The Accounting Technology Lab Podcast – Quantum Tech in Accounting – Oct. 2023

Technologists Randy Johnston and Brian Tankersley, CPA, discuss how a new generation of computing will impact accounting firms.

Use the podcast player below to listen.

Transcript (Note: There may be typos due to automated transcription errors. Also, due to the intro to the podcast, add apx. 10 seconds to the time stamps below.)

SPEAKERS

Randy Johnston, Brian F. Tankersley, CPA.CITP, CGMA.

Randy Johnston  00:03

Good day. Welcome to the Technology Lab. I’m Randy Johnson with my co host, Brian Tankersley. Brian actually conceded it was okay to talk about today’s topic. I’m excited. He’s kind of with me on it. But I’m gonna get him convinced and you convinced why quantum computing and accounting is such a big deal. Now, first friends, you have to remember I’ve got kind of a strange background because I actually have a light physics degree. So I’m actually a physicist, in addition to being computer science and accounting and social science and lots of other background. I love learning new stuff. And I recently completed a book which I recommend that you read called Quantum supremacy by Dr. Michio Kaku. And, you know, I have been following quantum for a long, doggone time. And I know that when quantum computing arrives, it will be a new generation of computing. Now Brian and I first started working on this together a long time ago. But we did a complete emerging technology session in 2016 and 17. That took about 1500 hours to write for 100 minute CPE session. It was

Brian F. Tankersley, CPA.CITP, CGMA  01:27

and honestly no human being could get through the slide deck and 100 minutes. Okay, so it was, it was a it was a heavy lift to get that one done in four hours instead of instead of two.

Randy Johnston  01:39

But you know, I talk fast, and I’m excited about it. And I want you to be excited.

Brian F. Tankersley, CPA.CITP, CGMA  01:44

There’s not enough I’m going to some days.

Randy Johnston  01:48

Some days there’s not. But here’s kind of the deal friends, quantum computing is close. And in fact, the way to think about quantum computing is I have claimed for quite some time that when we get to a 500 qubit computer that will be able to outperform any computer in the world. You know what the current fastest computer in the world is just down the road from Brian. But right now Intel’s tunnel falls chips are 12 qubits, Google’s bristle cone chips have 72 qubits. And the current IBM machines, I think have 480 cubits. Remember I said 500 cubits is my my tipping point. But we’ll talk more about that. But what’s happening here is the traditional computers that we run on today are commonly referred to as von Neumann computing, or Turing computer designs. And they’re basically ones and zeroes. And so everything is translated into ones and zeros. And we’ve built a whole industry around this, this approach. And that’s the way computing has been done since 1946, I believe is the right date. Now, as it turns out, these new computers instead of being made with ones and zeros are made with qubits. And conceptually, the thing that I’d like for you to begin to think about is that the one and zero was a state of honor off with qubits, there are infinite states. And the best way for you to picture this is if you’re going to walk across a room from point A to point B. If we do traditional Euclidean geometry, you might remember from your education, this shortest distance between two points is a straight line. In quantum computing, there are an infinite number of paths from point A to point B. Now, if you start thinking about there’s an infinite number of paths, that means that the state of a computer can do almost infinite calculations on a qubit. And if we start putting a bunch of qubits together, we can do some real complex stuff. Okay. I hope you’re still listening, because I want to talk now about the way it works in accounting.

Brian F. Tankersley, CPA.CITP, CGMA  04:15

So wait a minute start. So so so we can be so we can have the ledger simultaneously in balance and out of balance and an infinite number of ways if we’re doing this, or we really would be optimization solutions here.

Randy Johnston  04:30

Well, that would be true, Brian, and we won’t go into the Schrodinger cat, you know, analogy. But yes, you could have simultaneously in balance and out of balance, you know, the old accounting joke, how much is two plus two? What do you want it to be? Right?

Brian F. Tankersley, CPA.CITP, CGMA  04:51

Or the other one, there are three kinds of accountants those that can count and those that can’t.

