Transforming Knowledge Work with AI with Lindsay Sterrett at OpenText World 2024

This is Textron tv. Hello folks. We're back at Open Text World in Las Vegas, and we're talking about reinventing the knowledge worker with my friend Lindsay.

Lindsay, welcome to Show. Thank you so much. You know, I read this study just the other day and it surmised that the number of management positions was increasing, and it suggested that the reason for this was because more of these AI agents needed to be managed.

And so people are starting to reinvent some workflows. And I wonder if the term knowledge worker itself is becoming obsolete because we're all becoming maybe knowledge managers. That's actually a very interesting way to put it.

We haven't thought about it that way, but something that we're talking about this week is about the use of AI to elevate or reinvent the knowledge worker, right? That, um, instead of our best and most talented people in the company spending their time managing information, um, you know, trying to work through these manual arduous tasks, right? That there is an opportunity for knowledge workers to leverage AI to put the information to work for them.

Mm-hmm. Right. Um, so in, in the sense of, um, you know, what you said is, what was the term that you used?

Knowledge management. Uh, Knowledge managers. Knowledge managers, right.

Um, I think it's really about, um, enabling knowledge workers or knowledge management to leverage information in more powerful ways. Right? I think also one of the dirty little secrets about this business in general is that there are people who are specialists, they have a lot of knowledge, and they're working on their own projects, but everybody knows that they exist and they keep asking them questions that interrupt what they're trying to do because they're trying to be helpful with other folks.

Will it become easier to kind of distribute knowledge in a way that doesn't always require a human to sit down with somebody to do that? Absolutely. And, and maybe we can consume all this stuff in a better, more efficient fashion.

That's Exactly right. Um, the whole premise of Aviator, um, which is something that we've been talking about this week, is the opportunity to leverage interactive chat interfaces, natural language questions, to interact with information in entirely different ways, right? So when we think about really document heavy processes or aspects of the business where in the past somebody may have had to manually comb through hundreds of pages of documents, imagine something like that being as easy as just typing a question into a chat to get a nice consumable summary or an analysis of a piece of information, right?

That's just the beginning though. How are we gonna manage all these AI agents that are working on our behalf? I know you have like 15 of them now, or what you call, um, The aviator Aviators, um, which are actually super sets of a hundred or more agents themselves.

Um, will we as humans be able to manage that workforce? And It's a really interesting topic and something that we've been excited to talk to our customers about. Um, many of the customers that we've spoken to this week are at different stages of their own AI journey, and whether that's at the start of kind of ideating the strategy, upskilling the workforce, or even going, you know, full through to adoption.

And, um, a big topic that we've had with some of those customers this week has been the topic of upskill in particular, right? Because the lack of, um, the right skills or the resources is often what's holding organizations behind from being able to take that next leap into ai. Um, so to your question, I think that there is, um, a lot of learning and we should all be ready to embrace new skills, and that's what's gonna help us take it to the next level.

I'm not sure looking at the agents what skills I need to learn because it seems like the agents are pretty, uh, simple to use. They're kind of, um, self-evident, and I'm not, and I didn't feel like I had to become a prompt engineering expert anymore if I look at these agents. So every, we kind of moved beyond prompt engineering now and we are gone to a higher level.

I think you're right. Um, it's been very much about, um, simplifying the knowledge work or that end user experience versus more of the technical, um, things that are happening behind the scenes. I'm a marketer, right?

So I'm, I don't claim to understand all the, the deep things that are happening underneath the hood, but it is that simple in some cases. And, um, I think that again, it's, um, there's an opportunity for all of us to be thinking about what the next jobs are that these AI tools or these, um, interactive chat interfaces are gonna unlock for us. Is it your sense that the folks that we used to call knowledge workers, um, are they excited about this or are they, is there a little fear in, in trepidation in, in the mix as well?

And how do they kind of navigate that? That's A great question too. Um, I think that, uh, there's an obviously an opportunity for AI to eliminate a lot of, again, those like manual tasks and that like, you know, the, the arduous work again, right?

That things that are holding us back from being able to do more meaningful work and make those strategic decisions. Um, but I don't think that we're at a point where AI is going to replace knowledge workers. It's really about taking them, allowing them to, to be elevated within the organization.

