The Accounting Podcast

There may be errors in spelling, grammar, and accuracy in this machine-generated transcript.

David Leary: [00:00:05] Of the tariff impact. Us importers paid $175 billion in tariffs in 2025. Small businesses paid $55 billion of that. Coming to you weekly from the OnPay Recording Studio.

Blake Oliver: [00:00:25] Hello and welcome back to the Accounting Podcast, your weekly roundup of news in the profession. I'm Blake Oliver.

David Leary: [00:00:31] I'm David Leary.

Blake Oliver: [00:00:32] We've got lots to talk about this week. David. The Iran war. There is an accounting angle. There's a financial aspect to it. This that actually could prove critical to the Trump administration, to the United States. I have a follow up on Bot Keeper, the collapse of Bot keeper. And Rico Moreno, the founder and CEO, spoke to accounting today. We will rip that apart. David, you've got something about ChatGPT 5.4, the latest model beating accountants. A developer has created a cloud skill for taxes. The Uber Uber employees recreated their CEO with AI. This and lots more. I have two experiences I want to share using AI to file a city tax return, LA city business tax return, and using ChatGPT to help me with a bank rec. So we've got lots to get to. David, let's first thank our sponsors.

David Leary: [00:01:29] Yeah, our sponsors this week we have uncW Kenan-flagler Business School on pay and cloud accountant staffing.

Blake Oliver: [00:01:36] Hey everyone. Let's face it, the job market is especially tough right now. But every industry needs. Accountants and accountants are always in demand. In fact, employment for accountants is projected to grow 10% through 2026 faster than most other professions. That's where U.n.c. Kenan-flagler is master of accounting program comes in. It's one of the top ranked Macc programs in the country, with 98% of students accepting a job offer within three months of graduation and earning more than those with just a bachelor's degree. If you're currently working full time, raising children, serving in the armed forces, or living halfway around the world, they're highly flexible. Mac program can also fit your lifestyle. You can choose their 12 month on campus program or their online only option, where you have up to 36 months to complete your degree. Plus, you'll join the powerful 46,000 strong UNC Kenan-flagler alumni network connections that will serve you throughout your career. If you want to set yourself up for a lifelong career, pick the Mac program with proven ROI. To see why you should get your Master's in Accounting at the UNC Kenan-flagler Business School, head over to The Accounting Podcast. That's The Accounting Podcast dot promo. And now we got to start with the war and the numbers. Let's set aside whether it's right, whether it's wrong. Let's set aside the politics. Let's just talk about the money, because in conflicts, money often decides who wins, especially if it drags on a long time. And the United States does not really have a lot of appetite to spend a lot of money on foreign wars at this point. So I wanted to highlight this article I saw in the New York Times. Uh, it is called Iran's drones cost a fraction of the US weapons shooting them down. This is by New York Times business reporter Farah Stockman. And she highlights or they highlight a big financial difference in the cost of the drones that Iran is flying out to bomb neighboring countries, and what it costs us to shoot them down with our missiles. These suicide attack drones cost basically.

David Leary: [00:03:50] They're little drones. There's nobody on them.

Blake Oliver: [00:03:52] Well, they're about 11ft long. Okay. They look like tiny stealth bombers and triangles, and they're 11ft long and they carry an explosive charge, and they just go at something and hit it and blow up. And they can be launched from a truck, which means that they are easier to conceal and hard to preemptively target. The Shahed one, three six can travel about 1200 miles, and Iran has so far fired more than 2000 of these drones since the US and Israeli attacks began. And the cost of these drones is just 20,000 to $50,000 each. They are built with commercial off the shelf electronics. Russia mass produces one like this for use in Ukraine. Iran may have produced many thousands. We don't know exactly. So these drones are 20 to $50,000 each. The Patriot system that we use, the Patriot Pac three interceptors, cost more than $3 million per interceptor shot. So we are spending if we use a Patriot system to shoot down a drone, we are spending $3 million to shoot down something that costs 20 to $50,000, and we don't actually have that many of them. Lockheed Martin made 620 of these in 2025, and that was a record. So there aren't that many of them and they're very expensive. So the cost ratio is not in our favor. It can be 10 to 1 and potentially 60 to 70 to one in terms of favoring Iran.

David Leary: [00:05:28] And we're flying $35 billion jets around, right? Yeah. The cost for these jets.

Blake Oliver: [00:05:34] I mean we can't shoot down drones with jets.

David Leary: [00:05:37] Yeah. Yeah.

Blake Oliver: [00:05:38] Right. We can bomb with them. But like if we want to defend our allies, you know, it's very, very expensive. So I wanted to highlight this on the show because when it comes to war, like money matters and costs matter, right? That's why Russia is having a problem right now. They are bogged down in Ukraine and spending a ton of money in Ukraine, and they can't make any progress. So it's become this like stalemate situation, a war of attrition and a war of financial attrition. And my concern is that given these economics, the US could end up in a similar situation.

David Leary: [00:06:14] Because, yes, because we built a military industrial complex to fight countries like Russia, who was also doing the exact same thing. Let's build $300 million jets. Let's fly them around up in the sky at each other, shoot each other down, huge bombs, big, huge rockets that can go miles and miles and miles. And this is like that next generation of warfare. Yeah. It's actually really scary when you think about the, the volume of it. Like, like if there's one, let's just say like a crazy bird was going to attack you, you probably could fight it off. But if you have a swarm of like 200 bees attacking you, it's a lot harder.

Blake Oliver: [00:06:47] Warfare has changed. And perhaps the United States military hasn't caught up yet. Now, we do have a newer low cost option from Raytheon called the coyote. It's estimated to be $126,000 per interceptor. But that's still multiple times the cost of an Iranian drone because those are 20 to 50,000. This is $126,000.

David Leary: [00:07:12] And how many of these have China built for themselves and at what cost? They're probably like 1500 bucks. They're just launching them everywhere.

