John Gronen, CFO at Yooz, discusses driving efficiencies in procure to pay, fraud mitigation and the changing shape of the finance team

Speaker A: Welcome to CFO Insights, the leading podcast for finance professionals in disruptive tech, brought to you by the startup CFO community. I'm Guy Hutchinson. I'm the host of the podcast as well as being a tech cfo.

Speaker B: In this episode we're going to talk

Speaker A: to John Gronan, the CFO at Used the Purchase to Pay automation platform. John brings 25 years experience across finance, operations, sales and M and A, including previous CFO roles. In our discussion we talk through his views on AI deployment in procurement to pay and mitigating M against fraud. Many CFOs will be fascinated to hear how he shapes the growth of his

Speaker B: team and mentors others to help navigate essential career decisions.

Speaker A: John, welcome to the podcast.

Speaker C: Yeah, thanks Guy. Thanks for having me.

Speaker A: It's all right, it's all right. It's always good when you've got somebody on where, um, they've got a fantastic CFO story in terms of their career. But of course in your case you're also at ah, a business which is actually selling its product into other CFOs. And so you've got a boot in both camps. You've got the perspective of being in a tech business, but also, um, knowing that a lot of your customers, the senior stakeholder, is also someone like yourself.

Speaker C: Yeah, yeah, it's been kind of interesting. Um, you know, the career path has been kind of a little bit wild and then like you said, to now wind up in a company where I literally sell to my peers is kind of fun. Makes it really easy.

Speaker A: Yeah, yeah, yeah, I can imagine. Uh, and um, be good just to hear a little bit about the kind of things that you did early, ah, on in your career. Just like set the foundational steps towards the role that you have now.

Speaker C: Yeah, so my first job out of school was actually with a guy who was buying a bank in Arkansas and branching it to Little Rock and then opening a broker dealer, like an investment company and a trust company. And so at the beginning it was just the two of us. We wrote a placement memorandum, went out and raised capital, did due diligence on the bank and then ultimately purchased it. Um, so for a first job out of school it was amazing. Like short of going into, you know, venture capital or private equity or investment banking, you know, I got exposed to banking and how finance works, fundraising, M and A, due diligence, kind of all of that at once. Um, and such it was. Since it was such a small team, I wound up doing a lot of it. Very hands on, which normally as a young kid you wouldn't get to do that. And then once we had bought the bank, um, I was a financial and marketing analyst, but I was kind of doing a little bit of everything. And so we opened new branches. I was figuring out how do you buy an atm, um, and how do you set up bank software and all these things that made me kind of think outside the box and really set me down a path of always being curious about things and always wanting to learn more and, you know, figuring out how to research things on my own. Like I did the pricing for the investment company and the trust company. You know, I had no idea how to price those things. You know, I didn't have enough money to have investments at that point in my life. So it was, it was really interesting. And it was the early days of the Internet and so literally I'd spend my nights and weekends looking at how other companies did it, how do large banks or small banks or mid size price their services and then coming up with a schedule, um, that I thought made sense for us. Um, and so it was really a great job. And then once the bank was kind of up and running, um, the CEO loaned me out to a venture capital group and then I was writing business plans for entrepreneurs who wanted to raise capital, um, and helping them kind of go through that process and put together presentations and do the research and the financial model and all of that. Um, so it was just an amazing job right out of school. So I credit, you know, French Hill, uh, representative French Hill, uh, you know, for all of that, of setting me down that path. Um, it really was one of the best jobs I could have had, you know. And then I moved into telecom and then into staffing. And I spent about 11 years in IT, staffing and solutions, and then went to converged infrastructure, um, which is large data center, um, hardware that's kind of put together. So it was a joint venture between Cisco EMC and VMware. Their first project was iCloud. So they literally built the data center for iCloud. Um, and it went so well that they turned it into a business that grew like crazy, um, and ultimately was acquired by Dell. Um, and then I got into payments, um, and so I've been doing payments for about the last 12 years. Um, first with a company called VPay that was in the insurance space, kind of sitting between an insurer or a third party administrator and the provider, a doctor, a hospital, a clinic, whatever, um, and making those payments and moving data. And then went to a company called Siteline Payments that does payments in the gaming space. Um, so online sports books, casinos. And the interesting product there was they had a digital wallet with a physical card and a bank account behind it that would allow you to load a slot machine or purchase chips at a table using your mobile phone and then when you were done, cash back out to that phone. And the interesting thing in that industry to me was it was the ability to put your thumbprint on really changing an industry like the casinos, if you think about it, 100% cash business. You walk in with cash, you gamble, you cash out. At the end, you walk out with cash. So changing that to more of a digital environment where you could use a mobile wallet to load. Even though I'm not big on gaming or gambling, it was really interesting to potentially change an industry. And so I was really excited about that one. Um, and then I've now been at ah, use for the last couple of years and uses AP Automation. So it's a purchase to pay, uh, solution for CFOs and controllers, um, to automate the ingestion of an invoice all the way through the coding of that invoice approvals and then ultimately the payment of that invoice, um, using AI and machine learning to extract all of the information off of that invoice and take the processing of an invoice from about 11 to 12 minutes on average down to about 2 minutes. Um, so that savings for a finance department is just gigantic. If you can kind of think about that Math, like saving eight minutes per invoice on 100 invoices is 800 minutes. Now if you're 1,000 invoices or 10,000 invoices a month, that savings is just massive. And the ability to buy back time from your employees create some happiness for them. They're not, you know, mindlessly just keying in invoice after invoice. You can use them to do things that are more value add for the organization. Um, and then also the system helps prevent fraud, which I think we'll talk about a little later in this podcast.

