Welcome to the episode with Tom Fox, the Compliance Evangelist. Tom is a renowned expert in legal and regulatory compliance with an impressive portfolio that includes founding the Compliance Podcast Network and authoring 30 books on compliance and business ethics. Known as the Voice of Compliance, Tom brings a wealth of knowledge and practical insights to the burgeoning area of AI compliance. In this episode, Tom and I discuss this crucial intersection of AI and compliance in detail, exploring how businesses can navigate this evolving landscape safely and effectively. Now get ready for a fun interview with a top legal mind. Let's get started. Welcome to Using AI at Work. I'm your host, Chris Dagle. Each week we'll be learning how today's business owners, entrepreneurs, and ambitious professionals are getting more done with smart use of tomorrow's tech. Let's get started. Okay, everybody, welcome to this week's episode of Using AI at Work. We're very fortunate to have Tom Fox, a compliance expert uh beyond Compare, uh, joining us here on the podcast. So, Tom, if you don't mind, before we jump into the nitty-gritty, kind of share uh a little bit about your history and how you ended up here.
SPEAKER_01Well, Chris, first of all, thank you for inviting me uh to come on your podcast. A huge fan, first-time guest. Uh I have been in the legal and regulatory compliance space since 2007. I um started, I went out on my own in 2010. I started blogging. That led to podcasting. That led to the founding of the Compliance Podcast Network. I've written 30 books in this arena. I commentate, pontificate occasionally, practice law, but I don't say it very often. I don't say it very loudly for fear that someone might hire me. And I generally uh sort of do all things compliance. I'm the voice of compliance and I'm the compliance evangelist.
SPEAKER_02Fantastic. That's a topic, as I mentioned before we started recording, that hasn't gotten a lot of attention here on the podcast yet. However, um, as more and more adoption of generative AI happens in small businesses, it's very important for them to be aware of. So I'm kind of curious how how have you seen that compliance conversation related to AI change over the past 18 months or so?
SPEAKER_01So it's been a fascinating exercise for me and probably for you as well to watch this grow literally in front of us. The people that were the true pioneers, however, have been doing this, turns out, for quite some time. And so I was at a compliance conference in early April where uh a woman who has been in this field for about 10 years said, Hey guys, we've been doing this for 10 years. And guess what? It's just a tool. And she really laid down the basics of that. And that has really sort of struck me when she said that. I had always felt like generative AI was a tool. In fact, the first time I used it, I thought the only experience I could compare it to was when my parents bought my sister and I Encyclopedia Britannica when I was 11 in 1969, and I thought the world was at my fingertips. I thought the same way the first time I used generative AI. But it's just a tool. It is a sophisticated tool. It is a, it may be a generational tool, and we are exploring the parameters and limits of that. So it's it's been fun to see it change and morph. Obviously, uh November 2022 with ChatGPT was huge for many people as it put it on our radar, and I say I because I include myself in that. So the struggle has been this is a tool we don't completely understand how it works, at least some of us laymen don't. So we have to begin to figure out what are the inherent risks, legal risk, business risk, compliance risk, et cetera. And that's sort of what we're struggling through and trying to, once you figure out the risk, then you can put a risk management strategy in place. You can train your employees on it, you can monitor that strategy, and then improve it as necessary. So it's a process, it's a risk, and once we can slot it in to the risk management strategy, I think we can begin to answer some of the questions that many of us have right now. Fantastic answer.
SPEAKER_02So I'm an entrepreneur. A lot of people who are early adopters of tech tend to be entrepreneurial. Um, and as an entrepreneur, the the mature choices of compliance and risk management seem to be something that's not the first thought when it comes to using these tools. What are the risks for some of these SMB, small to medium businesses who are diving headlong into the use of it or exploration of it without a consideration of the compliance and regulation of it, internally or externally?