Blake Oliver: [00:07:20] Yeah. So it's interesting to like, think back to wars that we study in history. And I remember when I studied World War Two in high school, learning how Europe totally underestimated Germany because of their tank power. Right after World War One, the European allies built a giant line of defense against Germany, like basically trench trenches and, uh, bunkers to defend against like an infantry attack. And Germany just went around it with tanks. So new technology makes old warfare tactics obsolete. And that might be a challenge for us. And it might be a challenge for us economically too, because the more we spend on wars which get very, very expensive, the higher our debt grows and the less credible American debt becomes in the international market, the more interest rates rise. And eventually we can't pay back our debt and we default on our debt, and we lose our status as the world's reserve currency and therefore our global hegemony in the process. All right.

David Leary: [00:08:34] The numbers are staggering. The the price difference is insane, right? Like it's, it's I've seen these articles like this. It's like 25 grand and they're making thousands and thousands of these things and we're building one jet at the same time. It's just. Ah, I don't know. You hope, you hope somewhere there's a strategy somewhere. You know, somebody's buying them through China and bringing them into the States so we can fight the same way. It's a little scary.

Blake Oliver: [00:08:59] All right, David, let's talk about, uh, ChatGPT 5.4. You've been reading about this. I just plugged it in to one of my latest workflows. I'm excited to see what it can do.

David Leary: [00:09:12] Yeah. So at the highest level from their press release or their blog post about this, I'll read this one sentence. We put a particular focus on improving GPT five point four's ability to create and edit spreadsheets, presentations, and documents. Sounds like a lot of work accountants do, right? Um, so they designed this model to be handled for complex professional tasks. So it just wasn't like, hey, it's just the new next generation of the model. This model was purposely designed and tweaked to do professional tasks more reliably, Efficiently and autonomously. It combines improved reasoning, coding, tool use and computer interaction abilities and can perform multi-step workflows across software, spreadsheets, documents in the web with less human intervention. And they may have less hallucinations as well. And so this is where it's interesting, some of their stats. So for knowledge work performance, it's beating or tying industry professionals 83% of the time. So if you give an accountant a job to do across spreadsheets and software in the web, it's essentially doing it 83% of the time the same way. And they measure this from a thing called GDP, Val. And that's a benchmark used to perform or to measure real world knowledge tasks. It covers up to 44 occupations across nine industries that can contribute the most to the GDP. So jobs that contribute to the GDP, they're focusing on automating those jobs and you know, and it can, it do a sales presentation, can it do accounting spreadsheets, can it, uh, do scheduling plans for medical or manufacturing diagrams? That's a type of, it's kind of whitewash, uh, white collar management work, if you want to think about it that way that it's doing so. And then on spreadsheets, it's, uh, versus the 5.2 model, it's scoring 87% versus 6.8%. So it's, it's moving the needle another 10%. It's getting close to that 90% success. Now on everything.

Blake Oliver: [00:11:13] I'm not surprised, David, because I've been using AI to start doing some tasks that are mundane but not easy. Multiple websites. Here's an example. I have a client. I'm embarrassed to say I still have a client. I have a client that I have not been able to shake since I sold my firm. It's a family friend. I still do the books, keeps me fresh. Uh, but of course I go against all the advice and I do I do this. Um, I used Claude Kirk to file the city of Los Angeles business tax return for my client. So what is that? It's. You have to get the gross receipts of the business for the year. You have to go on the city website. You have to put in an account number. All this stuff, it logs you in when you do that. And then you got to pay the tax and put in the account and routing information, right. And send the ACH. And so this is something that we couldn't do just months ago. And now Claude Cowork did it for me. So what I did was I said, I need to file the City of Los Angeles business tax for a friend's business, search my email for info about the account and help me file it on the city website. Claude has access to my gmail, so it went to the gmail and it searched for the email thread. It found the attachment in the email, which was the notice the postcard notice from the city of LA.

David Leary: [00:12:45] And first try. It's not like, oh, then you don't have to say like, that's not the right email. Try again. Did it find it right away?

Blake Oliver: [00:12:51] It found it right away. And it got the business details from the PDF. So it was able to use that information. Then it went to the City of Los Angeles website that was on the notice. And it put in the account information and it was able to successfully log in as the business. Now it got to the point where it needed the business's gross receipts. And so it asked me what to do. I said, go into zero. That's the accounting system for this company and pull it. So it navigates the zero website. It has me log in, it goes to the correct report. It pulls a PNL for the year. It changes the date range on the report and it grabs the gross receipts amount. Now the mistake it made was it pulled the accrual number. It didn't go to cash. And maybe that's not a mistake. Maybe that's just because I didn't tell it what to do, but it pulled the number on an accrual basis. So I wanted it on a cash basis. So I corrected it, but the number it pulled was correct.

David Leary: [00:13:55] I'll pause you right there. I've had to correct that on my business tax return with at least 2 or 3 different accounting firms. Like I have access to QuickBooks, they run the reports on accrual, do all my taxes. I'm like, wait a minute. Yeah. So this is not this is a human mistake as well.

Blake Oliver: [00:14:11] So it pulled the right number and then it entered it. It answered all the supplemental questions. It filled in the declaration with my name and contact info. It used the wrong email address. It used my business email instead of my personal. But that's also something that wouldn't necessarily have known to do. And then it submitted it and it got a confirmation number. And then I asked it to draft an email to my client confirming that I had done all this, and it did that, and it created this beautiful table with all the details like the confirmation number, the account number, the gross receipts, the exemption status, the $0 amount due. And yeah, then I just sent the email. So this is a task that maybe if you filled out a time sheet, it would take you 15 minutes, 30 minutes. Claudio just did it and arguably could have done it on the accrual basis. And nobody would have cared because it was under the threshold anyway. So this is where we're at now with AI agents. And that's why I'm not surprised to hear that a company that invested a lot in building agents that can do a partnership tax return, right, can do that because this off the shelf tool can just like file a city business tax return.

David Leary: [00:15:23] So I have a related story. This is a developer, this developer created a clod skill to do taxes. And he wrote a whole blog post about this. And he argues that AI can now handle surprisingly complex tax prep at Zero software cost. Obviously you have to pay for the AI, but zero software costs. Um, so he used Claude plus IRS forms to complete a 42 page 2025 federal return. After he abandoned TurboTax, he just decided he doesn't want to use TurboTax anymore. Um, he had a couple big core lessons in this. One was Claude is not best to use as a chatbot. Answering tax questions, but its real power is being a document analyzer for mapper. Calculator checker and a second pass auditor. Um, the biggest failure in this whole system for him though, was getting it to fill out the IRS Free File fillable form online. And when he discovered it's much better just to get the blank PDFs and tell Claude to just fill out the PDFs, and then he can then print.