Speaker A: Yeah, yeah, absolutely. That was a, uh, really, really interesting summary. So your, your early career, you actually did things that were almost like being a founder. You were involved in really early stages where the team is small and so if they're doing things, actually that's not at all a classical finance career where somebody might be in audit or working their way up the ranks in a finance department. It's really quite, quite, quite interesting. It sounds like it's really helped you expand your curiosity and helps you to get into some businesses doing things that are kind of really leading edge.

Speaker C: Yeah, it has, like, it. It. That job really set me down the path of always being curious about things and always wanting to learn more and not thinking inside the box, like, this is my job and this is it. It's always, how do I get involved to have the biggest impact in the business, and how do I learn more about new things and not being scared to. To research, um, things and figure things out on my own. Like, that was, like, key to me because I think a lot of young people come out of school and they get put into an audit job or an accounting job or a financial analyst job, and you're really kind of told this is the box that you sit in. Um, and you do that job for a little while, and then maybe you get promoted and the job kind of changes, but there isn't an opportunity to do something completely different than what you learn. Like, I had no idea how to buy an ATM machine, but none of you know, and so you go, okay, who manufactures ATM machines? You know, there's NCR and there's Diebold, so let's go figure that out. Um, and so it really, like, created this. This culture within me of you're just always being curious, always wanting to learn, and never feeling like I can't do something. It's more a matter of if I don't know how to do it today, then I'll go figure out how to do it today and I'll solve that problem. And, And I've tried to carry that through my career. And then even now, as I have, um, younger people that are working for me or working with me, I try to instill that in them as well. Of, uh, always be curious, always be learning. Um, you know, never think that you can't have a broader impact outside of your role. Just look for opportunities to do that. Um, and it's really, I think, you know, as I think through some of the people that have worked for me in the past, you know, it's. It's fun to watch their careers and where they've taken their career and. And they've kind of had that same mentality. And so I try to pass that along now to my employees.

Speaker A: Yeah, that is fantastic, actually, that, that piece where you've discovered that that's important part of your Persona helps you to work with really interesting businesses that are growing and really going somewhere and being able to identify what people in your team also need to be doing at the stage that they're at, in their Career to have those chances later on. It's a fantastic thing.

