UiPath Forward 2024 | Bobby Patrick, UiPath & Craig Le Clair, Forrester Research

Clips
News
More from UiPath Forward 2024

Bobby Patrick

CMO

UiPath

Craig Le Clair

VP, Principal Analyst

Forrester Research

AI workforce automation: Balancing innovation with workforce challenges

Artificial intelligence is at the forefront of shaking up industries, with AI workforce automation poised to transform how businesses operate.While the promise of increased efficiency and innovation is clear, the challenges that come with integrating these technologies — such as ensuring governance, security and minimizing workforce disruptions — are just as significant, according to Craig Le Clair (pictured, right), vice president and principal analyst at Forrester Research Inc. As AI becomes more central to mission-critical tasks, companies are focusing on how to build robust automation systems that are resilient, reliable and trusted

play_circle_outline Exploring the Impact of Technology on Workforce and Job Roles with Craig Le Clair's Book

play_circle_outline Discussion on governance, security, and trust in automation

play_circle_outline UiPath's approach as a central orchestrator of AI agents

play_circle_outline Optimistic paths for middle and service workers in the age of automation

play_circle_outline Focus on AI-native companies disrupting industries with cost-effective solutions

play_circle_outline Market growth and acceleration of agentic automation compared to RPA industry

Info
Transcript

Bobby Patrick, UiPath & Craig Le Clair, Forrester Research

Bobby Patrick

CMO UiPath

Craig Le Clair

VP, Principal Analyst Forrester Research

TheCUBE's live coverage of UiPath FORWARD 2024 features Dave Vellante, co-host and analyst, along with Craig Le Clair, VP analyst at Forrester, and Bobby Patrick, CMO of UiPath. Le Clair discusses his book, "Random Acts of Automation," focusing on technology's impact on the workforce. Patrick stresses governance, security, and trust in automation. They explore AI's role in enhancing human action and challenges of maintaining purpose-based agents. AI's potential to transform industries, value of operations-focused agents, and need for compliance and security a... Read more

explore Keep Exploring

What is the purpose of the book that is being discussed in the text? add

What are some of the key topics of discussion and focus at the FORWARD event? add

What problem does the speaker believe the industry can solve with the help of an abstraction layer created by AI interfaces? add

What is the optimistic path for middle-skilled workers using AI in the future job market? add

What are some key factors that are often underestimated or overlooked when it comes to AI native companies and their impact on industries? add

What is the significance of being with Craig in relation to putting RPA on the map and the growth of the agentic automation market? add

bolt Powered by CUBE AI

Bobby Patrick, UiPath & Craig Le Clair, Forrester Research

search

Rebecca Knight

>> Good afternoon everyone. Welcome back. We are here in day two of theCUBE's live coverage of UiPath FORWARD 2024. I'm your host, Rebecca Knight, alongside my co-host and analyst, Dave Vellante.

Dave Vellante

>> Thing is there's so much content going on, we have to miss some of the keynotes to be doing CUBE interviews.

Rebecca Knight

>> I know.

Dave Vellante

>> We're going deep on some of these CUBE interviews though. It's been good.

Rebecca Knight

>> Luckily, we have some of the stars of the keynote stage joining us here. I'd like to welcome Craig Le Clair. He is the VP, principal analyst at Forrester. Thank you so much for coming on.

Craig Le Clair

>> Great to be here. Thanks for my return trip to CUBE. That's great.

Rebecca Knight

>> Indeed. A CUBE alum. And another illustrious CUBE alum, Bobby Patrick. He's of course the CMO of UiPath.

Bobby Patrick

>> I always enjoy being here.

Rebecca Knight

>> Thank you so much. So Craig, I'm going to start with you because we got to do a quick plug of your new book, which is coming out in January. It is called Random Acts of Automation. Congratulations. Not your first book, I should mention. Tell our viewers a little bit about the thrust, the idea that you're putting forward here.

Craig Le Clair

>> Yeah, so the basic idea is that all this technology we're talking about here, whether it's agents or agentic, is really going to affect the workforce. And a lot of the books about this are written for the business community or they're written for the tech people like me and Bobby and you, this one's written for workers. The idea is not to try to forecast how many jobs will be lost. I think we've realized that we can't do that, but really to look at the shifts in the workforce that are going to occur. So you might be a middle worker and worried about becoming a gig driver because you see what's happening with the agents that are doing your work. You might be a professional that's worried that your credentials are not going to matter that much with LLMs and AI bots that are doing it. And you might be a service worker that's wondering what these android robots are doing coming into your workplace. So this is a blueprint for workers so they can navigate this path towards really human-capable automation is the term I use in the book, and hopefully will be read by workers and help them. So if you're a mom and your son's been in the basement for six months playing video games, go out and get this and give it to them. Say, "Get out"-

Rebecca Knight

>> And read it yourself.

