In this Mixture of Experts interview, ai.work co-founder Maor Ezer joins theCUBE + NYSE Wired co-host Gemma Allen to explain why “internal service” remains one of the last stubbornly manual frontiers inside the enterprise. Ezer argues that nearly half of internal workflows are service-oriented – think IT, HR, procurement and finance requests – yet still run on tickets, wait times and brittle handoffs. To this end, ai.work’s answer is an “AI worker” model: governed, policy-aligned agents that can take a request from systems such as ServiceNow and execute end-to-end work with human-like reasoning and controls.
The discussion digs into what “agentic” actually means when it moves beyond buzzword status and into operational reality: multi-agent orchestration, task decomposition, validation steps and context management that can handle non-deterministic workflows traditional automation never cracked. Ezer also outlines the human side of adoption – how frontline teams become the bottleneck (or the accelerant) – and why reskilling matters as roles shift from doing tickets to training and supervising AI workflows. He closes with ai.work’s enterprise focus, its discovery approach using historical ticket data, and early performance markers such as autonomous ticket-load handling and faster time-to-impact.
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
theCUBE + NYSE Wired: AI Agent Conference. If you don’t think you received an email check your
spam folder.
Sign in to theCUBE + NYSE Wired: AI Agent Conference.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open this link to automatically sign into the site.
Register For theCUBE + NYSE Wired: AI Agent Conference
Please fill out the information below. You will recieve an email with a verification link confirming your registration. Click the link to automatically sign into the site.
You’re almost there!
We just sent you a verification email. Please click the verification button in the email. Once your email address is verified, you will have full access to all event content for theCUBE + NYSE Wired: AI Agent Conference.
I want my badge and interests to be visible to all attendees.
Checking this box will display your presense on the attendees list, view your profile and allow other attendees to contact you via 1-1 chat. Read the Privacy Policy. At any time, you can choose to disable this preference.
Select your Interests!
add
Upload your photo
Uploading..
OR
Connect via Twitter
Connect via Linkedin
EDIT PASSWORD
Share
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
theCUBE + NYSE Wired: AI Agent Conference. If you don’t think you received an email check your
spam folder.
Sign in to theCUBE + NYSE Wired: AI Agent Conference.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open this link to automatically sign into the site.
Sign in to gain access to theCUBE + NYSE Wired: AI Agent Conference
Please sign in with LinkedIn to continue to theCUBE + NYSE Wired: AI Agent Conference. Signing in with LinkedIn ensures a professional environment.
Are you sure you want to remove access rights for this user?
Details
Manage Access
email address
Community Invitation
Maor Ezer, ai.work
Exploring the Advancements and Challenges in AI Agent Deployment
John Nay, founder and Chief Executive Officer of Norm Ai, joins theCUBE's special presentation with NYSE Wired, focusing on the upcoming Artificial Intelligence Agent Conference 2025. Hosted by John Furrier, co-founder and co-Chief Executive Officer of SiliconANGLE Media, this insightful discussion covers the pivotal developments in AI infrastructure and the regulatory complexities faced by enterprises.
In this episode, Nay shares their expertise in regulatory AI infrastructure, particularly as it pertains to AI agent deployment in highly regulated sectors. The conversation, hosted by Furrier, delves into the evolving landscape of AI technology, compliance challenges, and the strategic initiatives underway at Norm Ai to address the pressing issues surrounding AI deployment. The discussion provides valuable insights for both technology and policy influencers.
Key takeaways from the discussion include the emphasis on the need for dynamic, real-time compliance frameworks that align with regulatory standards, as emphasized by Nay. Furthermore, the episode highlights how enterprises can leverage existing compliance structures to integrate AI technologies more effectively, offering a glimpse into the future of AI agent scalability and regulation. The conversation underscores the importance of bridging the gap between engineering, policy, and technology for sustainable AI innovation.