Blake Oliver: [00:16:25] And mail it into the IRS.

David Leary: [00:16:26] Yeah, I'm assuming he mailed it eventually. Um, now he has all the details and all the things he did, but he has if you want to do this yourself, he actually has code to do this or the or the steps. So first he says, have code clod fill out the PDFs directly, skip the free file with the IRS. And he says, just download your current blank forms from the IRS, upload them to your cloud project and have cloud complete them, print and sign and mail. So yes, so he's physically mailing these. So we're using AI over here to send paper to the IRS.

Blake Oliver: [00:16:58] So what tax return did he do at 1040.

David Leary: [00:17:01] It was.

Blake Oliver: [00:17:03] I'm assuming it's like.

David Leary: [00:17:04] 1040.

Blake Oliver: [00:17:04] Tax return. Yeah.

David Leary: [00:17:05] Yeah a ten and a 1041 for two trusts that he had to do.

Blake Oliver: [00:17:11] David let's thank our next sponsor. And that is on pay. Are you tired of payroll headaches getting in the way of the client experience. You want to deliver manual workflows, creating bottlenecks, compliance nightmares and endless support calls that go nowhere. There's a better way for your team and your clients on pay is the payroll partner that accountants and bookkeepers actually love. Why? Because it's easy to use, packed with value, and backed by support that actually supports you. Their team gets rave reviews for being fast, expert and actually reachable when you need them. Onp handles the heavy lifting. You get a dedicated onboarding coordinator who sets up worker profiles and transfers year to date data from previous providers, all at no extra cost. There's seamless QuickBooks and Xero integrations. Eliminate manual journal entries, and they support any type of businesses you serve farms, restaurants, nonprofits, you name it. Onp can handle unique requirements without adding complexity. And Onp keeps pricing simple, too. Everything your clients expect from multi-state filing to off cycle pay runs. It's included. No hidden fees, no surprises. To book a demo, head over to The Accounting Podcast. That's The Accounting Podcast dot promo 1099 o n p a y.

David Leary: [00:18:23] Before we transition. But my favorite part of that story, though, is somebody on Twitter put a screenshot of the GitHub repository for this cloud skill and the replies in the comments from Tech's Twitter. You know, because obviously all the accountants and tax Twitter people are all in the thick of tax season right now. And all the comments were like, can I quit doing my returns tomorrow? Can I quit? I've been waiting for this my whole life. It's, it's really funny. Like the spirit of everybody's like, thank God this tool exists and I can stop doing taxes.

Blake Oliver: [00:18:55] One tool that we haven't talked about a lot on the show is Google Gemini. And I started playing around with it because I heard that the model got a huge upgrade and now it's equivalent to ChatGPT. And I have to say, yes, it is. It's amazing. And I love in Chrome that you can just open up Gemini in a sidebar. I guess this is like copilot in Microsoft Edge, and you can just chat with it about whatever page you're looking at. It's so amazing and fast, but I want to highlight a use case that maybe people don't know about. I heard that Gemini has video and audio generation capabilities. And it's like really good. So I decided to ask it to write new theme music for our podcast. David. So I just put in a simple prompt, I want to make new intro music for my show. The Accounting Podcast and Gemini said, that sounds like a great project. I've composed a new intro track for the accounting podcast that blends a professional feel with a modern, approachable vibe. The music is ready for you to listen to directly in our chat. Let me know if it strikes the right balance for your show. Would you like to listen to it? David?

David Leary: [00:19:59] Let's go for it.

Intro Track: [00:20:01] Welcome to the accounting podcast, A Journey through the numbers. The ledger is open. Now we're balancing debits and credits, navigating the fiscal year, finding the narrative in the data, making sense of it all the balance sheets, a work of art.

David Leary: [00:20:26] So bad.

Intro Track: [00:20:28] Fiscal responsibility.

Blake Oliver: [00:20:32] There we go.

David Leary: [00:20:34] Oh, boy. It's got a long ways to go.

Blake Oliver: [00:20:36] Okay, okay. But you know that arguably I didn't give it very much guidance. So I went through a few iterations and here's the the track that I came up with that I actually think is, is not horrible. Take a listen to this.

David Leary: [00:21:12] It's much better. Much better.

Blake Oliver: [00:21:23] Not, not, not terrible. I mean.

David Leary: [00:21:28] I don't know if I got a lot of people don't see this because they're not watching the live stream, and it's usually hidden before we go live. You play that music and I'm dancing to it. I'm getting fired up for the show. And like, neither one of these two you presented to me gets me a little like, ready to go. Okay, so I don't feel like I'm getting hyped up. I need to be more hype. Music you got.

Blake Oliver: [00:21:47] Okay, so here's what I'm going to do. David. I'm going to ask it, um, to, I'm going to ask it to like, I'm going to give it that context. Uh, this needs to be music that, uh, will hype David Leary up as he's like, walk on music. You ready to start the show?

David Leary: [00:22:06] Yeah.

Blake Oliver: [00:22:07] Like walk on music. Let's see what it comes up with. It's surprisingly fast. It's clarifying the objective. It is verified David Lowery's podcast role. And, uh. Oh, it.

David Leary: [00:22:22] Make sure you didn't make it up.

Blake Oliver: [00:22:23] I guess so it's thinking. All right, well, we'll move on. Come back to this, perhaps.

David Leary: [00:22:32] Well that's working. I can tell you about how Uber's employees have created a clone of their. An AI clone of their CEO.

Blake Oliver: [00:22:40] Okay.

David Leary: [00:22:41] So. And let's just call the CEO's name Dara K, because I'm not saying the last name in a million years. Um, Uber employees have created a clone of the Uber CEO using AI. And the reason why is they use it for presenting pitches and ideas before going to talk to this to him.

Blake Oliver: [00:23:00] Wait before they what.

David Leary: [00:23:01] They before they go meet with him in person. Let's say you have an idea or a slide deck or a presentation. You're going to go present to the CEO at your company. You want to practice it. And this lets people practice this before they get there to do and prepare for the questions that he might ask, etc..

