Audit & Beyond 2025 | Happy Wang & Anton Dam, AuditBoard

Clips
More from Audit & Beyond 2025

Anton Dam

SVP of AI & Product

AuditBoard

Haibei (Happy) Wang

Chief Product & Technology Officer

AuditBoard

From toil to traction: How AI is overhauling the audit

The increasing complexity of governance, risk and compliance, or GRC, often leaves internal audit teams caught up in manual processing, preventing them from taking on more strategic roles in the business. Enterprises are in need of a connected risk platform to automate busywork and raise productivity.To address this demand, AuditBoard Inc. is integrating new artificial intelligence capabilities, including its new Accelerate offering, to speed up the auditing process. The aim is to unify all risk data and insights, according to Happy Wang (pictured, left), chief product and technology officer at AuditBoard

play_circle_outline Introduction of new products and features at Audit Board's event, including Accelerate.

play_circle_outline Enhancing Auditability and User Empowerment: The Future of AI with a Focus on Reliability and User-Centered Design

play_circle_outline Empowering Auditors: How AI Elevates Roles from Manual Tasks to Strategic Contributions in the Workplace

play_circle_outline The need for AI governance and risk management due to emerging AI technologies.

Info
Transcript

Happy Wang & Anton Dam, AuditBoard

Anton Dam

SVP of AI & Product AuditBoard

Haibei (Happy) Wang

Chief Product & Technology Officer AuditBoard

In this Audit & Beyond segment, AuditBoard leaders Anton Dam, senior vice president of AI and Product, and Happy Wang, chief product & technology officer, join theCUBE’s Rob Strechay and Rebecca Knight to unpack the company’s newest AI capabilities – including the launch of Accelerate. Wang explains how Accelerate doubles down on purpose-built, human-in-the-loop design to embed AI directly into day-to-day workflows: document intelligence to turn sprawling docs into clear narratives and flowcharts; scenario planning and stress testing that elevate teams from o... Read more

explore Keep Exploring

What advancements in AI features were introduced in the core workflow last year, and what investments were made this year to enhance those capabilities? add

What are the processes involved in ensuring user involvement and traceability in AI evaluations within enterprise software systems? add

What role does AI play in transforming job functions and enhancing productivity within an organization? add

What are the challenges and considerations involved in managing AI governance and risk? add

bolt Powered by CUBE AI

Happy Wang & Anton Dam, AuditBoard

search

Rebecca Knight

>> Good afternoon everyone, and welcome to theCUBE's live coverage of Audit Boards, Audit and Beyond. I'm your host Rebecca Knight, alongside Rob Strechay. Rob, we are both just from the main stage this morning, what we're heard a lot of illustrious executives from Audit Board talking about all the new products, the new features, the new bells, and whistles, lots to talk about.

Rob Strechay

>> Yeah, a ton to talk about, and I don't think there's two better people to talk about it with because a lot of the innovation that's coming out, and really taking the toil out of audit, and risk, and compliance is right here.So, very excited.

Rebecca Knight

>> Indeed. So, with that, I would like to introduce our next guest. We have Anton Dam, SVP of AI, and product at Audit Board. Welcome, Anton.

Anton Dam

>> Thank you. Great to be here.

Rebecca Knight

>> And Happy Wang, chief product, and technology officer at Audit Board. Thank you so much for coming on, Happy.

Haibei (Happy) Wang

>> Thank you. I'm super excited to be here.

Rebecca Knight

>> Yeah. So, actually I'm going to start with you happy because of what we were just talking about here with all the launches of these new products, and features including the launch of Accelerate. Can you walk us through a little bit about how this solution builds on Audit Board's existing capabilities?