SPEAKER_01Let me be maybe push back a little bit on your introduction, not your question. Because what I find is entrepreneurs are typically process-oriented. And that's what I want to emphasize that compliance is a risk management strategy that is process-oriented. And for an entrepreneur, uh, certainly you can be ahead of the curve, you can uh be the first to adopt, you can run very quickly and fastly. But the people I've interviewed who have been successful in this have a process. And that process is they see something, they uh do some market research, try to figure out what the user experience might be, try to create a minimum viable product, have beta testers test it, and then take it to market and uh either get funding for that or self-fund and then either sell it and move on to the next entrepreneurial thing, or uh they grow that. So I find actually entrepreneurs are very process-oriented, particularly ones who've done it more than once or twice. And that's so what I want to suggest to the audience is this is going to be really no different. The compliance perspective on this, this is not gonna tell you to stop. We're not gonna tell you you can't do this. We may say you ought to put some guardrails around this for your own protection. But if we can put it within the process you're comfortable with and people understand this is just another tool and another business process that you can make a lot of money out of, uh, that's really the message I want to get across. So the Yeah. And uh the issues are from less compliance and more legal, are what, you know, are there any copyrights? Are you getting information that's copyrighted? Uh are you putting information in in the form of your queries uh that is giving away confidential information protection that you have? So if you have a product, you haven't trademarked it, you haven't patented it, but you've kept that information in that process confidential, you put it into Chat GPT to ask a question with prompts, guess what? It's now in the public arena. Uh then what do you get out of Chat GPT or generative AI? And and the one example, or one of the ones is a lawyer we all laughed at was the lawyer who turned in legal briefs that Chat GPT wrote that had fake citations. But as funny as that may have been, it it illustrates the next point I want to emphasize. Chat GPT is going to give you a first draft of something. Generative AI is going to give you a first draft of something. Now, if I put a compliance question in, I already know the answer. So if I get a first draft of something, I'm going to know if it's correct. But if you don't know if it's correct, you have to research that answer to make sure it's correct. So don't ask something that you're going to depend on the answer on without first double checking it. Number one. Number two is if you're going to ask ChatGPT to write a paper or a legal brief or something, uh, it's a first draft. Do not cut and paste, put it on a Word doc, and send it in, even in a PDF, uh, because you don't know what's in there. And if you turn it into a judge and the judge reads your cases, it turns out they're fake. You're going to be paying a monetary penalty for that. So right now I think some of those challenges are uh almost basic. Uh we're moving into uh another arena or another area where such things as real-time deep fakes, information barrier, evasion, uh, and uh software detection, verification are becoming more important. But those are really still sort of at the very early stages. Um for a business, I would I would agree with you 110%. Embrace this. It's not gonna replace you, it's gonna make you faster, quicker, better. If you can figure out, and and I'm not a prompt expert, but if you can figure out how to do prompts, you are gonna be exponentially better than your competition. And I've seen people give presentations at conferences, you know, right in front of us up on the screen. So I know there are those people who can do that. Um for me right now, it it gives me more information, more uh data that I can use. Uh, and so I'm just sort of exploring those parameters. But I I would say embrace it wholeheartedly, check the results you get, think about what input you're putting into it because you don't want to give away anything that's confidential, and then use it as a first draft of a something.
SPEAKER_02I love it. So, what would be for the the business owners or the team leaders out here who are excited about introducing AI into their business, but want to make sure they're doing it, quote unquote, the right way. What would be, from a compliance perspective, what would be a good first step?
SPEAKER_01A good first step would be uh determine who in your organization is going to have access to it. My wife's company just bought licenses for Chat GPT. So I sort of watched her go through that process. And they actually limited uh with uh several thousand people in the company, they bought 200 licenses because they really wanted to beta test within the company. So if you're a large organization, you might want to consider that. Uh if you're a smaller organization, I would say uh the most important thing is really to understand how it can help you and explore. Uh some of the best advice I hear at conferences, take a weekend, you know, take a week off from work and just start working through it and see what you can get and utilize it that way. The once again, the dangers are uh using it as an end product because you don't understand the source of that data.
SPEAKER_02Instead of, yep, we call that human in the loop over here because you can't trust the information. So one of the things that I hear a lot um from other resources is to one of the first steps that a company should do is establish that generative AI use policy. And that's essentially uh in line with what you were just suggesting with your wife's company as the example, that they've got people using it. And will those individuals typically inform the creation of the use policy for the rest of the corporation or the enterprise?