In this Mixture of Experts interview, ai.work co-founder Maor Ezer joins theCUBE + NYSE Wired co-host Gemma Allen to explain why “internal service” remains one of the last stubbornly manual frontiers inside the enterprise. Ezer argues that nearly half of internal workflows are service-oriented – think IT, HR, procurement and finance requests – yet still run on tickets, wait times and brittle handoffs. To this end, ai.work’s answer is an “AI worker” model: governed, policy-aligned agents that can take a request from systems such as ServiceNow and execute end-t...Read more
exploreKeep Exploring
What is the problem being addressed with AI technology in enterprise workflows?add
What is the future direction of the class or team in relation to AI and automation?add
What impact is the introduction of AI workers having on human roles and performance in the workplace?add
What is the current status and future plans regarding team growth and funding for the company?add
What is the main area of focus for the speaker regarding AI in enterprises?add
>> I'm Gemma Allen here at our studio with the New York Stock Exchange. This is theCUBE with NYSC Wired, and we are talking to some of the leaders who are going to take the stage at the AI Agent Conference here in New York in May. Joining me now is Maor Ezer. Co-founder of ai.work. Welcome, Maor.
Maor Ezer
>> Thank you, Gemma. Great to be here. What an exciting venture.
Gemma Allen
>> For sure. A lot of noise behind us here today.
Maor Ezer
>> A lot of noise, a lot of money and action happening.
Gemma Allen
>> A red day in the markets, but look. Okay, so fill me in. ai.work., to those of us who may not be familiar with the company, what exactly do you do? What problem are you solving for?
Maor Ezer
>> Yeah, today about 45% of internal workflows in enterprise... I know it's not as fancy as changing the world, but 45% of internal workflows in enterprise are service-oriented. So these are basically employees that need certain stuff from IT, from HR, from procurement, from finance. And today it's fully manual. So what we're doing is we're building AI workers that can basically fulfill a lot of the internal service. So take IT help desk. Instead of you having to submit a ticket and waiting three days for someone to answer you and then a human needs to take care of that, we now have built complex agentic technology that can take that request right there from the ticket that was created in ServiceNow, for example, and operate that ticket like a human being aligned to policy, fully governed with all the bells and whistles and fully execute that ticket.
Gemma Allen
>> So essentially create a record of the intersection where an employee works as a system within a company and also learn something from those patterns and those behaviors, I'm guessing, right?
Maor Ezer
>> Yeah, of course. We do put in some learning and we understand the different ways, but we're very focused on the execution of the need of the employee first.
Gemma Allen
>> So tell me a little about the space you're entering from a competitive perspective. A lot of SaaS players are doing a little bit of this themselves. Like you mentioned ServiceNow, Workday, other IT help desk ticketing systems.
Maor Ezer
>> Yeah.
Gemma Allen
>> Are you coming in to replace an incumbent or are you coming into work next alongside them?
Maor Ezer
>> That's a good question. Look, I think AI in the last couple of years, specifically in the last year, has kind of sent everybody thinking, what do we do with our... Every vendor is thinking, what do we do, right? They're stressed. They're even more stressed than the guys down here, right? And they're trying to understand what's going on in the market. And there is a big shift. There is a shift that we haven't seen in our generation yet as big, maybe on premise to cloud, but this is like 10 times more. And I don't think it's about me replacing ServiceNow or ai.work. I'm sorry, replacing ServiceNow. I think it's more about the evolution of the enterprise and what we really need in the enterprise to solve said problem, right? But yes, we do take a lot of the business logic in servicing the person and automating that out of the incumbent software. And it's a very noisy market.
Gemma Allen
>> For sure. The term agentic, it feels like it's become somewhat of a buzz term, right? A catchall. It has a lot of different connotations all of a sudden. From your perspective, really though, it's about pattern recognition, about being able to better the process, right? One agent can really have multiple minds in terms of what it learns, how-
Maor Ezer
>> It's amazing.
Gemma Allen
>> Adjusts and reiterates. Talk me through the technology side of this from your perspective in terms of what you're building in particular.
Maor Ezer
>> Yeah, this is a new art. I have to say that every couple of weeks we look at, let's say, a use case that was run by one of our customers or something that we are experimenting with and we're just surprised every time of the results. This is like someone gave you a really nice kind of playground and you can now build whatever you want. So this technology is, you hear a lot about agentic because it's just this new toy or this new life-changing technology that everybody can shape in a little bit of a different way in how they execute on top of it. And that's the beautiful thing. And I think we've built the full agentic system, multi-agent, context, everything you hear, all the buzzwords, but it's really interesting to see an AI worker receive a task and then call up another agent and that agent executes one piece of the task and you don't need it to understand the full context of what it's operating. And then another skill, what we call agent to do something and then do some validation and all these pieces and you see it orchestrating this complex workflow, it's just mind-blowing.
