visual on that backend as well. So this is how we're really going to start to revolutionize different ways that people interact with Workfront every single day. Again, going to our philosophy about a customer-oriented vision, we want to make sure that we are helping people regardless of what their role is, how they're using Workfront, making sure it's easier for everybody.

Now, before I go into the screenshots about the Assistant and walk you through a few use cases, I want to talk about what's really unique to how we have structured Workfront and we're structuring our AI assistant. At the top layer we have a micro frontend that gives you your UI where a user can engage and write the prompts and get information back. That's great. But we've also created a brain that's one layer underneath that, that acts as a router for information so that we can do a lot of really cool things that help organizations be more successful. So what you'll notice is we have these things layered as experts underneath. These experts are specific AI solutions that we build to handle a particular task inside of Workfront and also make sure that Workfront can be hyper contextually aware of what a user's doing to give them the best answer. So if a prompt comes in, it goes to the frontend, it goes through the router to the right expert or a series of experts. We get a response from the information from those experts, we then look at it and sort of combine it inside of the brain and we then output the best, most accurate answer we can for a customer. So we're increasing the quality of our responses. Since we have these micro LLMs that we're working with, it actually speeds up the time of response as well. But there's some other unique things that also come into play. Number one, since we have these different experts, that means we can use multiple LLMs, and different large language models are better suited for certain tasks. So you have some large language models that might be very good at one thing and very bad at another thing. That means that we can always route to what's the best LLM for a particular task or for a particular need. Again, making sure we have that accurate information. It also gives our customers flexibility. It means that you don't have to have our assistant either enabled or disabled like a binary state. If there are certain things that your organization doesn't have approved through legal or compliance or just doesn't feel comfortable with, you can turn off some of these capabilities individually and still get a lot of the value. In some scenarios, individuals either have to adopt an assistant because it's built on just one LLM, or they have to not adopt it because they're not approved for that. We offer a degree of flexibility, and I'd say the next thing is we have an enhanced degree of security. We value security at Adobe, and we know that your information is important to you. So as you start typing in, we don't pass information from one assistant to another, and your information is never going to merge or mesh with another customer's information. We could have all that controlled by the way that we've designed our AI Assistant.

So a few things, like you can see the AI assistant, and this is a real snapshot of work that we have. You can see the AI assistant right there on the right-hand side for a user who's working in Workfront. This user happens to be a little bit new to the Workfront. They don't know all the good things that they can do. So they've asked, how can I create a request queue, that's the basic thing. So we give multiple types of ways to help this user. One, we give them information, sort of a how to guide of what to do. This helps our admins 'cause people are no longer going to admins. We give them some basic information, some links they can click on. But since we have those experts and we're contextually aware of what the user has on their screen, we know that they're in the fresco, copa bean voyage project. We ask, we give them a quick action for, do you want to just go ahead and create a request queue for Frescopa Bean Voyage? Yes, we taught them how to do with the text. But now we're just giving them one button away from being able to do that right away, instead of them having to navigate and figure out how to apply, what we just wrote out for them and summarize for them. Or if there's an existing queue, we can provide a link and take them directly to that queue. Lots of options to help the user know what to do next, without them having to think about it and move on. So in this scenario, not only do we provide the information, but the user has said they want to create a request queue. So much like I mentioned with the reports where once something is done, a link will be provided to it within the assistant. Not only has the request queue been created, but the user now has a link to it that they can directly access and go look at that request queue. Again, something that previously they would have to, if they didn't know how to create a request queue and that's something they need to get done, they would have to go ask an admin, maybe search on experience league online, and it would've taken them much longer to get this done and they may not have even done it correctly. Now, just a few words in typing away, they've been able to get this done for their organization and again, enable work to start, actually happen.

Here, we'll step through some of the additional typing that you can do. But what I wanted to point out first is, we saw some of those quick actions that were generated based on the responses that the individual had or their interactions with the AI Assistant. We also have other quick items that individuals can immediately do, prompting users with things that they might need to get done. Let's say it's the end of the week. Perhaps they need to start logging some time or they have approvals outstanding. We can present some of this information to the user and help them enable themselves to get work done more quickly. But for this scenario, we'll type in a new thing for the user to show you how we can actually make work happen. The user is saying that they want to actually create a new multistage approval workflow, something basic, something they've typed. Well, we have the context of where they are in this particular campaign. So once the user types it, we've created a multistage approval for the user via the assistant for the proper campaign. So the user didn't have to type in information about the campaign, do a lot of the initial legwork that would take place to get that done. AI took care of a lot of that for them. And based on the information that it saw in the brief and the summary information from that project, it initially created three areas in this multistage approval. But in this particular situation, the user said, I'm going to need a final review. That probably wasn't clear in the project, but I'm going to need a final review. And they added a fourth step. The AI, since we have those experts running the background to keep contextual awareness, recognizes that and even sends a message back to the user that says, you've added a fourth stage via the UI. So this is just an example of how we can leverage the AI Assistant type in things that you historically would've had to do manually, or assuming that you knew how to do it. If you knew how to do it, you have to do it manually. You can type it into the AI Assistant, get a lot of that work done. Still you as a conductor can shift things around and make the music happen as you see fit. And our AI Assistant stays aware. It follows you, much like the orchestra falls, the conductor. It falls along with you, so that you can continue to make magic happen together.

On this one you can see this. You can see this new space that we're seeing that we had created for our creatives and marketers. This is the new landing page that we're going to start using for them instead of Workfront. And it has all the information that's contextual for a particular project right then and there. But we also want to put information and power at the fingertips for these users via the AI system. So the AI assistant with prompt users in this space to do several things, summarize unread information or documents, or what should I follow up on. Let's say you've been on PTO for a while and there's been a lot of noise and information in the comments, and you've come back, and you don't want to read every comment. You can ask, what should I follow up on? Or can you summarize the documents, or what actions do I need to take first to accomplish this project? Again, making Workfront to be more accessible, easier, more powerful, and not just doing it for people who are planners, but for also people who are creatives and marketers, so they can be more engaged and also find Workfront more delightful for themselves to use and have all the information they need at their fingertips without having to navigate and do a lot of complicated work on their own end. So we're really passionate about, from top to bottom, whether it be creating value streams that mimic our own content supply chain, and helping people through a single workflow or through introducing assistants that are going to be able to help guide you and help you get work done all along the way. These are the things that we're doing. We'll leave you with one last bit. You also probably saw today the keynote that Adobe is partnering with Microsoft to do some interesting things with AI and copilot. We're going to start connecting some of our capabilities that you see here today to copilot, to be able to introduce what Workfront can do across all those different surfaces. So now not only will you be able to accomplish a lot of this in Workfront, but our goal is to be able to help you do it elsewhere as well. So please do stay tuned. As that partnership continues to grow, you'll continue to see more information and more news about that. And the last thing I'll say is, what you saw with the assistant is actually in Alpha right now. And we will plan to move this into a beta stage as we get towards the early part of the summer. So if you have a CSM or a salesperson that you work with, please let them know. We'd be more than happy to get you onboard for either that Alpha or that Beta.

And yeah, thank you all very much.

[Music]