Speaker C: Yeah, yeah. And it's, it's just really rewarding. Like, I try to, I, uh, spend a lot of time mentoring and coaching kids that are coming out of either high school or college because you really don't learn in college. Like, what's out there. Like, I had no idea when I left college what was actually in front of me. Like, I just got really, really lucky, uh, you know, in winding up working for French and, and if I had probably wound up in a bank other than his, it would have been a probably completely different career for me. Um, and so I try to pass that along. So I spend a lot of time with college age kids that are coming out saying, look, these are the avenues that you can go into. Here's what accounting looks like, here's what investment banking looks like, here's what PE or venture capital looks like, here's what the corporate world looks like, and here's the career paths for each of those, you know, and kind of the pros and cons. If you're going to go into investment banking, venture capital, private equity, you know, first three years of your career, you might as well write off your personal time like it doesn't exist. And so the expectation is you will be at their beck and call for, you know, three or four years until you get enough seniority and somebody else takes that role. So, and walking through that, you know, here's how much money you can make in each of those different careers and the, you know, depending on what you want in your life, you know, you need to weigh those, um, those variables and decide what's best for you. Um, but it's trying to help kids kind of understand that. So now I've got probably five or six kids that I mentor on a regular basis of, you know, anytime they're talking about career changes or what to do or whatever, they give me a call and we'll like talk through it and just say, look, here's my opinion. You do what's right for you, of course, but, but, you know, having been around the block a few times, you know, this is what I think is kind of out there for you or what might make sense or, you know, how to handle a situation. And so it's kind of fun, it's rewarding.

Speaker A: Yeah, yeah, yeah, sounds it, sounds it. And you're quite right in that window where you've recently graduated and people are still super young. Right. There's a huge amount of luck in play in terms of what Happens with your early career. And if you've got some really experienced people around you, just help you to, uh, navigate those bigger decisions. That can be a really big differentiator. It's a fantastic thing.

Speaker C: Yeah, yeah. And it's, you know, generally speaking, your, your parents don't know. Like, if you're going into a career field that isn't what they did, they have no idea what's out there either. So they can' even really help you make those decisions and, and see the potential that's out there. Um, and then it, to me, there's always a little bit of luck involved with it too. Like you can, you can shape that, you know, by the school you go to or how well you do in school or whatever. But at the end of the day, you know, you could wind up in Deloitte doing, you know, audit work for the rest of your career and be incredibly happy. But you didn't know there was something else out there that might have made you happier.

Speaker A: Yeah.

Speaker C: You know, and just that opportunity to, to understand what's out there and make an educated decision is, is one of those things that I try to help kids with because I think it's important. Um, you know, otherwise you just have a very kind of myopic view of, of what that career path looks like because you're coming out in finance or accounting. Um, and you really don't see that there are these other things there. Like, I didn't know, you know, venture capital existed until well after I was at the bank. Um, you know, prior to that, I had no idea what it was or how it worked or how private equity worked or, you know, investment banking or anything else. It's something you learn kind of down the road and by then it's a little too late to shift into that career path. It's something you almost have to do straight out of school. Yeah, so, so that part to me is always kind of rewarding. I enjoy, you know, mentoring kids and, and trying to help them along and then kind of following them through their career and see where they go.

Speaker A: Yeah, I bet that is fantastic. And turning to, um, use where you are right now, uh, so use also must have had quite the journey. Right. Because the procure to pay setup. Right. Even if you go back to say, Netsuite 20 years ago, those were sort of rules based procure to pay systems that were maybe back then a bit clunky. Used to tell it what the rules were then. There's been a window of time where it's been machine learning, where like the platform has been able to learn some of the rules for itself. And now the last, whatever it is, two years, maybe more now actually, um, AI is a big feature. How has that informed thinking, uh, at use?