Blake Oliver: [00:23:17] Okay. Got it.

David Leary: [00:23:18] So I see this and my, my $0.02 on this is I don't think this is going to eliminate the CEO's job, but I do see it eliminating a bunch of middle managers that are like gatekeepers.

Blake Oliver: [00:23:29] Yes.

David Leary: [00:23:29] If you work for a big company and you've done this, you get a chance to present the CEO and you have layers of managers making you we'll take that, that, that image and that, you know, that one box on the PowerPoint, move it over this way a little bit and round the corners. Don't use that yellow. Use a slightly different yellow. Like that's the kind of stuff you get in all these layers of management. And when probably if you practice it, you're not going to ever get that feedback from this AI chatbot because it doesn't matter. And the CEO is never going to say anything about those things. That's who this thread is. You're bypassing all these gatekeepers between you and the CEO now.

Blake Oliver: [00:24:05] And the gatekeepers might suppress the negative information that you might share with the CEO.

David Leary: [00:24:10] Yes.

Blake Oliver: [00:24:10] That is well, that's another issue in big corporations. Is that like the stuff that matters doesn't get up to the CEO because everybody wants to soften it so that it's.

David Leary: [00:24:19] Emperor's new clothes? Exactly.

Blake Oliver: [00:24:21] Yeah. That's interesting. Well, David, I have follow up on bot keeper Enrico Marino, the CEO of the now defunct Bot Keeper, did an interview with Accounting Today Chris. Gaetano wrote about it and Enrico explained more about the shutdown, the sudden shutdown of Bot Keeper. And I've got some questions, some follow up questions that I would like to ask Enrico about this. So Enrico Marino claims that the company could no longer stay in business. He claims that the realization that they were going to go out of business happened in only eight days, with the most intense chaos and pain occurring over 72 hours. I'm curious how you could not know until eight days before that your company is going to run out of cash.

David Leary: [00:25:23] Well, especially if you have this amazing AI. Bookkeeping system that should be giving you a beautiful cash flow report constantly. Yeah.

Blake Oliver: [00:25:31] And he was in San Francisco meeting with zero about getting special API access just two days before the collapse. Seems like he should have been working on something else.

David Leary: [00:25:42] But I think what happens in these situations, all it takes is you think you're going to refinance something and something's not going to get refinanced, and that person calling for payment of their note and then it all dominoes, right? Like in general, there's always this one. One, uh. Debtor is not the right word when you, the person you owe money to the creditor creditor, right? One creditor calls on some note. And if you don't have it, it just dominos your cash flow.

Blake Oliver: [00:26:08] Yeah, well, that's just being over leveraged, right? Yeah. Uh, okay. Here's a more troubling thought. 30 to 40% of revenue at bookkeeper came from only ten customers, 30 to 40% of their revenue. That's what Enrico said.

David Leary: [00:26:25] When you. Does it define customers like, oh, it was from ten firms that had thousands of small business clients using it? Or is it from like ten cuts, ten client clients?

Blake Oliver: [00:26:36] It sounds like they were accounting firms. And Paul Marino also blames the downfall on M&A activity. So he said that accounting firms have been consolidating and that what happened is that I guess bookkeepers, big customers started consolidating, combining, merging. They didn't realize it was going to happen and then they lost business as a result.

David Leary: [00:27:08] I don't follow this. So like it's what's happening is it's not so much consolidation. There's roll ups happening. So in theory, there'd be more clients to implement by keeping Quibron. Right? This doesn't make any sense to me.

Blake Oliver: [00:27:22] Well, I guess if one firm was using bookkeeper and merged with another one that wasn't, and the other one had another solution, maybe they would drop bookkeeper.

David Leary: [00:27:30] Okay. Got it, got it there. Okay.

Blake Oliver: [00:27:31] But still also doesn't quite add up to me because mergers don't just suddenly happen.

David Leary: [00:27:37] Like usually these companies take years to get a decision finalized. And I mean, we're talking about accounting firms here, right? They're not ripping out software like instantly.

Blake Oliver: [00:27:49] Now, I spoke with an I spoke with somebody familiar with the matter of Bot Keepers downfall. And I learned some new information that gave me, uh, like an idea of how this happened. Meaning how did how did Enrico Marino manage to raise tens of millions of dollars from venture capital? When he was essentially running a services business. They were doing outsourced bookkeeping and accounting in the Philippines with potentially some automation, some AI, although we really don't know how much. And I doubt that there was actually very much at all.

David Leary: [00:28:31] And we've talked about this before. Is these over when we covered this about blockchain and all the other ones is they were trying to get SaaS based multiples with the raises for a service business. And that's kind of what was Uber's problem. A little bit. Uber didn't realize how much of a service business they really were.

Blake Oliver: [00:28:47] Right. So how did bookkeeper manage to raise at a SaaS valuation when they were essentially a services business? I've always wondered about this because VCs aren't stupid. If they look at a company and they see that their service is in there, they're not going to give them the multiple A software business does. And bookkeeper went through multiple funding rounds, right? They did a seed round of 4.5 million, a series A of 18 million, a series B of of 25 million and a series C of 42 million.

David Leary: [00:29:20] Close to 90 million total. Right. Or 100 million total. Somewhere in there.

Blake Oliver: [00:29:23] The lead investor was. Yeah, the lead investor of the series C round was Grand Oaks Capital, led by Tom Golisano, the founder of Paychex. They had previous investments from Gradient Ventures. That's Google backed Greycroft. So how did they do this? Well, I dug into it and in August 2021, just three months before they closed their series C, Bot Keeper sold its Direct Business customers unit. Their internal bookkeeping company, their services business to an accounting firm called Benchmark Cloud Accounting. They then and this I have from someone familiar with the matter. And they acquired this firm. So they acquired this firm called Benchmark Cloud Accounting. They bought it. Then they sold their bookkeeping customers to benchmark cloud accounting and directed benchmark cloud accounting to purchase a multi-million dollar per year Bot Keeper license. That is how you turn service revenue into SaaS recurring revenue.

David Leary: [00:30:31] Companies are doing this right now. Yeah.