SPEAKER_01Uh they can be a part of that, and I would suggest you always try to have your employees as a part of the input into any policy you create. One to create buy-in, but two, there's a subject matter expert. So I'm a lawyer, I can write you a policy on anything. It may or may not be correct. So you want to bring those uh uh I'm I'm a big believer in the UX in everything. And that includes writing policies and procedures. So get your users, have input, write your policy that sets the parameters, but that's just one step. You may need to write procedures around that, but the key is to train your employees. Don't write a policy, slap it in your policy manual and say, Well, didn't you read the policy? No, I didn't read the policy. I'm actually trying to make money. I'm not a lawyer. Yeah. So uh you've got to train on that. And then the next two steps are monitor the results of your policy and improve or continuously improve as appropriate. So at this point, I would take a look at everything probably every 90 days or at least every six months, because of the changing nature of generative AI and what's available out there. Last week I was at a podcast conference that was podcast uh on AI, and the last conference I went to was in January, and I could not believe how much we'd advanced since January. Wow. So, you know, it's constantly evolving.
SPEAKER_02So this um I like the idea of checking in as the as the policy is in use and people are uh deploying AI in their business to have this regular tempo of checking in to make sure that are we still following the guardrails? Does it still make sense? Is this too restrictive or anything like that? Who in a company typically would lead that initiative?
SPEAKER_01The initiative should be led by uh the compliance function or the legal function, if you're too small to have a compliance separate compliance function. But I can't emphasize enough, it's all about the UX user experience. You can't write a policy that's inapplicable and then expect people to follow it. That's why you need the input on the policy. And then set some controls around, whether it's reviewing of uh certain outputs of generative AI, you have a committee looking at uh report or some sort of major project out of it. So you put a second set of eyes on what's being created. And then if if you can see that it's working generally along the lines that the policy lays out, then you can move forward. But if something's not working, if some types of prompts or some types of information are giving you uh false leads or uh incorrect information, uh that needs to be communicated as well. Or human in the loop. I love that.
SPEAKER_02I I actually heard that first term from the guys at the Marketing AI Institute. And if you're not familiar with them, uh follow them. They're fantastic. Um, you know, you mentioned something a moment ago about, hey, I'm an attorney, I can write you a policy on anything. For the individuals who are out there who are listening to this and they're like, you know what, we need to do this. What should they look for in counsel that will let them know, hey, this person knows what they're talking about. We're we're like, we're gonna get a valid and applicable uh compliance output from this from working with this attorney.
SPEAKER_01So the first thing to do is just ask those direct questions. Uh there 98% of all lawyers say we could do everything.
unknownYeah.
SPEAKER_01And it took me a long time to learn not to do that. Uh, but uh if somebody comes to me now with something I can't do, uh I don't want to learn it at this point in my career. So I say I'm not the guy. So that's number one. Number two is could you give us some clients that we could uh essentially referrals that we could look at? So it's it's almost basic due diligence uh to look, check out their website, see if there are resources on there, if they've written about it. That gives you a sense that you know they may have been in this field for 18 months, but that may be 18 months more than the other lawyers in this town or this community.
SPEAKER_02And that makes a bit 18 months in the the AI space, tech in general, but particularly generative AI is uh that's an eon for sure. Um, are there particular resources online that might help if somebody's maybe a solopreneur and maybe they don't have the budget to engage with an attorney for this? Are there resources online where somebody could self-educate a little bit more about the compliance environment for AI?
SPEAKER_01Hmm. There is no resource or database for that as yet. I think some of the compliance large trade organizations are moving in that direction. But as of now, uh, and so for instance, if you called me or sent me an email and said, Tom, I need one, two, three, and I didn't have it, I would simply call two or three or four of the people who are much more advanced than me and say, Do you have one, two, three? And um there are people who have self-educated themselves literally starting in November of uh 2022 and are at the leading edge of that, but there's no overall resource base or repository that I can point you to today.