Gemma Allen
>> I want to talk to you about context, but before we go there, in terms of the IT or ticketing, broadly speaking, that experience, right? We've all had experiences where you log a ticket for something that seems basic, takes days to get a response, whether it's about your pension plan or your access to an application from home, whatever it is, right? We've all been there. But a lot of that work, it's scripted. You're following humans in responding to help desks are following a script, right? How do you see the role of humans playing out as companies like yours and products like this become more and more successful? Where do you think that class will go?
Maor Ezer
>> So let's break it down for a second. Is it scripted? If it was scripted, then there would be software for it, right? I don't have to have a sales call with a lead and go file it in the cabinet and write it on the piece of paper. I just spoke with Gemma, right? I have a CRM that solved that problem for me. There is a reason that this problem is not solved. The scripted, what's called like the if this, then that, or deterministic automation did not solve the problem. And what we're seeing with agentic is kind of that new way of reasoning and thinking and using kind of the decision making that a human would use to execute said workflow, right? So it's not that scripted in terms of that manner. And what's going to happen with the humans is the most interesting piece, right? And I do this all day. We're seeing the IT AI worker, Henry that we just launched, we're seeing him handle becoming a top performer in the team, number two or number one even in probably 60 days.
Gemma Allen
>> Wow.
Maor Ezer
>> So imagine that you have a team of hundreds of people and suddenly you have this new AI worker that is top performer fast. Like, would you fire your top performing employee? No. But what's really interesting is yes, the world will change, not even related to us. The world will change and human roles will change. And some of the roles we're doing didn't exist 10 years ago, right? So it's okay. But we're seeing a huge trend of a lot of employees, if you take IT, for example, that are actually stepping in, they're training AI, they're starting to build AI workflows and agents, and suddenly they have a new skill. They're even more valuable to the organization now that they've re-skilled. So I don't think it's about AI here to replace. I think it's about AI just moved the entire kind of platform for skills in the organization. Some of it the AI will take. Some of it, humans will take new skills that we didn't think about.
Gemma Allen
>> So Henry, he is learning a lot about different employee experiences. He's becoming a mastermind in repeat problems. He is seeing possibly the same trends, the same frustrations time and time again, right? And that data is powerful.
Maor Ezer
>> Absolutely.
Gemma Allen
>> It's powerful in a number of settings. How do you think about the transferable longer term opportunity for a company like this? Where do you see this headed?
Maor Ezer
>> I think that's the beautiful thing because right now, companies that want to do change management, for example, and they want to swap ServiceNow for whatever, for Freshservice. It's very hard because everything is already embedded and it takes a long while to change up software. But if you look at what's happening, since the business logic is being extracted from these platforms and now it's shaped into AI, now the underlining systems can be changed faster or tomorrow when you do this. And also that content now, the AI starts becoming a more dynamic brain. So like you're saying, it ran app access control requests a million times. It now really can build the policy for the company much better than we built it before on some document and confluence, this is the policy. Now it's a dynamic policy. It learned. Oh, the CEO has asked for access for this app. Oh, give the CEO whatever they want, right? Ooh, new piece in the policy. I've learned something, right? I just gave an edge case, but it's a really interesting place where we're moving as humans who manage workflows very statically to these dynamic, smarter workflows.
Gemma Allen
>> So your buyer persona for this, you work mostly with IT environments, correct?
Maor Ezer
>> Right.
Gemma Allen
>> But it has a lot of transferable opportunity. What sort of profile of customer are you targeting? Is it SMB, large enterprise, and what sort of stage or maturity are they already at?
Maor Ezer
>> Wow, that's a good question. That's a good question that we deal with every day. As much as I'd love to service SMB and mid-market, because that's kind of like where we "grew up", but you just see a direct correlation between one, the maturity and second, the amount of tickets and employees. As much as the volume is bigger, when you solve a problem, it's the same problem for a small company and a big company, right? It may be more complex in a big company, but when you solve a problem and you take on the ticket load, for example, and you start putting chunks of it autonomously, that saves them a lot more money and time and productivity to their employees if they're bigger company. So to your question, we cater more to enterprise. Let's call it large mid-market to enterprise. And we just see them a little bit more mature with their tech stack too. So if you know how to buy enterprise software, if you bought Workday and you bought ServiceNow, you understand what it means to implement software. And sometimes the SMBs, they just want a quick fix, right? So one, the problem is not as painful in SMB. Second, they're not as mature in implementing software as an organization, not of course as individuals. So we cater to the enterprise, we focus a lot on CIOs that are already doing AI implementations that the teams... the teams, you're probably going to ask me this, so I'll say it in advance, but the teams, the frontline workers, they've been the hurdle right now to stalling AI adoption in part, right? The technology, everything, right? But management want to push it sometimes, and the workers, they just don't have an immediate urgency to do it. And what we're seeing is the better organizations that adapt and transform faster are the ones that management and frontline workers are very aligned, and then it just goes fast, right?