Blake Oliver: [00:30:33] Yeah. That's how you convert a services business into a SaaS subscription business with annual recurring revenue. And it's kind of genius. I mean, I got to give it to Enrico. Wow. So that's that's the bot keeper story. That's how they did it. We finally know the mystery has been solved.

David Leary: [00:30:55] Yeah. For my very first interaction with bookkeeper, they were at a conference. I was at Intuit and I remember they were at the conference. Big, huge booth. And they had the QuickBooks logo. We integrate with QuickBooks. We do stuff with QuickBooks. And at that time I'm running the QuickBooks App Store. I know who's connecting to QuickBooks through API calls. I'm like, they have no API connection to QuickBooks. I'm like, how are they doing this? What are they doing? And I kind of went over the booth and got a little demo. And I think I talked to Enrico actually. And like the first screen they showed me is like, oh, you invite in QuickBooks, you invite by keeper as your accountant, like, like through the ProAdvisor invite your accountant feature, right? And that was always like, that doesn't make any sense. Like if you're going to claim you have an integration, you need to use APIs. Yeah. Now eventually, I think they did do some APIs to move data around. But, but I was always like from day one, it was always presented as something that wasn't really right.

Blake Oliver: [00:31:53] David, I want to thank our next sponsor and that is cloud Accountant staffing. Are you tired of the endless search for qualified accounting talent? You're not alone. Growing accounting firms are struggling to find available and affordable team members when they need them most, and Cloud accountant Staffing has the solution with their revolutionary candidate portal. Unlike traditional staffing agencies that waste your time with sales calls, paperwork and deposits, the Cloud Accountant Staffing Candidate portal gets you instant access to highly vetted, qualified accounting professionals. While other firms make you wait weeks or months, Cloud accountant Staffing delivers candidates in just weeks, and their boutique support ensures you're getting quality talent that's both available and affordable. No waiting, no hassle, just top talent. Right now, the Cloud Accountant Staffing Candidate portal puts you in control. Browse live candidates, make selections on your timeline, and build your offshore team without traditional headaches. To find, review and book interviews with potential team members, all in less than ten minutes, head over to The Accounting Podcast dot promo. That is The Accounting Podcast dot promo code A.

David Leary: [00:33:00] Since we're talking about questionable behaviors, I have a story that is probably just starting to hit come out and probably will be a real story soon, but I want to at least cover it on our show. So it's a.

Blake Oliver: [00:33:10] Real story, David, because you're breaking it right now.

David Leary: [00:33:12] I'm not breaking it. It was broke. I just stumbled upon it, luckily.

Blake Oliver: [00:33:16] So it's going to be big.

David Leary: [00:33:18] I think it could be. It has a possibility to be big and bad. Um, so basically this Blake is are the earnings of companies audited by KPMG being leaked on Polymarket. So I saw a tweet that basically questioned this. And then there's a whole Substack written up about this now. Are you familiar with a lot of the prediction markets? We've probably seen the headlines. People bet that the Iran war was going to happen, and they placed a bet the day before and they think, yeah, they're trading at the white House.

Blake Oliver: [00:33:47] I saw this like some some representative was on Instagram pointing out that somebody made this huge bet on call XI, that the US would attack Iran and made like half $1 million. And it must have been somebody with insider information, like about the attack who did it.

David Leary: [00:34:03] And so everybody thinks a lot of these prediction markets are full of insider information. And the markets themselves, like Polymarket. I know I heard an interview with the founder. He's kind of okay with that because he looks at his prediction market as he wants it to be accurate. So if somebody has inside information and they place a bet that's going to help the market be more accurate, that's their point of view. Well, Polymarket lets you I think a lot of it's based on crypto and you basically are anonymous on Polymarket. And you can bet on all kinds of things sports, you can bet on election outcomes, whether or not we're going to go to war. But you can also bet on whether or not companies will beat their earnings in the next quarter. Oh, and so what this author of this post did, they dug in and they noticed that a group of Polymarket bettors, who are all anonymous, have been really good at winning their bets, betting on whether or not companies would hit their future numbers. And the one thing all these companies have in common is that they're all audited by KPMG.

Blake Oliver: [00:35:04] Wow.

David Leary: [00:35:05] And some of these anonymous accounts have only bet on the KPMG audited companies. They've never placed a bet on a non KPMG audited company. Now the amounts are still small. It's like 1000 here, 5000 here, 4000. Here is their winnings. So there's not like this massive like Bet and a huge pullout. But like there's inside information. People are willing to do that. Are they doing it on the real derivative markets as well? Right. And this is when I first saw this. It reminds me of conversations we had about audit before. How do we know there's good audits happening? I always said like accounting firms should be able to short companies as a reward for doing a good audit, discovering the bad stuff. And here, this is a good example of it, right?

Blake Oliver: [00:35:48] Well, so this is what this could be basically, is that it's somebody with access to inside information about these companies that works at KPMG or knows somebody who does, and they're getting that information before the street gets it, and they are betting on these prediction markets to to win. And they're doing it on a prediction market instead of the real stock market because the SEC will come after you for betting.

David Leary: [00:36:11] Yeah. You can't do it.

Blake Oliver: [00:36:12] With insider information on the stock market. But as far as I know, there's nothing illegal about doing it on a prediction market or, I don't know, maybe they're not aware of it. Like and well, the prediction markets are anonymous. The stock market isn't.

David Leary: [00:36:25] Yeah. So all the all they can do is analyze these people are making similar bets or the same bets, like what's in common, what's in common. Somebody's determined that all these companies.

Blake Oliver: [00:36:34] Yeah. All your stock trades eventually get tied to you as a person, right? It's like there's a there's a trail. But with prediction markets, you can be completely anonymous because doesn't it use crypto?

David Leary: [00:36:43] Yeah, it uses crypto.

Blake Oliver: [00:36:44] Yeah.

David Leary: [00:36:45] Wow. So it could be one person with multiple accounts. Nobody really knows. The only thing they know is these companies are audited by KPMG. Like this will probably be a bigger story if there was a war going on maybe. Right. And, and how like, how does the SEC prevent this? Like how, how do they find a firm for this? They, they, how do they, how they work backwards to prove it.

Blake Oliver: [00:37:06] Well, that's the problem. That's the problem with crypto, right? It's like the anonymity also makes it impossible to regulate.