Speaker C: Um, it was one of the things that drew me to use. So when I had left Sightline, um, and was taking a little bit of time off and was starting to look for new roles. One of the things that caught my eye about use is they've really been doing AI for about the, since about 2010. Um, so they had built their own software to kind of look at an invoice and how do we extract data off of that invoice? Ahead of kind of the last couple years of, you know, taking Claude or, or Chat GPT and kind of layering it on top of a somewhat aging platform, um, and using that AI to read the invoice, they were kind of way out ahead of that. Um, and so they were using, you know, ocr, the optical character recognition originally, along with some other programs to, to extract data. Um, and then it just kind of grew from there to the point they had their own model. So one of the things that was interesting to me is they had then about 18 people who were PhD level engineers who do nothing but think about how to use AI, our AI, um, within the system to do more and more automation for the user. And um, and so it was interesting to me because they weren't kind of, you know, I joined in in 2024, middle of 2024. They weren't kind of just running out and grabbing Claude and throwing it on top and going, we use AI. You know, it was in the DNA of the company almost from the get go. Um, and it's a company that spun out of a company called itsoft that did document management. So if you think about what AP is, it's effectively document management. You're, you know, ingesting a document, storing it, you're reading some information off of it and then you're kind of layering all of the accounting controls you need like approvals on top of it. Um, and so it was really interesting and I like where the company's going. We invest. Uh, like I said, there's now about 24 people who are doing that AI engineering piece. So we invest quite a bit of money in just continuing to advance the product. We're a product focused company. Um, as opposed to just kind of throwing a on top and saying yeah, we're good, um, we continue to evolve that piece. Now we're using AI to look at Fraud, um, to understand the invoice better, to do line item extraction and PO matching, to be able to reconcile, um, a vendor statement, which is kind of a hassle for accounting groups. So when your vendor sends you a statement of here's all the invoices that we sent you and what's been paid and what's not paid, the system can actually reconcile that for you. You know, it used to be you had to go dig out the invoice and literally tick it off on a list, printed out the vendor, um, uh, the vendor statement and literally tick off each invoice. And so now the system can do that for you. And all of it's kind of geared toward, like I said, buying back time from your employees and creating a little more job satisfaction for them and allowing them to focus on more value added services for your organization.

Speaker A: Yeah, yeah, those are really clear leaps in terms of, uh, the kind of features that, uh, a finance team could just have at their fingertips and know that platforms like yours can essentially kind of do those things for them. And even I can recall kind of early in my finance career, um, joining as the finance person for a kind of vendor meeting where part of the meeting was just to compare, uh, your payables and their receivables or vice versa. It seems crazy now I think about that because obviously good technology will make mincemeat of that exercise, but people used to meet in a room to solve those things and kind of figure out why they didn't have the same numbers each way. Um, so I can see it's a huge step in terms of what Hughes has, uh, delivered. Um, but one thing that we were keen to touch on was actually the risk piece around fraud and, um, your perspective as a CFO on, um, how you need to be lifting the game in the finance team and dealing with an escalation where fraudsters are also using AI and finding sophisticated, um, harder to spot means to instigate or start a fraud. Be really interesting just to hear you talk about that, explain how you've seen those things pan out and the way that use helps, helps kind of businesses manage that.