Gemma Allen
>> I mean, I heard this week, I hear it all the time now, AI might not necessarily take your job, but if you don't use AI in your job-
Maor Ezer
>> Exactly.
Gemma Allen
>> You're putting yourself at a much bigger risk, right? You need to use it to your advantage, to harness it in some way.
Maor Ezer
>> At the end of... We saw what happened in Amazon yesterday, right? 16,000 people were let go for various reasons, but you would never let go to the person that is currently doing AI transformation and is leading the pack on that, right? And I think that's what people need to think about.
Gemma Allen
>> So right now, this is employee to company. It's solving for tickets between me as an employee to my employer, right?
Maor Ezer
>> Correct.
Gemma Allen
>> Seems as though there's huge amounts of opportunity for external ticketing to-
Maor Ezer
>> Right.
Gemma Allen
>> Customer to customer, client to customer-
Maor Ezer
>> Support.
Gemma Allen
>> Support across the board, because those tickets are often very high stress rationale, right?
Maor Ezer
>> Yeah.
Gemma Allen
>> Talk to me about the product map. Where are you going with this?
Maor Ezer
>> So customer support is less on our product roadmap. And it might sound funny, but it's actually an easier use case because it relies less on complex enterprise systems, and it's more kind of what's called rag, right? Just answer or I need to change my ticket for the flight, right? And it can automate that process pretty quick. So that's a more actually more mature AI market, customer support. We're very focused on the internal service, so we're focused right now on IT. We're doing builds for customers that are in procurement, in HR operations. That's a great place to start as well. Huge value in finance in those worlds.
Gemma Allen
>> These conversations are prospective customers of yours, especially from a practice perspective. Talk to me about the discovery and the execution. If I hire ai.work to automate or to bring Henry in and lead the charge for my ticketing in, let's just say a HR space, how long does it take to roll this out? How plug and play is this?
Maor Ezer
>> Yeah, it's a good question. So what we do is we run... First of all, we run our discovery agent. What we do is, you've got a huge ticket backlog or like old tickets that have already been finished and that's a gem for data for AI because there's so much information. What are the requests? Was it handled? How long it took? Comments, right? There's a whole textual base. So what we do is when we start with a customer, day one, give me all your ticket history, let me scan it. We run our agent, it reclassifies, recategorizes all the tickets, and then it looks to our platform, the integrations we have in place, the use cases we've already done, and it comes back with a roadmap for potential agentic automation.
Gemma Allen
>> Oh, wow.
Maor Ezer
>> And then we sit with a customer, all right, here's the roadmap. Here's the low hanging fruit, high volume, high potential for automation, let's go, right? And we start going one by one and we're seeing, I want to be not too enthusiastic with the ROI, but we're seeing three to six months, we're seeing the AI worker become number one. We're seeing us handle about 30% of ticket load autonomously and then another X percent assisted.
Gemma Allen
>> Wow. So, Maor, this isn't your first rodeo.
Maor Ezer
>> Yeah, it isn't.
Gemma Allen
>> This is your third company.
Maor Ezer
>> Correct.
Gemma Allen
>> Talk me through a little bit about your own journey, your founder journey, I guess, which always interests our audience. You said you just moved here to New York from Tel Aviv this year.
Maor Ezer
>> Yeah, correct.
Gemma Allen
>> You're doubling down on ai.work.
Maor Ezer
>> I wanted the coldest winter in New York history. Here I am.
Gemma Allen
>> You wanted to make snowmen.
Maor Ezer
>> I just wanted a really nice coat and I didn't have a reason to buy one in Israel. So we moved here. No, I'm kidding. Yeah, we moved here building out the go to market and the business here. I wanted to be closer to the customers.
Gemma Allen
>> Okay.
Maor Ezer
>> One of the things that I learned the most in my entire career is, yes, you can build amazing companies in other territories and there is incredible talent, but being close to the customer that you're selling to, nothing beats it. You just understand the culture and the mindset of the company or even the country that you're in better. I myself, this is a third company. I had a music licensing marketplace, believe it or not, back in 2008.