SPEAKER_02Kind of a dangerous environment for somebody that's trying to set this up if there's no uh there's no established protocol necessarily. So are there any legal um uh, I don't know, bills or anything coming down the pipe that that a user, a power user of generative AI in their business should be aware of?
SPEAKER_01There is some discussion. Congress is moving now towards data privacy and data protection. Generative AI will be down the road. So there's nothing significant for any of us to watch. But if something does come out, um obviously that will be uh something we should all uh think about and look at. There are governmental agencies that are starting to do enforcement actions, but they're around companies that claim uh they are using generative AI to develop a product or service and they're not. So in the financial advisor space, there was recently two enforcement actions against companies that publicly said we're using generative AI for uh market uh you know selections and they weren't. Um and so uh it's called AI washing. Um, but if you uh claim you're using it and you're not, then you can get in trouble for false advertising, essentially.
SPEAKER_02It's the complete opposite of what I was going to expect. I thought that you were gonna say if you're using AI and not telling people you can get in trouble. But actually, if you're not using it and saying that you are, you can get in trouble.
SPEAKER_01There, I think there's many people who believe sort of morally you should tell people something like that. Uh, on the other hand, uh, as a lawyer, I never tell anyone, well, I went to the library and looked this up and I put it in my brief. Uh I'm not offended to tell people I'm using AI. I just don't see that it's needed because I use it as a first draft.
SPEAKER_02Okay. So big distinction there for the listener. First draft means that's going to get modified, or again, that human-in-the-loop term. Tom is going to be reworking the output as compared to copy and paste like the attorneys and that anecdote that he shared. Uh, and that story, I think, is still going on. I heard something recently about that, that they kind of have doubled down on the fact that they were right, but uh use that as a cautionary tale, ladies and gentlemen. So um, one of the things that's very interesting about this time is that not only are you uh a compliance expert, particularly when it comes to the topic that we're discussing today, but you're also uh a podcaster talking about this topic regularly. And before we jump into kind of the geeky stuff about how you're doing that using generative AI, first I want to say that that makes you uh a better voice to address this because you're using the tools. You're not just looking at uh the theory of it, you're actually seeing these in use and can um have a more, I guess, informed decision. But in in the uh in the as far as topics when it comes to the compliance podcast network, I would assume that AI is is becoming a more regular conversational point in those episodes. Indeed, I have a podcast called AI and Compliance.
SPEAKER_01Love it. So, yeah, so on number one, it is. It's a topic that's front of everyone's mind. As I said, this conference I just attended in April, uh, where the woman spoke about it. And um, so that was that's one angle. But it's the use of AI in podcasting, that's the where I do have the most experience because it has given me the ability to create an AI voice, to create AI images, uh, to do video graphically video things that I've never had the skills to do. It has given me um information on topics, that's the Encyclopedia Britannical part, um, at literally in seconds. So if I type in a Sherlock Holmes short story title, you name it, and I say, give me 10 ethical lessons, I get 10 ethical lessons. Now I have to edit that, but it's given me the ability to create more, to create faster. Uh, there are multiple tools in the podcast world that are generative AI based that create show notes, they create blog posts, they create outlines, they create social media shares. Um and so that part has uh greatly decreased the amount of time it took me to create a podcast. And uh the best example I can give you is uh one of my top podcast service offerings. It's a five-part series. It goes out five days in a row, 15 minutes each, blog posts, show notes, videograms, edited white paper based on the blog post, uh, all in a package, one package for a customer. It uh prior to Chat GPT, it took me 20 hours to do that. Earlier this year, I had set aside two days to do that series, started at eight, at eleven that morning I was finished. I just and it was the only time I sat straight through to do it, so I knew what the time was. And I just I was absolutely stunned. But here there's even a coop coda to that story that's better. So the guy that developed that tool was a solo entrepreneur. Entrepreneur in college, and I emailed him. I said, You're not gonna believe the time savings you just give me. And it turned out he was having a terrible day, something had gone horribly wrong, and he got my email and it just made his day. Yeah. So, you know, if somebody gives you or creates something that you use, tell them thank you. It actually means a lot.
SPEAKER_02Great tip. Let me ask Tom, have you always considered yourself kind of tech forward?