Speaker C: Yeah. So we've seen over the last few years the amount of fraud in ap, you know, grow like crazy, just like it is in, you know, payments and everything else. Um, fraudsters are now using the same AI that we're using to fight them to create that fraud. Um, and so it's a little bit frightening if you think about where this is kind of going. The ability to use AI to take my likeness Take my voice, um, do a Microsoft Teams call or a Zoom call into my team and say, hey, can you wire money to XYZ company or this bank account? I need this done immediately. And if your team is not aware and not paying attention, it's very easy for companies to lose money. Um, and I think I read a stat the other day that on average a company loses 5% of its revenue to fraud. Um, and that's the average. So you're obviously going to have some companies that have no fraud and some that have a massive amount. And you know, there's a couple of huge cases, including Facebook, who famously lost a crazy amount of money to fraud, um, through ap, um, people turning in invoices that were fake invoices. And the company was so busy and growing so fast that they were just paying these invoices until someone stopped to say, wait a second, what is this? Um, and I think that's where AI is really powerful, like uses building AI into the platform or has built AI into the platform to be able to look at an invoice and start to help you decipher whether it's a legitimate invoice or a fraudulent invoice. So it can look at. Have some of the major things change that would allow fraud to happen. The remit to address or the bank account details or any of those kind of things. Has it changed from invoices that you've previously seen or that you're seeing from other customers? Um, so use, you know, obviously has about 6,000 customers in 52 countries. And so we have the ability to look across all those, those companies and see, you know, who else is paying salesforce.com for example, and do the remit to and the bank, uh, account information match what we're seeing from those or has there been a change for this customer and highlight those. Um, and the other advantage, like I mentioned earlier about taking an invoice from, you know, 11 or 12 minutes down to about two minutes, that's time the AP clerk can stop and do a deep dive on a new vendor. For example, as they're setting up a vendor in their ERP system, being having the time to go look and say, does this vendor make sense and what are we buying from them and are they legitimate? Do a little bit of research on it, um, to make sure they're legit. And then using the AI on top of that, as you're ingesting the invoices, you can see, do these invoices make sense? Have we seen it in the past? Have things Changed. All those are helpful to helping you prevent fraud. Um, I think AI will exacerbate the fraud problem for a while until everyone kind of starts using a tool similar to use either ours or know one of our competitors who's doing kind of the same thing, um, to help prevent that fraud. Now we, you know, have been doing it for a couple of years. Ours continues to expand. So I tend to, you know, be a little jaded and think ours is better than, than our competitors. But, um, you know, the reality is everybody's using it differently and, and, and I think everyone's using it for the right reasons to try and combat that fraud. And it scares me a little bit of, of where this is going. The, the lack of guardrails around AI and the ability for the fraudsters to gain access to that is a little bit frightening. Um, you know, the fact that I don't control my own image anymore. The second you, you know, do a podcast or, or you know, put something on TikTok or whatever, your likeness is out there, your voice is out there and they can use that small sample to, to, you know, create an image, a AI driven image that can converse with you and I. Now, fortunately, right now it's, it's not as good as it probably needs to be to, to be 100% effective. Um, you know, when you say something that the AI version of me, there's a hesitation while the system's thinking about what it needs to say back to you that, that you or I wouldn't have. Um, you know, you ask a question and I answer immediately. There's no, you know, period where you're thinking for, you know, 20 seconds or so or 10 seconds of what the right answer should be or researching it. So it's a little more fluid, but I think the system will get better over time. Um, so if you're not using good AI to fight the bad AI, you're already kind of behind the ball. Um, and so that's what I tend to tell CFOs that I'm talking to when I'm, when I'm doing something like this or at a conference. It's almost always, look, you need to be thinking ahead. This is a very real problem and it's only going to get worse. Um, so you need to be using the same technology to fight that fraud.

Speaker A: Yeah, really interesting. So I was involved in setting up broad, um, account quite, quite recently and of course with Claude, you can click the option on the paid accounts. The model doesn't learn from your private data, so Actually, the question I've got about users, because you guys have got hundreds of customers, there's a pattern occurring in one customer and they spot this. Like, actually, this is a fraudulent invoice case. Uh, we've caught it, but there's a pattern here. Does your AI, um, capture the data around that pattern so that the other customers benefit?

Speaker C: That's what we're working on right now. It's exactly that. The ability to actually highlight to other customers using that fraudulent vendor that, hey, here's a warning. We see that you have this set up as a vendor in your, uh, instance of use. Um, and we are seeing fraud with this. We've seen an invoice that was changed, and so somebody's mimicking the vendor or it's just straight up a fraudulent vendor. So, yeah, that's coming in the next few months. The ability to actually send alerts to customers. Um, because I think that's very important. Like, once you see it, it's, you know, the fraudsters don't just target one company. They, you know, target one company. It works. And then they go target a couple of hundred companies, um, you know, knowing that some percentage of those couple hundred companies are going to fall for it.

Speaker A: Yeah, yeah, yeah, yeah. That sounds like a fantastic feature. I think it'd be very well received. Uh, super interesting. And, uh, yeah, I think this fraud piece is quite kind of multifaceted, really, because as you hinted at earlier, um, people that would have been hired into a finance team five or 10 years ago might have expected a high proportion of their time to be quite transactional, just getting through, you know, AP, AR Bank Recs, uh, then are freed up. And it's a lot more interesting to be, um, on alert for things that might be suspicious and to be following up on items like that rather than just under pressure because you've got so many invoices to process that week. How has this helped you to sort of think about shaping the culture of your team? Right, because presumably, as you see those benefits trickling through, some things are happening that are different. You're building a different kind of finance team.