There's an old saying that says, if you do something enjoy, you never work the day in your life. Are we getting to that point now where we can focus on the stuff that's fun and all the other stuff that gets in the way will just be automated? That's the vision, right?

I think that's, that's absolutely the vision. So that again, those most talented people in your organization are no longer bogged down with managing information, but that they can focus on doing, um, things that are going to deliver, you know, it's innovating new products or it's, you know, expanding meeting new customers or expediting medical care, things like that, that are much more impactful. Mm-Hmm.

Is there gonna be large scale re-engineering of some of the workflows that knowledge workers are tied to? Because it seems to me that I'm gonna have all these new capabilities and maybe doing things the same old way might not make as much sense. So are we gonna take a moment to kinda rethink all those workflows?

I don't know if we're rethinking the workflows as much as sort of re-engineering the way that they happen, right? So again, um, one of the awesome examples that we showed in one of our sessions this week was taking something as simple as a sales contract approval, right? And being able to use an interactive chat interface and natural language questions to actually prompt that workflow and then automate some of the things that are happening on the backend, such as when, you know, in this particular example, the document needed a certain sales director approval, right?

Once it was approved by that person, the AI is actually then taking it to the next step of being able to PDF and water market, right? So the step itself or the workflow itself is very much the same. It's just how it's actually happening.

That's that much easier for the knowledge worker. Do you think maybe eventually we might restructure organizations in the age of ai because a lot of the tasks are created and roles assigned to tasks are based on the fact that some thing had to be manually done. Well, if it doesn't have to be manually done anymore, can we maybe take a step back and say, Hey, what is your job description and, and what might that look like?

Yeah, I think you're absolutely right. Again, it's, um, not so much about replacing jobs, but freeing up knowledge workers for the next job, right? And that's, that's truly the reinvention of knowledge work.

What are you hearing from customers? What are they struggling with in this whole conversation? I mean, you've been walking around the show floor here.

What's the feedback you got so far? So again, um, we have customers at all sorts of stages of the journey. And the things that we're finding are most common from a, a challenges or limitations to getting started perspective is, um, again, the, the lack of, uh, sort of a formulated foundational strategy, um, having the right skills or the resources to actually implement that strategy.

But also, and perhaps most importantly, um, and why OpenText becomes so critical to this conversation is that data readiness perspective, right? And so where we come from in the content management, the document management business, um, and our customers who've been with us for a long time and they've, they've trusted us with their unstructured information, um, they're coming from a standpoint of great advantage because they've already invested a lot of time, um, and effort into getting the information prepared, metadata and, and all those things that makes gen AI that much more impactful. One of the things I hear people kind of struggling with a little bit is that, um, the responses from the AI agents are probabilistic and they might not be the same over and over again.

And if you're a knowledge worker, you're kinda like, you know, driving a process that's a little more deterministic. So the fact that the thing tells you something that's similar but in a different way kind of drives you crazy. Again, that's where, um, I think OpenText customers have an advantage where, um, we're through the concept of something that we call the business workspace.

You're curating information that's all related to each other, and by virtue of asking questions against documents and data that all are, are, you know, related to a certain use case or, um, what have you, that the results that you get back are that much more relevant and accurate. Hmm. How do you think this will play out when I, as a knowledge worker will have my small army of AI agents and you as a knowledge worker will have your small army of AI agents and somehow or other we're all gonna collaborate together and to do something.

How does that process work in your mind? Because everybody's gonna have all these different AI agents? That's a another excellent question.

Um, I think that that's something that we're still trying to explore right now. And one of the things that we're hearing a lot from customers is, you know, how can we anticipate that all of these, um, AI agents, as you say, how will they interact? How will they interoperate, right?

And so we have some really excellent partnerships with organizations like SAP, Salesforce and Microsoft that we're really going to build up, build upon and leverage to make sure that we can, um, deliver impactful solutions to our customers that allow for that AI to ai um, integration. Do you think I'm gonna have like one kind of senior level AI agent that manages all the other AI agents? Is this gonna be like Downton Abbey where there's a head butler Perhaps?