Craig Le Clair

>> Read it yourself. We all need it.

Rebecca Knight

>> Indeed.

Craig Le Clair

>> We're all going to have to deal with this.

Rebecca Knight

>> Indeed.

Craig Le Clair

>> But thank you very much. That's unexpected pumping of the book.

Dave Vellante

>> Yeah, good luck with that.

Craig Le Clair

>> Thanks.

Dave Vellante

>> It's a big effort to write a book and a lot of time and energy goes into it. So hope it takes off for you. Bobby, you gave takeaways from day one, which we really appreciated. Rebecca and I unpacked them very briefly, but while we have the author of those here, totally agree, single agents are kind of boring. Co-pilots, Benioff calls it Clippy. I thought that was very funny.

Bobby Patrick

>> Clippy 2.0. That was good.

Dave Vellante

>> And agents can't be... That's kind of new sort of thinking in the marketplace. Don't trust the agents, don't give the LLMs your credentials. The bots can be trusted. So I want to test that with you and that's where the guardrails come in. And then you're going after what we call the agent control framework, and you're putting a big invest... That's a high value piece of real estate there. So how is that resonating with the customers that you've talked to here?

Bobby Patrick

>> Yeah, there's two things at FORWARD which are pretty interesting and resonating that I... Well, one is the amount of discussion on governance and security and trust. The fact that we've had this woven throughout our entire event really, all our keynotes, it's sort of unmatched. Most people, Benioff, many of these guys, are just talking about these little agents and they have no governance, no control, and I think we bring a level of seriousness to it, which we brought in the world of automation. If you look at RPA, the areas that robots are used are mission-critical, they're business critical today, right? They've got to be highly resilient, highly reliable, very trustworthy. And so we believe we have a unique opportunity here to actually be at the center of the orchestration of agents that are LLM-based, robots that are highly resilient and precision-oriented, people because it's all about the workers, and making them more creative and more productive, and then models. And then we added another level of complexity to this and say, look, we're also going to be very open as a framework for the company. You can bring your own agents, you can bring Benioff's Salesforce agent for this siloed CRM stack if you need it for a certain use case, and we can orchestrate that. You can bring your own model. We announced several partnerships here with Inflection AI, Anthropic. We also have H Company and OpenAI and others. And so we're going to be the Switzerland, I think, and that ability to orchestrate the reality of the complexity that our customers know is what's needed to deliver AI in real work and real complex work. So it's resonating very well. I was worried because this is a big move and I couldn't be more thrilled by the excitement.

Dave Vellante

>> So I'd love to test some of that with the independent analyst. Craig, you said that you used to work for the dark side, but then came to the analyst... And then you, I think honestly and transparently, said, "That's not always so pretty either." So we know that well. But UiPath is essentially making the case that you really can't do effective agentic without automations/RPA is what my inference is, do you buy that?

Craig Le Clair

>> I do. In other words, it's pretty simple to me. A lot of the AI is helping with the human intent understanding. It's helping to reduce the friction between the applications and so forth, but without taking that human intent and doing something with it, getting something done, it becomes kind of sterile. So a lot of my analyst colleagues say things like, "If these process platforms, the mature ones, if they don't build AI into their platforms, they're gone." And I kind of say, "Well, you know what? AI needs the action component as much as the action component needs AI. It's a nice collaboration." And I'll build a little bit on a saying over here Patrick was saying, that in the early days of the RPA, when I covered it in 2016, there was so much talk about these credentials that these bots had, and we're going to have to put them in a vault because if a bot was created and had access to those credentials, they're going into the banking system. So the whole metaphor of having a trail of when the bot was developed, how it was onboarded into the company, how it was off boarded, understanding whoever touched it and versioned it, that's the kind of thing that these guys have been working with for a long time. And a lot of folks coming into this have no clue.

Dave Vellante

>> So yeah, I earned that trust. Bob, did you have the 8 billion quote?

Rebecca Knight

>> I do.