Speaker C: Yeah, it really is. It's changing the way I think finance teams work, um, completely. Um, you know, even at kind of our level, like at the CFO level, our roles have changed from where they were 10 years ago. You know, 10 years ago is still very transactional. Produce the PNLs, understand what it says, talk to the board or the investors. Now it's become a much more strategic role. How do you drive the strategy of the company to get a exponential return on whatever it is you're doing. Um, and so I find that even as myself, I'm looking more and more about how do I use AI to automate some of these processes. Um, and the goal is not necessarily to get rid of people. In some cases it's cost avoidance. You're not needing to hire an extra, extra person, um, but you're buying back time from some of those employees. You do have now to focus on higher level work for the company, um, which is more rewarding to them. Um, so personally I'm seeing a, um, reduction in turnover with our employees. I think the job satisfaction is much higher because we're using them to do things to kind of support that strategy focus. Um, so people that used to be doing just processing invoices manually, like, you know, I think back a couple companies ago, I literally would get a stack of invoices that was 2 inches high with checks paperclip to them, and I would have to initial the invoice and sign the check. You know, it took 30 or 45 minutes, you know, twice a month to do that. Um, you know, and then when the auditor showed up, you had to go dig up the invoice, dig up, you know, out of the bank a copy of the check and match them all together. You know, the interesting part of using an AP automation solution is you can have one system of truth for all of that. So you ingested the invoice, you have all the approvals of that invoice, you have the coding of that invoice right there. Um, since it interfaces with the ERP and pushes the coding to the ERP using the machine learning and everything, um, and then you have the payment of it all in one spot. Um, so you're really cutting that piece down and you're freeing up employees from having to support the audit for that. So again, just more time for them to do value added business. So we're doing a lot more cost analysis and vendor analysis, spend analysis, um, a lot more forecasting, um, for cash flow and things like that because you have more visibility into all of that. You know, AP automation gives you a lot of visibility into just looking at the reporting it generates, into kind of how's your spend been over a period of time. How would you expect that to be in the future? Um, so kind of the next evolutions of our system are to start getting into doing more cash flow forecasting for you, um, using the AI to actually build out based on your last six months of invoices. Here's kind of how we see this trending going forward. Um, and then as you're scheduling payments and everything else, it can obviously forecast that for you as well. Um, so I think it's really helping shape a more strategic finance group, um, and a less transactional one, which is an interesting change.

Speaker A: I can see that. And it sounds like a lot of these, uh, tasks that they're taking on that are new, like the financial analysis and the cash flow forecasting. This is partnering. Right? So they're less deep in the finance bunker. Uh, they're more out there with the business kind of partnering m on things that are more commercial. Does it mean that sort of finance people are going to become like slightly different Persona of finance person over time will be successful?

Speaker C: I think so. I think it's going to require that people are like you said, more capable of interfacing with their peers out in the broader organization. You know, my team talks to sales, they talk to the implementation team. You know, they're talking to the account managers. They talk to basically everyone. Um, you know, and when you're in finance or accounting, often you, you know, you do, but at a kind of lower level. Now we're having higher level conversations with them. Where are you taking the business? What are the things you're going to be focused on over the next three or six months that we need to kind of be thinking about and how do we fold that into as we're shaping the, the department and we're shaping the forecasts and things like that that we need to be thinking about. So there's a much tighter partnership I think now between finance and the rest of the organization than we've seen in the past.

Speaker A: Yeah, yeah, yeah. And that makes a lot of sense. It's just also more fun, I think in terms of how people spend their time at work. And uh, yeah, really interesting just to hear about that. And like, I think in terms of your kind of career journey, this piece around, you've had this quite entrepreneurial early journey. It's led you towards tech businesses. You're quite comfortable dealing with uncertainty, but that often comes in high growth. So that's quite interesting. Uh, seeing this transition in finance. I do wonder if the topic that we should sort of, um, end our conversation on should be AI. Right. Obviously maybe two years ago AI and finance was one topic. I think quite recently it's definitely not one topic. There's like 10 or 15 topics there now. Uh, and so we couldn't possibly cover all of it. But I'd be really Interested just to get your perspective on this moment. So we're in May 2026. For people that listen back on these things once and years afterwards, um, your, your perspective on how, how you see AI so transforming finance teams now and your, your sense as to what's important for somebody who might be like an up and coming cfo, what they might want to focus on.