Randy Johnston  04:56

Yeah, that’s that’s the that’s the zone we’re working in here. But You absolutely took it to the right spot, because we’re going to have the capability of doing all sorts of new optimization models, all sorts of new predictive analytics, and all sorts of monitoring of data. In other words, the billions, some estimates, say, up to 40 billion Internet of things here in the planet, can all be modeled pretty close to real time with quantum computers. Today, there is way too much data. Now, if you’ve seen Brian and I speak over the last few years, we talk about a lot of different technologies. And we razz and joke about things like toilet seats. And, you know, some of the new toilet seats that he and I saw at CES this year, actually monitored all sorts of body chemistry functions. And if you

Brian F. Tankersley, CPA.CITP, CGMA  05:54

we’ve been looking at toilet seats at CES for at least 10 years together. And this is the first time they actually solved a problem that I could see having in my life, unfortunately, came a little too late for my parents. But you know, I think it’s just in time for your mom and Pam’s mom, potentially, because, you know, your mom is celebrating your 100th birthday next week. So, you know, and I think it’s, it’s pretty exciting that you can monitor some of these, you know, just by the fact that people use the toilet, you can monitor some of these statistics about their heart rate and oxygen concentration and things like that.

Randy Johnston  06:37

Yeah, and even beyond that, Brian, you know, I’m very sensitive to cancer issues. And most of the cancer tests in the market today require millions of cells, you know, I don’t want to be inaccurate, I believe it’s 2 million cells to be able to detect that there’s a cancer problem 2 million mutant cells. And in quantum computing, they’ve already figured out they’ll be able to detect cancer mutations in the 50,000 cell range, and you could do it in the toilet.

Brian F. Tankersley, CPA.CITP, CGMA  07:13

And I will tell you that the, the thing that the other thing that you need to know, this was announced yesterday, actually, Microsoft has a deal with page AI, where they’re going to look at all of the radiological films, so all the CTS, all the X rays and all the PET scans, and they’re going to integrate those and kind of be a tool for for medical practitioners. So just like that kind of stuff is happening in the medical profession, we expect this kind of stuff to happen in the accounting profession. And in unfortunate, fortunately, or unfortunately, it’s going to be available to the big guys on the side, on our side of fraud, and to the bad guys on our side of fraud. So, you know, let’s let you know, again, if we if we go through and look at some of these optimization problems, you know, if we, if we have all of these models that tell us what’s good and bad and ugly, and then we use quantum to solve for the optimal, optimal amounts to put in a fake invoice and those kinds of things, we can get some pretty convincing stuff, and honestly, potentially even some pretty convincing evidence coming from the bad guys. So it’s very important that we don’t tarry and we don’t mess around with these, this technology when it comes in. When you

Randy Johnston  08:35

and I want to extend this, it’s going to affect us as accountants directly, but my forecast is every industry will be affected. So not only will it affect you in your practices, it will affect every one of your clients in their businesses. I don’t think there’s an exception, health care, construction and so forth. Now, we want to use since Brian went down the bad day, bad guy approach for a minute and, you know, I learned from him, you know, the phrase, the bad guys are getting better faster than the good guys can protect us. And we recognize recognized that there was an event in the past Why 2k That’s when the computer programs quit working for December 3119 99. And a lot of people thought that that was a fake event. But what happened is a whole lot of computer systems got updated. So it turned out not to be as big a deal when it arrived because a whole lot of remediation took place. But we knew the date of y2k was December 3119 99. Along that same lines, there is a new phrase why to queue sometimes called Q day when quantum computers can break encryption. And what’s happening right now is spy agencies and bad Actors are warehousing data that can be decrypted. And this act is known as quantum hacking.

Brian F. Tankersley, CPA.CITP, CGMA  10:10

spirits that are in the Salt Lake City area down on the Silva silicon slope, the NSA actually has a data center in Bluffdale, Utah, that is one of the largest data centers in the world that is storing this data. And a lot of the computing that’s done on that data actually takes place at the Oak Ridge National Labs Center for Supercomputing just a few miles down the road from me.