Dave Vellante

>> Because I have a question, and I'm going to pick up on something that Bobby said about the Salesforce, I think you called it stovepipe or silo. What's the 8 billion quote you have?

Rebecca Knight

>> Up on the main stage, you put forward a utopian vision. Over the next 5, 10, 15 years, 8 billion people will say goodbye to a complex web of mobile apps, sites and databases as AI dissolves the friction between humans and technology. Please, I lost that.

Dave Vellante

>> Okay. We all do.

Rebecca Knight

>> .

Dave Vellante

>> But that would fly in the face of decades and decades of IT industry and application history. So here's what I want to ask both of you. You correctly, I think alluded to, look, Salesforce is going to do agentic and ServiceNow is going to do it, and Oracle is going to do it and Palantir and on and on and on. And your strategy is to, like you say, be that Switzerland across. Do you think, and they're going to have some advantages within their own stacks,

Craig Le Clair

>> Yeah. Their domain, they're going to be strong.

Dave Vellante

>> Do you think it's feasible that the industry will be able to create an abstraction, the industry UiPath, specifically create an abstraction layer and solve this problem that Rebecca and I and everybody so wants to solve?

Craig Le Clair

>> Well, the abstraction layer has been created by the AI interfaces. I mean, I talked on stage about turning piles of data into a conversation. You couldn't do that before. The insurance policies that we have tortured by, we both have places in Florida, by the way, totally tortured by.

Rebecca Knight

>> Oh yeah.

Craig Le Clair

>> They're as transparent as concrete. They're written by lawyers for other lawyers to read. They mean nothing almost to the average homeowner. What if you could ask questions. If I should pass away, how much will my heirs get? How much have I paid for this policy over the last 10 years? That's the friction that is going to be gone. So everyone's going to be able to do that. And any of the more general platforms for co-pilots and gen AI as well as the more specific platforms. And I think the specific part over here is dealing with these long-run mission-critical processes that have to integrate with all the old technology and do that well. They're long-running. They have all kinds of issues, the conversational tools, just never had to deal with.

Dave Vellante

>> Yes, please.

Rebecca Knight

>> A lot of those companies have a vested interest in keeping all that stuff exceedingly complex and opaque to the average homeowner. So I'm curious how that's going to work too.

Craig Le Clair

>> Well, if you're an insurance company from that side, you have an advantage. You have a customer experience uplift if you can provide that conversation with your data.

Rebecca Knight

>> That's a good point.

Craig Le Clair

>> And they'll do that because they'll be pushed to it. Now, insurance is a bad example because they're not the most competitive of areas.

Rebecca Knight

>> No, we all know it.

Craig Le Clair

>> But you'll see that across a lot of industries.

Dave Vellante

>> So when I game it out and think about this Bobby, you have to have the back-end connections into those enterprise apps because that's where all the business logic lives, all the data and the metadata.

Bobby Patrick

>> Or UI.

Dave Vellante

>> Okay, but that's your connection, right?

Bobby Patrick

>> Well, UI automation was the original, but then the API automation, it's either/or, right? We've already built an extraction layer. I mean, that's for deterministic. So in the world of UiPath today, there's API connections, so millions of apps and then the UI, you could do anything that a human worker could do. Now, AI adds a new level of abstraction to that, but I don't know a single person that doesn't want to figure out how to reduce the complexity. But they know that the reality is if you could just leave that stuff alone and put a layer on top of that and let the AI do the work on top, that might actually be the easiest route.

Dave Vellante

>> So maybe I'm overthinking it. I think I am. This is helpful because I've been hung up on this-

Rebecca Knight

>> Data harmonization....

Dave Vellante

>> data harmonization layer. But basically what you're saying is look, irrespective of that data harmonization layer, which by the way I think is really important to build a digital twin of your business and a digital representation of that business, but that's going to take a decade. Irrespective of that, you can, in theory, do with agents what a human can do. And so a human has to deal with harmonizing data. The agents will have to deal with the harmonizing data. Take the humans off -

Craig Le Clair

>> The challenge is as I get into, again, in my Random Acts of Automation, which I'm not pushing on here, is that a lot of jobs exist because of this friction today. The reason you have a lot of guys in the middle office, in the back office is because they have to deal with 27 apps that are confusing, and they need this tribal knowledge to be able to figure out and navigate these corporate systems. An average broker spends 17% of their time navigating these back-end crappy systems. So there's going to be a big push to try to make that as smooth as possible, and AI interfaces will give it-

Dave Vellante

>> So it's a journey to Nirvana, my Nirvana. Your Nirvana comes before mine. I think you said '27 or '28. My Nirvana comes in 2035, but I do think it's coming. I'm serious about this. I think the industries are going to be completely transformed by AI.