Speaker C: Yeah, yeah. So I think we're at a very interesting point in time. Like, I think it's more interesting than probably my whole career, at least in the 30 years I've been doing this. There's more advancements in technology happening in the last, you know, two and a half years than I've seen in the 27 years before that. Um, and I think where it's going, it's going to continue to go that way. And so that part, to me is really exciting. It's also a little scary. Um, and so I don't think there are proper guardrails around some of how the data is being used and what companies are doing with it. And there's, there's a program on Netflix called the Social Dilemma. That is the programmers who wrote all of the social media, um, platforms out there talking about those platforms and there's, there's. One of the programmers said something that has stuck with me and I watched it probably three or four years ago. And it's. What he said was if you're not paying for the product, you are the product. Um, meaning if you're not paying for Facebook, using your data is the product that they're selling to make money, um, and then not picking on Facebook. It's like that for all social media. And to some extent right now it feels a little bit like AI is that way too. Um, so I think the guardrail that I think needs to be there is when you give access to any of the AI platforms, the commercial LLMs that are out there, you're effectively giving up the rights to your data. Um, and so unless you have it carved off into a private model, um, if you're using just, you know, Copilot or Claude or Chat, GPT or whatever, xai, whatever out in the market and you're uploading a file to it, you no longer own that data. They do. Um, and I think at some point that's going to become an issue for companies like you want to kind of, especially as a private company, you want to protect your financial data, your HR data, all of that. Um, so there's a really fine line in my mind around how you get the Fundamentals, full advantage of AI and all the power that that brings you while still protecting your data, um, and the confidentiality of that. Um, and so I think those pieces still kind of need to be worked out a little bit. Um, because I'm a little personally fearful that, you know, if I upload an HR file, for example, that has all of our employees and all of their statistics and use AI to do some analysis on it, that all of a sudden that's floating around in the ether and potentially being sold to someone else without my knowledge. Um, and so that's the piece that I think there should be a little more guardrail on. How do you protect and ensure privacy of that data? Um, I think once that piece is worked out, the only other piece that scares me is really kind of the pricing piece. Right? You can't spend the amount of money that those companies are talking about spending on data centers without changing your pricing model from where it is today. Um, because those economics just don't work. Um, and so I think at some point there's potentially going to be a, uh, resetting of the pricing model for those tools. It can't be free forever. Um, and so either they're selling data on the back end to make money or they're going to have to charge more on the front end to make money. And so I'm a little fearful with some of those tools that the more you embed those in your platform, um, and once you're kind of hooked on that and it's core to what you're doing, they have the ability to change the pricing and you have no ability to unhook that. Um, because it's now exactly how your product works. And it's one of the things I love about this company is the core processing AI is our AI. We don't use an LLM for any of that. Um, and so we're starting to use an LLM for more agentic. For the ability to ask questions and have the system return things we have segregated off our data from the rest of the kind of open world. Um, and so I think we're taking all the precautions we can. Um, but I'd like to see those strengthened a little bit at more of a legislative level. Um, but I think once that piece is figured out, AI is going to reshape how finance groups work. No doubt about it. Like, I have so many plans in my head for how I want to use AI. Like, you think about even the collections process. Like, do you really need a human being to call the customer to make sure they got an invoice. Or can you use an AI agent to do that for you? Um, you know, AI is getting to that point where it is kind of hard to tell whether it's a human being you're talking to or, or a, you know, AI bot that's doing that call out. Um, and so I've seen some software out there that does, um, like prospecting. It'll, it'll call you and ask to set a meeting to do a demo. And it's a completely. An agent that's doing it and you. It's really hard to tell that it isn't a human being. Like, they're really good. Um, and I think over the next two or three years, it'll get even better and more lifelike. Um, you know, on one hand that's great. On the other hand, again, it kind of opens the door to fraud and other things. But I think the ability to use AI as it's evolving to evolve the finance group, um, and the broader company that, that you might be working for is huge. Um, you know, using that to, in marketing, for example, to write, you know, LinkedIn posts or social media posts and stuff. I think that exists today. And it's just going to get better and better. It has the ability to read other posts you've done and learn from M. How you kind of tend to write things. Like, it'll almost pick up your diction, your, your pace, how you, you know, tend to phrase things. And it gets better and better as it learns. Um, that piece to me is really, really interesting. You know, it's a little bit scary about, are we all going to have a job somewhere down the road? Like, you know, that piece. I don't know. You know, you had Elon Musk come out and say, at some point in the future, the government's going to have to support a large percentage of the population because they won't have jobs anymore that's possible, like AI could take over some of those jobs. Um, and that piece is a little like, concerning, as you think about it, of, of what does that mean for society and you know, what did those people do? And you know, we'll always need, you know, a little less skilled labor, um, and specific things like, you know, it's never going to, I don't think, 100% replace the accounting team, um, because you want people at the end of the day to look at these things and make a decision about them and decide where it's coded and the system can't always do that. For you, um, because sometimes it's a judgment call. And I think at least AI in its current incantation can't make judgment calls like that. But I think it's getting better and better and so it's going to be interesting as to where it goes. I think the opportunity for fraud will continue to get more and more, um, more and more serious for us. Uh, I think is probably the way to put it like it. That ability to fake someone out and get them to wire money out of the company or change a bank account without stopping to think is this the right thing to do because they truly believe it's me, um, is going to get worse. And so, you know, I'm telling people on some level you have to go back and get a little old school with that. Um, so we have a password before we can wire money out of the company. So if, even if it's me on the teams meeting asking you to wire money, the finance team is taught to ask the question and I have to give the answer before you can validate that it's me. And so we kind of went a little bit old school with that, um, to prevent these kind of AI deep fakes from taking hold here. Yeah, and I think that's one of the things I always recommend to other people is, look, it's just a good idea. Um, it's the one way you can validate if you have a code word just that your team knows and no one else. And I ask you what is that code word? But you can't give it back to me. You know, it's not you. So, so it's one of those ways to just validate. It's a little, you know, 1960s, but it's a good way to validate that you're not an AI, uh, deep fake agent.