David Leary: [00:37:13] Maybe we need a prediction market if KPMG is going to get in trouble for this. And we can bet against that.

Blake Oliver: [00:37:20] Let's talk about let's talk about Zapier. They released a new feature that I think is really smart. So in your Zapier workflows now you can insert a an AI guardrail step. So the feature is called AI guardrails by Zapier. It's a, it's an app that you can use inside of Zapier in your apps, like as a step, and it can detect more than 30 categories of personally identifiable information, including financial and contact data. Block workflows when sensitive information is found. Redact sensitive information before it moves to downstream systems. Identify prompt injection and jailbreak attempts. Screen for toxic content and analyze sentiment and return confidence scores. And I totally see this being a huge tool for accounting firms, because we have all of this information that we want to use with AI. But a lot of it is too sensitive to use with AI. That's the main reason most firms aren't doing anything with it.

David Leary: [00:38:24] So you put this in the step before your AI like, uh, sanitize this data before you do the AI.

Blake Oliver: [00:38:30] Yeah. So you could have the trigger and then you have the guardrail and the guardrail could be redacting sensitive information, detecting and redacting. And then you send it into the AI by Zapier step or wherever else you want it to go.

David Leary: [00:38:47] And if I remember correctly, 2 or 3 months ago, Zapier rolled out human in the loop, right? Yes. As well.

Blake Oliver: [00:38:52] Yes. They also have human in the loop steps. I haven't actually tried that yet because my, my workflows were all like automated.

David Leary: [00:38:59] So yeah, sign into the California sales tax website.

Blake Oliver: [00:39:02] Well, actually work has built in human in the loop. So when you ask it to do something, it'll first make a plan and then ask you to approve the plan. And then whenever it is going to do something like enter sensitive information, it will ask you, or if it's going to log into a website, it'll stop and ask you. It'll ask you permission to use certain websites like they've built in a lot of controls.

David Leary: [00:39:23] So now is this live? Did they roll this out or is this a blog post of a feature that's coming soon?

Blake Oliver: [00:39:29] This is live.

David Leary: [00:39:30] This is live. So you can start. So, so if you have client data being passed through Zapier into any AI tool, go add this step to your workflows.

Blake Oliver: [00:39:37] Yes.

Blake Oliver: [00:39:38] And I'm going to be giving it a shot and I'll let you know how it works out. Record numbers of workers are rating their 401 (K) savings. I spotted this in the Wall Street Journal. This is from data from Vanguard. And I bring it up because we've been hearing mixed stories about the economy. It seems like the economy is rebounding or the stock market is doing well. Sentiment has increased like the the AICPA just reported that CPA executives recession fears have dropped from 52% to 36% in just one quarter. And our optimism, the optimism of CPA is about. The US economy went from 28% to 39%, and we are more bullish on expansion plans. Now, more than half of CPA execs plan to say that their businesses are expanding. So we've got like good, good business economic news. But then on the individual side, we've got this troubling stat. And here's the chart. This is the participant use of hardship withdrawals from Vanguard.

David Leary: [00:40:47] Not just yet bleak.

Blake Oliver: [00:40:49] Oh here we go. There we go. This is the participant use of hardship withdrawals from Vanguard defined contribution plans. And you can see it starts in 2020 where it was under 2%. And now in 25, 2025, it's at 6%. So basically we've more than tripled hardship withdrawals. And that is when somebody takes money out of their 401 (K) to pay for a medical expense or to avoid foreclosure or eviction. And something you don't want to do because like that account is to save for retirement. And there can be penalties.

David Leary: [00:41:29] And the hardship withdrawals, withdrawal. It's not a loan against your 41K, right? Like the withdrawal just is straight up. You pay, you pay the extra 10% tax hit and you just take the money.

Blake Oliver: [00:41:38] I don't know, I have no idea. I've never done it. I don't plan to either. So you know, how are we doing? Um, it seems to be like a mixed picture. And when you look at inflation over the last, I don't know how many years it's been since COVID, right. We're basically up like cost of living is 20% up from just like five years ago. So you've got cost of living has gone up 20%. Salaries have stagnated, job opportunities are declining in white collar professions because of AI and some, you know, admin type jobs, right? So it's not a good market for workers. And our, our buying power has shrunk. And so that is.

David Leary: [00:42:29] That's creeping up, personal debts creeping up.

Blake Oliver: [00:42:32] And that's why I think the Republicans are going to have a really hard time in the midterms.

David Leary: [00:42:36] Especially when somebody's like, oh, we're, we're building $30 million jets to fight $50,000 drones.

Blake Oliver: [00:42:44] Yeah, yeah. And it doesn't help Trump. I think that like, Democrats are going to use the his promises not to get in to foreign wars against him because apparently now we're going for regime change in Iran, which is a quite a big shift, quite a flip, if you will. David, you had something about small businesses and tariffs. Follow up on.

David Leary: [00:43:07] That. So I saw an article on Yahoo Finance that small businesses, they're struggling to. They can't afford the legal the paperwork necessary to get their tariff refund. I guess that's what we're calling it, right? A tariff refund. Um, but I have some numbers on this. So of the tariff impact, US importers paid 175 billion in tariffs in 2025. Small businesses paid 55 billion of that.

Blake Oliver: [00:43:35] So that's a lot.

David Leary: [00:43:37] Roughly a third of all the tariffs were paid by small businesses. Um and with 97% of all importers are classified as small businesses. They just can't absorb the losses. To pursue this, um, and specific there's a backpack company, Dal, and they um. He doesn't see any possible way to get that money back. He just can't afford it. It just doesn't make sense for him to pursue this money. It's because of a wash.

Blake Oliver: [00:44:07] The legal fees to file a lawsuit to get your refund. And that's what you have to do right now, right, is you have to file a lawsuit. The legal fees are too much.

David Leary: [00:44:15] And so this is where I still think your idea was genius. Accounting firms spin up a service to help people get their refunds back, and you can buy. I have a better idea quickly.

Blake Oliver: [00:44:25] I have a better idea.

David Leary: [00:44:26] Use http PPP.