Haibei (Happy) Wang

>> Yes, sure. So, last year we introduced the AI features first time into our core workflow. If you remember we had AI intelligence recommendation. We have using AI to summarize a lot of description, generous issues. We got a really, really good wall factor when we used the customer so that we know there is a lot of appetite in this AI world, even in the risk audit compliance. So, this year we doubled down. We doubled down a lot of AI investments. That's why we created Accelerate. Accelerate has really try to build the same principle, want to purpose-build, have a human intelligence loop. We have intuitive user core use experience, and we invite even more workflow. So, if you look at today, we elevate to a document intelligence, you can summarize a lot of documents just spread out to the narrative, and the flowchart. We did scenario planning, which is really turn the risk people from operation to it just enable it, right? We take the massive data for them so they can look at data, create a scenario planning model, stress testing, so they become a strategic enabler. We also did a security questionnaire. We have tiles questioning, we can load into the spreadsheet. We're mapping them, and it's been the same format. So, literally we're upper level, all the AI capabilities really super integrate into a day-to-day workflow.

Rob Strechay

>> Yeah, Anton, I think, again, everybody looks at AI, and sometimes there's trepidation with it. Sometimes there's embracement. I think where you guys are really focused is on how does it help people reduce the manual work, and how does it become their assistant in this? Could you elaborate on, I think there's people still on the risk side are trying to figure out agentic technology, and agentic AI, and how it becomes audible, traceable, and secure. Because obviously you do that for people, but you're also having it in your product.

Anton Dam

>> Yeah. So, I think I just want to get this straight out of the way. We build for users, we build for humans, and that underpins everything that we do. And so the practitioner, and the organization are always fully 100% in control. That's the core principle with how we build these things. And you saw it in our keynote, and if you get a chance to see a demo from the folks behind us, you'll see that, too. Every single thing that happens, Every action, every record, every recommendation before it gets entered into the system of record either gets approved by human, or really there's a process. A user gets a chance to review, edit, and then accept into the system of record nothing's happens without the user. Now to your... That's the first half of my answer to your question. When it comes to the second point of like, hey, auditability, and traceability, the name of the game in AI is evals. And really what that means is how do you know that something actually worked? Right, now that is a super high level, was something completed? Did it align to the user for enterprise software like Audit Board to the organizational intent, right? Was it done? Was it done correctly? And then you get into other more traditional Gen AI evals around, hey, is it correct in the sense of that it makes something up, that it leaves something out, that it adds something while correct, maybe shouldn't be there. And so that evals framework to make sure that whatever codes live, and gets in front of a user is correct. And then second, over time, does it drift, right? You'll hear things about model drift, and all those kinds of things because hey, AI gets deployed, users use this. It's a live system. And again, to what degree do things change? Do models change? Do how people use it change? And all the guarantees that we have in place when we deploy it are still correct. So, that's observability. And I think it's also useful to remember that our goal isn't to have the best Q&A system for Audit, and GRC. Our goal is to save people time, and enable our users to be more strategic. Now what that means is that we need to enable our users what well AI the helps, and accomplish long horizon, or for more layman's terms long-running multi-step tasks. And so it's such a very complex evals framework that lives on top of the foundation models that our great partners at OpenAI, and other places provide. But really once you get that sort of right, then things really begin feeling like magic.

Rebecca Knight

>> Well, one of the things that you said that really stuck with me on the main stage was you were talking about how when you were first beginning in this profession, and you were meeting your new colleagues who all were super smart people but stuck doing this really repetitive boring stuff, and even young Anton could recognize there's higher value problems that we could have these sharp minds working on. And I know Happy you've talked about AI not just being about efficiency, but about empowerment. Can you talk a little bit about how you envisage this becoming playing a more strategic role in an organization?