Speaker A: Yeah, I see more of that. So I think the points you've made there are, uh, absolutely, ah, bang on in that there are situations where you want it to be automated. Like you might want to an early demo of a product done by an AI agent because you don't really need to meet the person at, ah, that point, you just need to tell it which features you want to see and it goes through the process of doing that and that gives you a sense as to what you're looking at. But maybe if you went to a second meeting for that vendor, uh, you'd want to meet some of the team to get a feel for the company. Right. And so it's this sort of this more discerning thought process about which tasks really need to be done by a person because people really count, and which tasks can be done by the bots because this is just a really routine, straightforward thing and probably not much will go wrong. Um, and really interesting just to hear you explain it in that manner. John um, John, this has been really fun. I've enjoyed having you on the podcast. It's been a superb conversation. Really fascinating to hear about your career and learn more about yous. Um, thank you very much for joining me. I wish we'd had more time, but it's been a great conversation.

Speaker C: Yeah, thank you for having me guy. I really enjoyed it.

Speaker A: Mhm.

Speaker B: I hope you enjoyed our discussion today. We're really proud of the podcast following we've built up as we run our um, podcast in conjunction with the Startup CFO community. We're able to access many of the experts who are changing the face of the modern finance function, allowing us to hold these discussions, playing our role in shaping the modern finance leader and informing career journeys. In the age of AI, we're seeing the most dramatic changes in how CFOs apply themselves to supporting high growth businesses. It's a time where peer support will add more value than ever before. And lastly, if you're not in our group already and want to join, just go to StartupCFO Tech and click to apply to be part of our exclusive community offering

Speaker A: it.