Blake Oliver: [00:44:27] A startup that uses AI to file the lawsuits to get the refunds. I mean, basically do the same thing that happened with Eidl and PPP, where all these startups popped up and they used tech to file for all of these. Uh, all the free money, right? I mean, maybe we could do the same thing. That's a $50 billion opportunity right there. And these businesses can't afford lawyers. So it's not like the lawyers are going to take that $50 billion.

David Leary: [00:44:58] Yeah. If you could somehow spin up a service that only cost 500 bucks or.

Blake Oliver: [00:45:03] Just took a percentage.

David Leary: [00:45:04] A percentage take.

Blake Oliver: [00:45:05] A percentage of their refund.

David Leary: [00:45:06] Like you or see.

Blake Oliver: [00:45:07] If you can. I mean, why not a startup could not a CPA firm, but a startup could. Uh, ibis in the live stream says class action lawsuit against the government. I don't know if you can. Can you do a class action against the federal government?

David Leary: [00:45:21] No, but I predict we're going to see one because of the tax data leaks. I do think we're going to see something eventually.

Blake Oliver: [00:45:29] Um, welcome to our livestream viewers, by the way. Uh, ibis great to see you, Allison. Ben Adler, Marcus BJ Glutz basketball. Awesome. Thanks for joining us live. If you've never joined us live, subscribe. Hit that notification bell icon on YouTube and you can get notified when we go live. We broadcast live stream most Fridays, but the time is unpredictable. So just stay glued to YouTube all morning long, potentially all afternoon.

David Leary: [00:46:07] You can tell we're ten days from the first tax deadline here because the, uh, the number of people popping the live stream is down a little bit this week, right?

Blake Oliver: [00:46:14] It's funny, busy season. We have a busy season slump and then it goes back up and we're about I mean, it's a it must be a busy, busy season because we're at like 20% of, uh, of usual.

David Leary: [00:46:27] Um, so I saw an article in Bloomberg tax from former IRS commissioner Danny Wuerffel, and it's about Donald Trump suing of the IRS. And he has a decent I like his take a little bit on this. So he brought up a couple issues that are as this goes through, the courts are going to be a possible issue. So if you think about it, the president in this case is the plaintiff And he's the defendant because the as head of the executive branch, he oversees the Justice Department, who normally defends the IRS in these court cases.

Blake Oliver: [00:46:57] Wait, so is it literally Trump be Trump?

David Leary: [00:46:59] Yes, that would be yes. That should be the episode title. Trump v Trump. Yeah. Um, now he does have the opinion that the court should probably, uh, appoint an independent counsel to represent the US on this. Um, but he ties a lot of this to back to IRS budget cuts, right? And if you think about this, um, as you cut the budgets of the resources, it cuts back on tech. You also, um, oversight, etc. also gets cut, right? And now we live in this world where, you know, there's constraining resources over here, but we're sharing data like how they shared all this data with Department of Homeland Security and Ice, right? And this is just going to open up a ton of new vulnerabilities because the safeguards aren't going to be there because they've cut the budget. And, you know, really at the end of this point of view is just like taxpayer privacy is fundamental. And this is what should bubble up from this court case. Um, policy choices about investment and technology and security have real world consequences. Like the budget cuts of the IRS caused this leak of Trump's tax data. And, and then lots of taxpayers tax data through these other sharing agreements. Um, and he says that these aren't partizan arguments. They are the fundamentals of a fair judicial process in a healthy tax system where the fundamentals are preserved, all Americans benefit. And I think I agree with this. Like, like we have to have a functioning IRS. It can't be compromised. We can't share data with the other government departments. We can't cut it down until it's it's since Reagan, it's just been crippled every year. We've talked about this on the show. Every year the IRS gets crippled more and more and more. And what happens? People like Trump's numbers get leaked.

Blake Oliver: [00:48:43] Well, speaking of crippling the IRS, the Trump administration has directed the IRS to terminate its collective bargaining agreement with the National Treasury Employees Union, which represents most IRS and Treasury Department employees. This follows Trump's Executive order 14251 issued last year, which removed collective bargaining rights across more than 20 federal agencies and departments. The Treasury Department had been temporarily, temporarily shielded from the order while the litigation played out, but that changed after an appeals court lifted a lower court injunction last Thursday. The Nteu is contesting the move. The union and says the IRS does not have the authority to end the contract on its own. So the litigation will be ongoing. But I could see this damaging morale at the IRS. They've already lost 27% of their workforce through layoffs last year. The IRS has been shifting some IT employees into customer service roles during tax season and requiring them to complete tax law training. So why would you be shifting IT employees to customer support or taxpayer support?

David Leary: [00:49:54] We're going to give you four weeks of training. Go start taking the calls. Yeah.

Blake Oliver: [00:49:58] I mean I guess they could do tier one support. Democrats have unveiled a bill to resurrect direct file. This is Senator Elizabeth Warren, your favorite David and Representative Brad Sherman.

David Leary: [00:50:12] Is this actually a bill or is this a letter about why it's legislation? Oh, real legislation.

Blake Oliver: [00:50:18] It's legislation. Uh, the Direct File Act, that's what it's called. The Direct File Act wouldn't just restart the program, it would make it a permanent IRS responsibility by requiring the agency to run a free online tax preparation and filing system. It would also block the IRS from signing agreements that limit its ability to offer free online filing. It would require annual reporting on usage, user behavior, and ways to improve access. It would push for integration between direct file and state tax filing systems, including grants for states and data sharing mechanisms. And it would speed delivery of third party income data to the IRS to help verify returns before refunds go out with the goal of reducing fraud. The IRS had launched a pilot of Direct File in 12 states in 2024, and it expanded to 25 states in 2025, before the program was shut down by the Trump administration and 94% of direct file users before the shutdown rated the experience as excellent or above average. That's extremely high. It was apparently very satisfying.

David Leary: [00:51:21] For a government delivered tech product. Yeah, you think about that for a government delivered tech product that's insanely high.

Blake Oliver: [00:51:29] And the full implementation of the direct file program was projected to save families up to $23 billion annually in fees, time and tax credits, money that they wouldn't have to pay to TurboTax or H&R block or your neighborhood tax preparer.

David Leary: [00:51:48] Yeah, that that million dollars Intuit donated to the Trump inauguration really paid off. You get a good return on that.