Bobby Patrick

>> enterprise software.

Dave Vellante

>> Yes, absolutely. But it's a step stone-

Craig Le Clair

>> Well, the enterprise software thing, we're looking at AppGen platforms that can allow you to build custom code, and you are essentially buying these enterprise apps because they embody best practices and they were pre-built. Well, if you can build them so easily, and if the best practices are actually infused in the general training of the LLM, their value is diminished. So I'm not as dystopian to say that they go away, but their ability to keep adding on layers of value is going to be diminished by the ability to take AppGen, kick gen AI and build software that's needed. And there are a lot of CEOs that are sitting there going, "I'd love to just get rid of Salesforce. I'd love to get rid of all this SAP stuff."

Dave Vellante

>> Klarna.

Craig Le Clair

>> What's that?

Dave Vellante

>> The Klarna example. Did you see that? They say they'd get rid of it. I like the phrase tribal knowledge. So these agents are going to be able to observe humans. And because humans reason, AI can reason. We see it reason now with o1.

Craig Le Clair

>> Basically.

Dave Vellante

>> It says it's reasoning, but it's getting there. Jensen says it's going to happen, it's going to happen. But my thinking is that agents will be able to follow the reasoning traces of humans and then adapt to them and learn. Do you buy that? How long is that going to take?

Bobby Patrick

>> The key's the guardrail. Because the reality though is how do you keep an agent purpose-based, right? And it's confined to the role it's trying to do, like for a BDR agent or for an event management agent that helps with event management and things and doesn't end up straying from its particular type role. This is why we always say for us, it's humans using agents, agents using robots, and can that agent then use robots then to harvest information within the enterprise back to that secure systems of record or whatever and pull that data in that's appropriate just for that role, to keep the role... It's a small language model, maybe SLM, I don't know. But I think you've got to have it very, very tightly role-based so that it has memory on that role, that it's tied to that role. And that's a challenge right now because we all use ChatGPT all the time, but it's got all this information and all this data and it goes to lots of different directions, and we want to keep it nice and tight. I think that's a big part of trying to ground the agents.

Craig Le Clair

>> And that's where the context grounding becomes really important. So you're basically taking the enterprise data and putting it into an environment that's then absorbed by the LLM, and that's narrowing, that's providing some of the guardrails that you want the agent to do. And that's what we're all trying to figure out how to do that the best.

Rebecca Knight

>> still a human in that process too, who is checking and making sure that it's not straying, that it is-

Bobby Patrick

>> Fine judgment. Absolutely.

Craig Le Clair

>> Yeah.

Dave Vellante

>> I would think you have a better chance of controlling those agents with a strong governance framework than you do of controlling humans. A lot of cowboys out there. Keeping people in their swim lane and organizations is very difficult.

Craig Le Clair

>> Well, I tell people, they talk about hallucinations of 3% and this and that. I say, "I got guys I've worked with that are hallucinating all the time. This is not a strictly software-based problem that we have." Right?

Rebecca Knight

>> So I want to go back to the workers because this is, I think where the rubber meets the road here in terms of what you foresee as the biggest changes that will affect the kinds of skills that organizations are looking for and the kinds of the ways in which we will do our jobs on a day-to-day basis.

Craig Le Clair

>> So I'll give you my two from again, the random acts of automation.

Bobby Patrick

>> Available on Amazon.

Craig Le Clair

>> Yeah, I think there's an optimistic path for the middle, that the middle using AI, can do the work of more highly trained and credentialed professionals. You'll see this, the nurse practitioner gets even a better salary and a better work. The paralegal, your law firm in the future will have a few licensed attorneys and then a lot of super paralegals. So there's an optimistic path. Now the numbers don't work out that great. Not everybody's going to be dumped out of their 22 million cubicles in the US and have the constructive ambition and the critical thinking to really move in this direction. But that's one, I think it's a very optimistic path for what I call human touch workers, which is the lowest level of service workers. All of this physical robotics that we're doing, which are now getting integrated with LLMs is going to make their jobs a lot safer. The three D's, dirty, dangerous, and I forget the other one, but the janitor that now is managing three janitorial robots, the security officer that's managing a fleet of androids, these jobs are going to get a lot better and safer and higher value, and they're going to get paid more. The numbers are still a problem, but for many, there is a path if you apply some of the principles again in the Random Acts of Automation to really move in that direction. But unfortunately, there's going to be a lot of people from the middle that are going to drop to the front line service worker category and many digital elites, that's the upper end category that will fall into the middle as gen AI disrupts their knowledge based work that they do. So there's going to be this shifting. That's why in the cover, you see all these workers floating and moving from... They're really just moving from where they are and how they move when they move is something that I argue they have control of.