Gemma Allen
>> We talking Maxstar days or-
Maor Ezer
>> But more licensing.
Gemma Allen
>> Okay.
Maor Ezer
>> Like if you need music for a commercial use, like in a video or whatever here playing in the background. And my last startup, Abbi, A-B-B-I, was sold to WalkMe. WalkMe is a digital adoption platform. We joined pretty early on and this was back in 2017, early 2017 or 2016. I might have been messing it up. And I spent the last seven and a half years there.
Gemma Allen
>> Okay, wow.
Maor Ezer
>> I ran the mobile line of business and then I was CMO of the company and we ended up growing the company tremendously into a NASDAQ IPO and-
Gemma Allen
>> We won't talk about that.
Maor Ezer
>> And it was acquired a year ago or a year and a half ago by SAP.
Gemma Allen
>> Oh, wow. Okay.
Maor Ezer
>> So I've seen a lot of internal use cases and that's where we came up. Me and Neer, my partner, he was also at WalkMe for 12 years, first employee, VPR&D and CTO. And we just saw how SaaS is breaking right in front of our eyes. So many apps, so much confusion, so much complexity and workflows and knowing AI and we were deep into it, we just said, there is a better solution now. There is an unlock of technology and let's do it.
Gemma Allen
>> And it certainly feels like the Wild West, right? So you've raised seed round.
Maor Ezer
>> Correct.
Gemma Allen
>> Are you heading towards a Series A? What is the next 12 months? You growing the team in New York, I assume, to have some company on the ground?
Maor Ezer
>> Yeah, we're growing the team and we will head into the A round soon enough. Knock on wood, it's going good. We did a seed round of $10 million. We were lucky enough to have amazing investors like A Star here in the U.S., Lul Ventures in Israel, First Minute Capital out in the U.K., along with some amazing people that joined the round and some amazing funds that joined the round. And yeah, we'll soon do the A and we'll keep growing. We just want to fully fulfill the vision of the AI worker. That's the focus. I'll be honest. I'm not focused on raising capital. I'm not focused on anything else and just seeing agentic AI and the AI worker really do what it's supposed to do. It's a hard challenge, right?
Gemma Allen
>> For sure.
Maor Ezer
>> It's not easy really convincing a company, an enterprise to kind of step back and give processes to AI to handle end-to-end, even if there are guardrails and governance and human in the loop and everything. So that's where I'm focused. I want to see those hard ROI metrics. I want to be able to say to you, in a year's time, sit here and say, our AI workers are operating IT all over the world right now.
Gemma Allen
>> Amazing. A word class product right-
Maor Ezer
>> Imagine Henry's CV. He's working in the best companies in the world.
Gemma Allen
>> Henry's going to be in demand. Well, you're going to be speaking at the AI agent conference in New York in May. I think you mentioned you might have some of your Henry fans joining you on stage.
Maor Ezer
>> Yeah. We're going to bring in great leaders. I'm going to host a panel about autonomous AI and AI workers in enterprise. We're going to have great speakers. I won't mention names yet. We'll publish the next week, but we're going to have leaders from top companies, from tech, from accounting, from different sectors of the market, maybe oil and gas too. And you'll see kind of how everyone is tackling it in a different setup, but seeing those ROI levels of just driving autonomous AI.
Gemma Allen
>> Well, we look forward to it. We'll be there too.
Maor Ezer
>> Yeah, me too. It's going to be an amazing event.
Gemma Allen
>> Absolutely.
Maor Ezer
>> I hope you're going to be there.
Gemma Allen
>> Oh, we'll be there. The Cube and NYSC will be there. We're covering it with Simon. So listen, Maor, thank you so much for coming on theCUBE.
Maor Ezer
>> Thank you so much.
Gemma Allen
>> Keep us informed as things progress for you. We certainly look forward to seeing what Henry can do in this world.
Maor Ezer
>> Absolutely.
Gemma Allen
>> Thanks so much.
Maor Ezer
>> Thank you so much for having me.
Gemma Allen
>> I'm Gemma Allen coming to you from our studio here at the New York Stock Exchange. This is NYSC Wired with theCUBE. We are talking to some of the leaders who are taking the stage at the AI Agent Conference in New York in May. Thanks so much for watching.
Gemma Allen