Blake Oliver: [00:51:55] All right.

David Leary: [00:51:56] I have an IRS story. Um, IRS says that meta, you know, Facebook owes $16 billion in back taxes.

Blake Oliver: [00:52:03] 16 billion for what, 16 billion?

David Leary: [00:52:05] So the IRS is claiming that meta has underreported about $54 billion in income and now owes 16 billion in back taxes and penalties. And this is all due to a case in 2010, um, when they transferred intellectual property, um, out to and they call it the meta's double Irish structure.

Blake Oliver: [00:52:24] Oh yeah. We've talked about this years ago before.

David Leary: [00:52:27] Um, so, so it's really starting to stack up.

Blake Oliver: [00:52:30] So the way it works is let me see if I remember. They, they, they sell the IP to the Irish entity and they license it back to the US company.

David Leary: [00:52:40] It's kind of similar to what you just said about barkeeper like you do this then for the money back.

Blake Oliver: [00:52:44] Yeah, it shifts. It shifts income from the US to Ireland. Correct. And then they keep that money in the EU. And as long as they don't repatriate it, they don't have to pay tax on it.

David Leary: [00:52:56] And the reason why it all happens in Ireland, because Ireland was smart years ago when they said every country in the EU can set their own rate. Ireland made sure their tax rate was cheaper than anybody else.

Blake Oliver: [00:53:06] Yeah, it's like super low, right? It's like what, ten? I don't know what it is. It was like super low.

David Leary: [00:53:12] And so this is starting to bubble up to the court. And the ripple effect of this one tax consultant estimates up to 700 billion in exposure across lots of companies. So when this settles this case.

Blake Oliver: [00:53:25] Wait so what's the argument? The argument is that Facebook did this improperly.

David Leary: [00:53:31] Yeah. They they apparently they based it on future earnings of some type instead of basing on actual earnings.

Blake Oliver: [00:53:41] Oh, so the way that they calculated the value of the IP. Yeah. Is it. Is it steak. I see. Not the fact that they did this. The actual Irish sandwich is still legal.

David Leary: [00:53:50] It's.

Blake Oliver: [00:53:51] Yes. It's just they they use the wrong valuation.

David Leary: [00:53:53] Yeah. But but apparently a lot of people play this game, right? There's a lot of companies playing this game. And so it's much bigger than the $16 billion.

Blake Oliver: [00:54:02] So the standard trading rate in Ireland. Profit generated from core commercial activities of a business is 12.5%. But a higher effective rate of 15% applies to large multinational enterprises. And that would obviously be Facebook there. Uh, passive income is actually taxed at a higher rate of 25%. Interesting. That's like the opposite of the way it's done here. Let's talk about Trump accounts. The AICPA is urging the Treasury Department and the IRS to automatically enroll eligible children in the new Trump accounts program created under the One Big, Beautiful Bill act. Aicpa wrote a letter February 25th and argued that automatic enrollment paired with an opt out option would improve access, reduce administrative friction, and prevent participation from being concentrated among families with greater tax awareness. So this is a interesting fact about Trump accounts that I didn't know, is that you have to actually go sign up for one. It's not automatic. So kind of sad actually. Like if you're a kid, right? Like the whole point of these Trump accounts is that we're giving money to the future. We're giving money to children. But if their parents are too stupid or ignorant or unaware of Trump accounts.

David Leary: [00:55:28] Don't.

Blake Oliver: [00:55:28] Have the resources.

David Leary: [00:55:29] To do it.

Blake Oliver: [00:55:29] Or don't have the resources to do it, blah, blah, blah, blah, right? Whatever it is, if the parents don't do it, then the kid doesn't get it.

David Leary: [00:55:36] Yeah, I could see forcing the auto auto enrollment making sense.

Blake Oliver: [00:55:40] Aicpa says the Social Security Administration could do it. Could do the enrollment because they already assigned Social Security numbers at birth. So just automatically give them a Trump account. Kind of makes sense, right?

David Leary: [00:55:51] Yeah. No, these are these Trump accounts. Is this money stored at the government or is this like, they'll put it in some other investment account at, quote unquote, bank partners or investment partners?

Blake Oliver: [00:55:59] I think it's the latter. I think it's it gets put somewhere like.

David Leary: [00:56:04] So I have to go like, like those.

Blake Oliver: [00:56:06] Educational.

David Leary: [00:56:07] Accounts for your kids, right? Or at least for 27 plans or whatever they're called for. 29. Yeah. Like that. Um, so speaking of people of organizations, I know you just talked about the AICPA has a recommendation. Nasba penned a letter urging the Department of Education to restore the professional status for accounting degrees. We talked about this before. They wanted to pull that back. And so in a March 1st letter, Nasba warned that excluding accounting from the professional category would sharply reduce loan access for graduate students pursuing CPA licensure. Nonprofessional programs can borrow up to $20,500 a year, but professional programs can borrow up to $50,000 a year. And Nasba argues that the change would create financial barriers for CPA candidates and worsen the profession's workflow shortage. Now, my $0.02 on this is like, this is one of those unintended consequences 150 hour rule. People wouldn't have to go pursuing their Mac or doing two extra years of education because of 150 hour rule and take out all these extra loans. But then I'm also kind of like, now Nasir's arguing that it's better that potential CPAs get into $50,000 debt instead of pressuring colleges with accounting programs to offer accounting degrees in a much more affordable way. That's what Nasba should be doing. They should be twisting the arm of colleges to not charge people 50 grand, to get two years of accounting education. That's what they should be doing. Like, why are you going after like justifying the debt. It's crazy to me. Go after the people ripping people off for this 50 grand.

Blake Oliver: [00:57:39] Thanks everyone, for tuning in to this episode of the Accounting Podcast. You can earn free continuing professional education for listening to this episode and all of our past episodes. Get the free earmark app, go to earmark app in your web browser, or download the earmark CPE app on the App Store or Google Play Store. You can sign up for free. You can earn one free CPE per week, and you can subscribe for the low price of $170 per year to get unlimited CPE every week, and access to premium content and support the work that we do here on the podcast and at earmark to make CPE accessible anytime, anywhere you are. David, always a pleasure talking to you. Thanks for thanks to everyone who joined us live and we'll see you around here next week.