And from my point of view, content aviator could really be that head ai, uh, butler, if you will. Um, something that we actually showed this week in fact was, um, content aviator, sort of the primary, uh, agent that you're interacting with, but um, in the backend it also has its own, uh, uh, tool or, or agent also, which was knowledge discovery, which is another really powerful tool, um, or a solution within our portfolio that does a lot of deep, um, rich media analysis and things like that. So it is certainly possible.

Can I ask you a question? Sure. All right.

So my favorite question to ask all of the customers this week was, um, okay, imagine every knowledge worker had an intelligent assistant. Um, what if you had your own personal intelligent assistant, what kind of impact would that have on your life and your work? I think it would make me a lot more efficient, but I'm not sure I would use that time wisely to do something fun and leisurely.

I would probably just do more work 'cause it's kind of how I'm wired. Um, so, um, but then again, my family might figure that out and they might be saying, you know, you don't have to spend as much time working on the weekends so we can go do this, this, and this. So I'm sure that, for example, my wife will probably have an AI agent somewhere that figures that out for her and she'll just start scheduling stuff and she'll say, devil be damned.

You're gone. A scheduling agent would be the best thing for all parents, wouldn't it? Yeah, That would be kind of cool.

Well, the kids will have one too, so then the kids will start scheduling things for parents too, so things could get a little kind of topsy-turvy and upside down. Absolutely. Let me flip that question back to you.

You're clearly on the front end of using this technology. How has it changed your job, your daily life? What do you know what's different today than from a year ago?

So I'm a marketer again. Um, we create a lot of content and one of the most immediate benefits that I think we're gonna see in marketing specifically is the ability to, you know, again, think about, um, we've managed a diverse portfolio of products and every time we launch something, there's new content that has to be created for those products. What if we could take, um, you know, release notes and things like this that are coming out of our product management team and leverage AI to translate that into marketing content, right?

Content creation and content generation is an super impactful productivity gain that we're gonna see the benefits benefits from very quickly. We have our own sales team and we're constantly now yelling at them going, Hey, why don't you just create your own marketing collateral? Exactly.

Exactly. It's a huge enablement tool. And even when we think about emails and, you know, maybe I can get some creative inspiration and, you know, help get a, an, uh, a content aviator to write this email for me, right?

Learn, learn to do it in my tone and my style, right? And I think the tone and the style is a, a, an important part of this equation. 'cause in fact, we were just having this conversation with, uh, Alan Shimmel who runs Textron, and I'm like, I see this content, but it needs more personality.

It needs you in that story and it needs to come through there. So, um, I wonder if we'll go through a phase here where everybody will be like just auto generating some content and letting the machine do it not, but not adding that human element that makes a difference. And that's where again, we can, it's a starting point and we, with our, you know, again, our skills and the knowledge that we have about the business can then come in and fine tune the result.

It's all about just those quick productivity gains initially, I think. Alright, so what's your prediction? 2025?

How fast is all this gonna come? I mean, is is that knowledge worker gonna be reinvented next year or the year after? How Fast it's a journey and things are happening so fast.

Um, it's hard to predict, but I think that we'll absolutely see from OpenText at least. Um, you know, once we get the, the, the phase that we're in right now again are like those quick productivity gains, automating, um, you know, some of these like, you know, common things like, you know, making information easier to find, right? That's an easy thing for us to do.

Um, breaking down content silos, creating or generating content, um, I would say by the end of next year that we'll certainly have made some impactful advancements on the workflow automation part. That's something that's gonna be really exciting to talk about next year. All right.

And last question. What do you know now that you kind of wish you knew a year ago about all this? Oh, tough question for the last question.

Um, I think that it's, you know, we're still, um, our customers are still on the journey, right? So one year later we've got, you know, every conversation that we're having is about ai and that's super exciting. Um, we have customers that are buying it and trying it.

Um, will we be at a point where everybody's using content aviator by next year? I hope so. Um, but you know, I think because of a lot of the unknowns and how rapidly the, the market is changing, um, the technology behind it is, is evolving that, um, you know, we're, it's still, we're still learning as we go, right?

All right, folks. Hey, you heard in here, instead of being a knowledge worker, maybe you become a knowledge manager and give yourself a raise. Hey Lindsay, thanks for being on The show.

Thanks so much for having me. All right. And we'll be back soon.