Dave Vellante

>> Given the lack of consistency and sometimes quality of LLMs and the expense, I mean, these things are expensive to run. And Bobby, you mentioned small language models, and that's really your game, is going into industry-specific situations and extracting or leveraging proprietary data for competitive advantage. So take us through your vision and maybe even timeline as to when the value of those types of activities will offset through better governance, proprietary data and maybe more efficient smaller language models, the value offsets the cost. How far away are we?

Bobby Patrick

>> You have a timeline, I think, didn't you? Well, yeah. We discussed this earlier a bit as well backstage.

Craig Le Clair

>> Yeah. I'm thinking '27, '28, our clients are saying, basically, "I'm not seeing the ROIs immediately from doing agent work." We say, "You just got to get on the boat. You just got to get on the path. They'll come." A lot of the agent work takes over more cognitive activities that aren't as easily ROI'd as the traditional UI paths. The RPA bot that's doing repetitive work. Something is done 1200 times a month by 37 people. It's a different game. It's different business value and ROI game with this stuff. But the end game is transformation, doing things very differently, introducing new products, new business models. That's the real opportunity here.

Bobby Patrick

>> And I think there's different categorization of agents. You have this in your talk, right?

Craig Le Clair

>> Yeah.

Bobby Patrick

>> And the agents that might have the biggest outcomes early may not be the ones that are the sexiest to talk about, the one that does all your flights and things. We unveiled the healing agent today, and one of our big focuses as a company is on reducing the change management required when an automation breaks, and we had this curve, we showed, Raghu showed, our CTO showed of how we track when automations break and the ability to actually make sure that we can repair them. And so this entire agent was built, LLM-based agent, we call it a healing agent. It's used on automation cloud, and it's going to make a order of magnitude change in that curve as we work towards reducing or eliminating errors in robot operations. So that is a real agent. It's LLM-based, it's got a real purpose to it, but it's not your agent sitting on your desktop. So there are different categories of agents I think that will... Customer service agents are obvious ones, but operations agents may be ones that have earlier value because they're more guardrailed naturally.

Craig Le Clair

>> I totally get that. And I basically tell our clients, "Go in the operations area for your first agents, know your customer application, your customer onboarding, mid-office and back-office, they're facing employees, they're co-workers with employees that are doing operational things. Absolute safest area, probably highest ROI area to get going with. Don't have too much customer interaction just yet."

Dave Vellante

>> Well, to your point, I mean RPA, ROI was a no-brainer. It was really clear. And then when UiPath came out and lowered the denominator, the cost, that was-

Craig Le Clair

>> That's why they're here.

Dave Vellante

>> It became infinite. Exactly. It's a little fuzzier, but-

Bobby Patrick

>> There's a lot of money on the table though, I think, in that whole equation though, Dave.

Dave Vellante

>> You think you should have renegotiated?

Bobby Patrick

>> No, the outcomes are so huge for our customers that , so it's kind of fun. It's great. The outcomes are amazing, but we left some money on the table.

Dave Vellante

>> It's interesting talking to the GSIs now who are leaning into AI and specifically agentic, they're trying to negotiate gain sharing. And you've got some experience probably with Forrester doing this, but I remember from my IDC days, it's always a funny thing. It's like, wow, that could be a big number. But then it's like, well, some customers say, "Great, you saved that." And maybe it's a one-time sort of uplift. I don't know. But I guess the point is, we had a conversation about productivity yesterday. We have not seen a sustained productivity boom like we have from the PC revolution, which lasted almost a decade, and we're promising... Essentially, the promise of AI is that we have a long sustained period. And that's where I think even though the ROI is kind of fuzzy, the transformation will become evident when we look back.