Randy Johnston  10:37

So there you can see this. Yeah, go ahead. Right.

Brian F. Tankersley, CPA.CITP, CGMA  10:41

So there’s, there’s there’s a lot of this, right now quantum is primarily provision of the largest companies, so the largest global enterprises, and nation states. And I think you’re when you just like when you started using computing, you used it in a shin a kind of in a timeshare basis. And I think the way you’re going to use it is going to be primarily through these data centers, because the requirements for these computers are such that it is it is completely unaffordable for you even consider them right now.

Randy Johnston  11:15

Yeah, but you know, along this line, Brian De, you can get trial loads on IBM Quantum computers in their timeshare, as you just said, but I want to remind our listeners that we’ve gone from huge room computers, mainframes down to mini computers down to personal computers, down to where our phones, the smartphones that we’re carrying with this are more powerful than the largest computers available when we put people on the moon. Well, the key thing I want you to hear on the bad actor side is all known encryptions are going to be broken with quantum computer. That’s what y to q or q day is about. And the bottom line here is that SSL stops working all the BitLocker stops working, everything we’ve got to secure our systems will stop working. Now, the National Institute of Standards has announced competition for quantum resistant algorithms. And they have announced the four candidates after a six year competition. So this has been going on, you know, for quite some time, the problems been known for 10 years plus, and the four algorithms that will be candidates are called crystals, Kyber, crystals, dilithium, Falcon, and sphinx plus, now, it’s computer science. People are trying to test these algorithms to make sure that they’re secure. Because although I think it has no value at this point, the Biden administration has set a deadline for new encryption by 2035. But the algorithm and the quantum computing capabilities that can break decryption, some experts are saying could happen as early as 2020, by 2029, using Shor’s algorithm. Now, I’m thinking that that 2029 date is late, I’m thinking we’re gonna see some of this first breakage occurring in 2025 and 2026. And it may be happening now, and I just don’t know about it, because you know, I’m just a simple guy from Kansas. But you know, the bottom line on this stuff is, it’s a real threat. And it’s a real promise. And you know, so when I start thinking about quantum, then there are applications that are very notable for an accountant. Among those are optimization. You know, we can do portfolio optimization, risk management, supply chain management, and we can analyze vast amounts of data, data amounts that so big, I’m not even sure calling it big data is accurate. There, you can just capture so much and run it. And basically, accountants can make informed decisions and improve business important performance by looking at these things. Further, it is hard for me to describe large enough data sets for you. Now, again, I tell you, I’m a simple guy. But I remember when McDonald’s was first starting out, and they were talking about millions served, and I just couldn’t wrap my head around that nowadays. It’s billions and billions. And, you know, I actually got a good smile from Dan hood, the editor at accounting today, who said, you know, I really liked McDonald’s burgers and I worked at McDonald’s when I was growing up and it’s like, well, I’ve talked about Ray Kroc and his processes for a long time and I get it, but I can’t wrap my hand around millions and millions read alone billions. And we are going to have large datasets that are beyond pay. bites. So just like we talked about in the generative AI session, Intuit has announced that they’ve got 60 petabytes of data that they’re working against. And that’s going to be pretty trivial in today’s world. So Brian, you appropriately started talking about fraud, fraud, creation and fraud detection? Do you want to kind of speak to that for a moment?