Haibei (Happy) Wang

>> Yeah, the world is changing. So, every single person, even my own role, how I see myself, I've been in tech for the last 20 years, how I lead an engineer product team even five years ago compared what I do today is already different. So, goal, one of our AI purpose is really is to amplify human potential. So, our auditor, CIO, CISO, we've been talking to them, there was a lot of part of work their passion doing, they love to but there are a lot of part of work they didn't want to do, right? It is very manual work, let's have audit. There was a lot of manual tasks. We know 70% time they work on the field work is necessary to complete audit planning, but it's not the most fun job to do. So, our goal is really to using AI technology to help them to get rid of those work no one wants to do, and really level them up. That's probably empowerment, because everyone want to see themselves strategic, be smarter. So, the way you elevate them is give them a smart task, and automate all the manual work, which is a lot of tedious work. The other way is we want to also, a lot of people in this role, when I interview this customer, they feel they don't have the voice on the table, because all they see is every time they go to give the report to their board, to the CEO, "Oh, here's my finding, here's the things you do it." And no one really appreciates that. However, what they really want to do is they always just happy. Sometimes I also feel if I can tell them early, I have a strategic view in instead just come in, give you a board reporting, and I actually can help a company better, and mitigate risk early, and more impact. So, how we actually make them be more strategic also have to free up their time. Because right now they're doing all those kind of tedious work. That's why all the work we try to do either is a summarize document, continuous monitoring shift, our compliance officer just look at evidence request every day, manually pull, and give the visual part. They can continue monitoring all the system. Those are very important. And the most important thing is we really think AI is also is a key component in really achieve our connect risk vision. So, this is our big ambitious for Audit Board. We want to be able to category, defining, connect risk, everyone talk about it. A lot of our competitors talk about it, but no one has done that beautifully. We think we have a lot of advantage, and we using AI power behind the scene, we can really connect all the data from the risk issues, compliance framework, policy, you name it. If you look at our connect is all the bottom solution, all tie back is the core data in the in the middle. So, our goal is really is not to replace your job, is to help you do your job better, smarter, and all of a sudden you become a... People look at you, you are a strategic player in a company. It is not just operator. Just always give you the...

Rebecca Knight

>> The boring reports.

Rob Strechay

>> Yeah, I think actually I loved the visualization with the bow tie analysis, and the scenario planning. I think in particular, I think the scenario planning got the loudest ovation in Audit Board, which I'm not surprised, because there's a lot of toil as I put it that goes into scenario planning. But you guys seem to really be using agentic technology, which is more than just a chatbot. And I think why don't you kind of talk to how you see agentic, because it looks like you're leveraging ML, and other things, and bringing that together in things like scenario planning that help them get a better view into that.

Anton Dam

>> So, I think that, well first the word agentic takes on a lot of meanings, but I sort try to keep it at least sort of how we view the world. And this is not getting into the semantics of what's an agent, what's agentic? No, but really it is like how do you help a user complete a long-running multi-step task. Now, short term, that means that you want to enable co-pilots that allow users for every step along the way to complete things very quickly with a lot of context automatically pulled for you. Meaning that typically today, especially in Workflow tools, everyone has 15 tabs open, and they're trying to stitch things together. No, really what you really should be doing is you press a button, or you type in a small little thing, and you get a result that you can review that distills all that information out there for you. So, then two things you get. One is the result itself, and also all the references. So, you saw an automated security questionnaire, you want to have some confidence that the system's not just making stuff up. And so where is it coming from? And again, to be honest, things do, it's not perfect. It gets things wrong at times, and that's why human of the loop, and the references are hypercritical. Now, where does this go? And again, the thing I want everyone to remember here is a 90% system for a long-running multi in a five step process will fail... If every single step succeeds 90% of the time it will fail more than half the time if there are more than five steps. And that's the thing that we take to heart, that hey, end-to-end agent autonomy, which is where we want to take things. We recognize that we don't get to 100%. It's long-running multi-step tasks. And so that's what we really need to be building towards. And so that's how we view, there's automations, the co-pilot, the assistive experiences.

Rebecca Knight

>> But so many organizations are wary of that, especially as they're experimenting more with AI. Audit Board serves more than half of the Fortune 500. Can you talk a little bit about what you are hearing, what feedback you're getting from those customers as they start to integrate AI into their risk?

Anton Dam

>> Yeah, I think I would sort of say there are three things. Number one is there is a top level mandate from senior management, if you will. In some cases it's the board. Sometimes it's CEO, but really...

Haibei (Happy) Wang

>> Sometimes more board.

Anton Dam

>> Yeah.

Rob Strechay

>> Yeah.