Rebecca Knight

>> I would say the reason we haven't is because of the friction. It's because we are all drowning in this admin that is affecting how productive we can be because we have to mess around with all these different apps. And so-

Bobby Patrick

>> I know....

Rebecca Knight

>> eliminate it.

Dave Vellante

>> So we think that Google and Facebook and Amazon and Microsoft are invincible, right? Because we know from history that's going to change. And so I expect there are going to be AI native companies that come out. They're going to do things for one-tenth of cost, we've talked about this, and 10 times faster that are going to disrupt industries. So my question is, do they have to lean in to RPA or will they invent something new? Is there something you guys got cooking in the labs?

Bobby Patrick

>> Well, there have been some startups in the agentic space that have had some great vision on that front, but the execution of the action side was missing. And I think that's always underestimated. And I think that the security side, it's not that it's underestimated, it's just it's so significant. So I agree, Rebecca, with the friction, but the security part of this because of agents and LLMs is so big that I think you got to break through both of those. But people often forget the importance in this whole thing of being able to go do reliable, resilient executions or actions based on whatever intelligence comes out of the system. And so I think that's where it's nice to have a very safe, reliable, trusted, proven workhorse that can do this. And that's where we come in. And I think that part of the story is underappreciated right now in the Benioff.

Dave Vellante

>> Well, I think compliance and governance and security is more important than ever. It's no longer the afterthought. We didn't even think about it in PCs and even... Cloud, okay, a little bit. But today, it's like table stakes. You can't do anything reasonable without it, or you're going to get in big trouble and cost your company a lot of money. And we've seen examples of that.

Craig Le Clair

>> Totally.

Bobby Patrick

>> Even on part of Inflection AI, they'll say, "I've been surprised by the amount of people who wanted to have meetings on that today," right? Because here it is on-premises, private cloud may be right, and it's kind of the antithesis of the all-go-everything-to-cloud story. But people are so worried about where LLMs and where that's running and what's in that data and who controls that. You heard the company, maybe you all didn't because you were doing great videos, live videos, but Eversource, the energy company, so she's on stage-

Rebecca Knight

>> They're coming on later.

Bobby Patrick

>> And they've shifted from an automation COE to an AI factory. It's an energy company. It's going to matter where a lot of this data sits, right? Particularly when it comes very operational for them, or it's security conscious. So the interest level in how to secure LLMs, it's really high, right? Or how to wall them off or guardrails. And I think again, that's been underappreciated.

Dave Vellante

>> Totally. Yeah. I talked to a couple of Wall Street guys about the analyst meeting that you guys had this week. They're very positive about Act 2. Of course, they want to see, okay, let's see the kind of a similar dominance that you had in the previous Act 1. And so it's there for your taking Bobby.

Bobby Patrick

>> So I'll say a few things about that, which is why being with Craig is so important. So Craig was there when... He actually put RPA sort of on the map, right? With the legitimization through your wave in 2017. And this was moving so fast, and I don't want to speak for Craig on this, but I think the thinking that is this similar about this, this is that kind of category. One of the other analyst firms has actually sized the market. I'm not sure if you've sized it yet.

Craig Le Clair

>> We have not.

Bobby Patrick

>> So they sized it, right? This is public data. It goes from zero to 4.2 billion in four years in software. The services opportunity is like 15 billion in new sales in that four years. Who knows if we're off by-

Dave Vellante

>> For what? For agentic?

Bobby Patrick

>> For agentic automation, for agents, not for robots. RPAs are separate, but the agent portion of that. Maybe we're off by a billion, two one way or the other, but the slope of that is pretty intense. But we need the education of the market. We need the analysts to be able to coalesce around this. We need the formalization of evaluation landscapes. But it's going to happen such a faster pace now, Craig, I think, right? It's not two, three years from now.

Craig Le Clair

>> I agree. I agree. It's at the same point where RPA was in 2016 and it's moving a lot faster than RPA.

Bobby Patrick

>> Right.

Rebecca Knight

>> Well, we could talk to both of you all day about this. This is a great conversation. Thank you both so much for coming on theCUBE, Bobby and Craig, and good luck with the book.

Craig Le Clair

>> Oh, thank you very much for mentioning it. I really appreciate that.

Rebecca Knight

>> I'm Rebecca Knight for Dave Vellante. Stay tuned for more of our live coverage of UiPath FORWARD 2024. You're watching theCUBE, the leader in enterprise tech news and analysis.