Brian F. Tankersley, CPA.CITP, CGMA  15:22

Well, I think the I think part one of the one of the challenges that we all have is with fraud detection, is that we look for these unusual patterns. And you and I have a few, you know, we have unusual patterns, you know, you and I are used to using descriptive statistics, you know, mean, standard deviation, you know, again, we look at things, you know, we expect, you know, I expect my, my cable internet bill to be $65 a month and my fiber Internet bill to be $125 a month. And if it’s plus or minus $10, I’m, I know that it’s outside outside of what it should be. So I think when we think about fraud detection, the thing I think you’ve got to we’ve got to worry about here is that is that the bad guys now can create these fake these fake transactions, that that have all of these, all of these numbers in here, you know, after, after Enron, Mark McGeary, I think published an article in the CPA journal about ratios that, that inform a inform a legitimate business. And we saw after after Enron, subsequent frauds, incorporated those ratios, and so those ratios were no longer predictive. Because the bad guys were already using them in those subsequent frauds. And so with this, you know, the, the standard benchmarks that you’re looking at, become become important here. Now, let’s talk for a second about that 60 petabytes that that Intuit has, you know, just for perspective, I have, you know, a small, I have all my entire music, pictures and audio, everything, all the content that I own is on a five terabyte removable hard drive, okay? And it is a ridiculous amount of content, it is literally hundreds of movies that I’ve ripped to video, it is literally 1000s 10s of 1000s of mp3 ease. It is, you know, it is all the digital pictures I’ve taken ever, plus all the ones I’ve scanned, and it’s about four and a half terabytes, okay? Now, a petabyte a terabyte is 1000 gig, okay? And so that means that a petabyte is a million gig. And so it into it here, we’re talking about 16 million gig of data. As we’re, as we’re going through here, and and it’s, it is absolutely amazing to think about the time it would take just to retrieve it from solid state drives or discs, to do these things. And the beautiful part about quantum now is it gives us a way to look at all that data and to make sense out of it with these optimate with these tools. Were in the past, it was just not possible with the Von Neumann, you know, floating point, floating point computers.

Randy Johnston  18:20

Yeah. And so Brian, you know, you actually segwayed over into another important topic. And a lot of our models today are running with these floating point chips like Nvidia, which contributed to their, you know, trillion stock market valuation and so forth. But I don’t think the floating point computing things are going to remain important if quantum kicks in. And at the risk of bending our listeners minds just a little bit further. Generative AI large language models running on top of quantum computing have even more capability than they do on von Neumann floating point environments. So I don’t want to give a ratio unnecessarily. But I’m thinking that they could be anywhere from 10,000 to a million times more effective on quantum than they are on today’s platforms. Now, if you just step back for a minute and say, Could that happen in three to five years? Yeah. And can you imagine generative AI being a million times more effective? And if I go back to your fraud claims? Can you imagine how good the bad guys can get with their fraud creation along the way with the AI advancements, and how hard it is for us to detect fraud in that situation?

Brian F. Tankersley, CPA.CITP, CGMA  19:48

Well, you know, it’s AI is already having huge impacts on on things like that. I read yesterday that Google is actually going to start requiring political can In this political season, to disclose when they’re using AI based deep fakes in their ads, that are that are that are video and audio of things that didn’t actually happen that was made up by AI. So we are living in interesting times.