Anton Dam

>> Teams have to do more with less. There's genuine beliefs in the potential of this technology. And so folks are figuring out how to do it. Now, so, that's number one. Number two is when they're trying to figure out how to make it happen, they're trying to review what their stance, their posture posture. And so this is why we acquired Fair Now, because governance of this is totally different. Actually the only other analogy I can think of is IT risk. 50 years ago, or 40, or 30 years ago, there is no real risk management. All of a sudden these compute things like computers, and interests are showing up. And that's an entire broad attack surface for the enterprise. We haven't seen anything since. And so now instead of like we're looking at lead sources when it comes to AI governance, who needs to manage AI risk? It's a recruiting team. It's the legal team that's reviewing vendor contracts. They have no idea what's going on. And so they're managing things in Word documents, in GRS spreadsheets, and whatnot. And they just need a tool that tells them what do they need to do in the world? What are some best practices, and how do they continue delivering their products, and their services to their customers while managing that risk? So, that's why the Fair Now acquisition for us is hypocritical to where we believe this world is going.

Rob Strechay

>> So, kind of final question to bring it home here, where do you see risk management going, and what's kind of your focus in the next year to come as you continue to differentiate yourself from your competitors?

Anton Dam

>> I can take this, too, which is, I think there, I would say two things. One is completeness. And so we're launching a lot of really exciting AI capabilities at this conference again, and I'm hoping that everyone that's watching this, everyone that sees the demos behind us can see really where this is going, the real potential of this, and how it can be impactful from day one. But it's not in every single workflow. And so, actually, I had another interview like yesterday, and one of the things I said is like we believe that we have to meet our users where they are. We love our platform. We believe that a lot of the work, and a lot of value can be accomplished, and achieved within our platform. The reality is people use the tools they use now. So, we have two choices. You can try to force people onto your platform, or you can meet them where they are. Some of our competitors, and key of competitors have chosen the former, but because we build for our users, we're meeting them where they are. So, Excel is the perfect example there. Most of our platform, most of our users live a lot of times in Excel. We're extending our platform into more Excel based extending our platform to deliver these capabilities that you've begun to see in our launches this week to Excel. And so they don't have to change everything that they're doing just to get this value, get these capabilities right away. So, that's number one. Number two here, that is, as you saw with scenario planning, it's this really powerful capability to sort of imagine, hey, what are the things that might happen? Now, a key component to that is data. Data, not just internal, but what's going on in the world. And so that combination between internal external, we've done a little bit of it, and we're going to do a lot more for different types of private, and public data. And so what I always used to say, I used to say that the best stories are always private. How do you leverage your internal data, but reality hits you when you walk outside. And so that combination is really what we're going to invest into going forward. So, again, completeness, and enablement is where we're going.

Rebecca Knight

>> Well, Anton just gave us a great thing, but Happy we're going to give you the final word here about where this is going.

Haibei (Happy) Wang

>> I think the most important thing is, that's why I keep going back, is I want Audit Board to be able to achieve this connect risk. When I say connect risk platform, it is not just a platform, it's an ecosystem. So, for me, risk no longer be isolated. There is a million types risk, because obviously AI risk is on top. You have a cyber risk, but every single risk they co-related, right? They connected together. So, you cannot treat, you cannot just admit it... What I try to say is you cannot just mitigate a risk in a siloed motion, okay? So, what we try to build is build a platform to connect everything together. You have visualization to see which part of the risk trigger the other part of the risk, and then on top of that, we create action plan for you, because we talk to a customer. Customer says, "Don't show me just a dashboard. Everything's red." I don't know what to do with that. So, we want give you insights. We will not only tell you where your risk come from, always tell you how to mitigate the risk. And obviously end of vision is always be able to predict, right? That's a lot more working too. I mean, predict, we certainly try to do some work here. Obviously we have to trigger the two model. We have to look at all the data points, look at all the historical pattern stuff. But if we also can really shift them all left to be able to surface risk a lot sooner than you actually coming to you, it will help to mitigate a lot of business

Rebecca Knight

>> Terrific. Well, Happy, and Anton, thank you both so much for coming on theCUBE. A great conversation.

Anton Dam

>> Thank you.

Haibei (Happy) Wang

>> Thank you so much for having us.

Rebecca Knight

>> I'm Rebecca Knight for Rob Strechay. Stay tuned for more of theCUBE's live coverage of Audit and Beyond. You're watching theCUBE, the leader in enterprise tech news, and analysis.