Randy Johnston  20:17

We are indeed, in fact, that signature inside those Google pieces, you know, is also being done by Adobe with their platforms. And I think those have to be done. You know, something that’s Gosh, old school, like 1980s old school, is every laser printer actually has a digital signature, but most people don’t know that it does. And that was put in there. So you can see which laser printer created any page that’s ever generated. But that’s, that’s probably a topic for another day. But to just close up on some pretty important things, cryptography, as you’ve already heard, will get broken. And so it’s going to be harder for us to protect sensitive information. And you know, when I think about tax strategies, and tax planning, we can optimize tax strategies by analyzing large data sets. And we already know that some of the tax planning tools, at least two of them have put artificial intelligence in there. But now, think about quantum powered AI doing that. Well, you know, again, we recommend, there’s such a fine read that to get. Dr. Michio is Cockos book, I always say his name wrong, but I love his topic. And you know, there’s going to be things in there that you will see that it’s like, huh, how is that possible? But Brian knows that I’m a kind of a weird science, you know, background book reader, and I like genetics and have assisted a, you know, genetic analysis company. And, you know, I watch a lot of things going on in genetics. But, you know, the way that our 20,000 genes or so the makeup, our bodies have been put together, quantum computing can help analyze that. And at the risk of topics that, you know, can be on things that are health and life related, things that are energy, labor related food, you know, how do we optimize this, because let’s face it until we got, you know, the non organic chemical fertilizers, there is a concern we’d run out of food. But, you know, there’s claims that quantum computing can help us cure global hunger, which is very interesting. Not to mention doing gene editing, as I’ve already alluded to, there is some take that the analysis of telomeres, which could make people immortal, is actually possible in this new realm. And there’s also things about fusion, that can be solved in quantum and simulating the universe. And I just gotta tell you, when I started thinking about this, Brian, I do not see a single industry that won’t be affected by quantum capabilities. And, you know, as I start mulling this over, I don’t very often wish to be younger. Because, you know, night like mum who’s celebrating that 100th birthday, as you alluded to earlier, I’ve always told people I want to live to be 102. Well, now I’m thinking I wouldn’t mind being young enough that I could see this whole thing unfold, which is probably going to take 50 or 60 years, I figure, but there’s going to be such stunning developments in the next five to 10 years. So like, we’ve warned you about generative AI, you need to be aware of what’s happening with quantum computing.

Brian F. Tankersley, CPA.CITP, CGMA  24:02

I completely agree with you, Randy. And I think the, I think, you know, I’m gonna quote AICPA CEO, Barry Milan son, who’s who’s famous for this statement. He has said, for for at least five years at the annual thought leader, gathering engage, that, that the pace of change that you experienced today, even though it’s more than you’ve ever had in the past, is the slowest things will be in the future. And I think, you know, when we look at the combination of generative AI and quantum, you know, we’ve been looking at Cloud and how cloud was going to change tax and accounting and how things were going to be, how problems are going to be solved different and I feel like we’ve entered act two of that now if it was a play, and we’re starting to see the solutions actually come out for generative AI and again, we’re still in Act One. And for quantum, you know, the solutions haven’t come out yet for that, but they are, we are close getting toward the end of Act One in, in quantum, and act two is coming a lot faster than we expect it to be. So it’s very important to think that you pay attention to what’s going on in the market, because there are a lot of there gonna be a lot of very interesting opportunities that are going to change the way we think about computing the way we use computing.

Randy Johnston  25:28

Yep. So Brian, I don’t know that I convinced you to be as excited about this as I am. But I for our listeners, very rarely do I say this is almost mandatory reading for accountants, but the book, Quantum supremacy, how the quantum computer revolution will change. Everything is written in such simple English, that you will find it very, very understandable, readable. And it will just open your mind to think about all the new service opportunities for your clients, and all the new opportunities in accounting. Because I’m looking at this saying, Wow, cloud computing, not particularly important in the scale of things financial reporting, predictive analytics, not as important, the they become tiny considerations when we start looking at these massive changes. And I don’t take talk hyperbolically Very often, Brian, you’ve known me a long time on this. You know, this is a revolutionary change to accounting. The only thing I don’t know is kind of when it will happen. Like why to cube not quite sure the date. But I can see the trains coming as down the track. And I want you to be you know, getting ready, accommodated as arrive. So Brian, I know I’ve gone on a long rant today, I didn’t let you talk as much as I probably should have. But you can see friends, I’m excited about this. This is a big, big, big, big, big deal.

Brian F. Tankersley, CPA.CITP, CGMA  27:01

Well, I actually have actually have a personal friend that I’ve interacted with for a number of years that actually design some of the supercomputers in Oakridge, and some of those data architectures and so forth. So this is, you know, this is an interesting topic, and it is definitely coming down the track. And I will just say, whoo, let’s go.

Randy Johnston  27:22

All right, well, maybe I got you a little bit more on board with me, so to speak. And you know, I love where I’ve been on train. So maybe we’ll get one of those trips together and one of them will be on quantum computing. France will talk to you again in another technology lab today.

=== END ===