SPEAKER_01Uh actually, I considered myself a tech phobe. And part of my origin story is um uh in my last corporate position, that's where I learned about compliance. My company was sold, my job went away. I decided what I really wanted to do with my life was race bicycles. So I went off on this great adventure to race bicycles. I was over 50. I was riding in the senior division until I got taken out by a Hummer on a training ride, which ended my cycling career. Convales, you know, after a few months, I had the energy to get on my walker and toddle into my office. And I had realized I was gonna have to go back to work. And at that point, I literally knew nothing about social media or tech. I didn't know anything about WordPress, I didn't know anything about Facebook, Twitter, LinkedIn, you name it. And the only thing I had on my hands was time. I had no money. So whatever I did had to be at no cost to me. And all of those tools, at least back then, they had a free component. And so I literally started at ground zero and taught myself.
SPEAKER_02So this was prior to November or this was after November 2022? No, no, this was 2010. Okay, okay. Now, were you uh because they've had Jasper and Jarvis and there was a Chat GPT, or there was a GPT-1 and a GPT-2. Were you using the tools prior to the release of Chat GPT?
SPEAKER_01Uh, you know, I dabbled around with Jasper and those tools, but at that point, this was like 2016, 2017, they weren't precise enough. They were too general. I could put a question in and I would get a general answer about what compliance was, but it didn't have any specificity. And the time it took me to rewrite whatever I got, I could I knew the answer and I could write it more quickly. So the I saw the basics of the tools. They weren't quite there yet. And it literally was uh November of 20, I thought it was 2022, but whenever ChatGPT was relate released, that's when uh I went into it whole hog.
SPEAKER_02So as soon as you heard about it, you were on the case, you were checking it out. Yes. Nice. Were you able to apply 3.5 when it came out? Were you able to start using it right away in your podcast process?
SPEAKER_01Uh as soon as I paid my $20 a month that I could get in, I was all in. Yep. I tried the free version and it could never, uh it was always backed up. So I just, you know, 20 bucks a month. For what I was getting, the savings and efficiency were far more worth it.
SPEAKER_02I would imagine in the past 18 months, your AI-powered processes have gotten a little more sophisticated when it comes to the the podcast network. Do you mind kind of sharing with us what that looks like? What how you're leveraging generative AI and not just the production of your own podcast, but everything that goes around that in running a business?
SPEAKER_01So the um I would say the tools have not become more sophisticated. I would say they're more tools. So on the content creation side, it's much greater amount of tools. That's the audio, that's the video, uh, that's the music, that's the graphics, that's the um transcripts, the scrolling transcripts, uh, that's the voice that you heard on uh my daily news podcast, uh social media share. Uh people love those and will engage much more on LinkedIn, Facebook, YouTube, Twitter, uh I suppose TikTok if I use that, but I don't. So um it's given me the ability to do things that I had to pay hundreds and sometimes thousands of dollars for uh before that time. So that's number one. Number two is I'm just uh I can't say that I'm the expert on this, but I have people who work for me who are experts who use the APIs to connect various tools. So you're we're recording this on Riverside. I can take Riverside, I can put it into a tool called DScript, that can give me a transcript, and from there I can create the video gram, the text of the videograms, I can give that to the audio guy, he can create the AI voice, and we can just go through that process uh almost seamlessly now. Uh the so it's really uh led to much greater efficiency for me and the ability of me to do more.
SPEAKER_02You know, you mentioned that you've got some folks helping you with this. And one of the biggest issues with with, I guess, anybody who was interested in doing this are one, where can I find the talent? Or two, where can I go learn the stuff myself? How are you sourcing uh trustworthy AI capable talent?
SPEAKER_01The um what's the right word? It used to be called side hustles, but those of us who have worked from, I've worked from home since 2010, every creative uh has the ability to market themselves literally internationally now through Fiverr or Fiver. That's one source. Uh I live in a small town of 20,000 people. I just went to the computer repair store and said, Hey, you know anybody who's doing this animation? And the guy said, Oh, yeah, you know, my buddy at Shriner's doing that. Uh so I mean, you can ask around and find out the people. Um if you have uh a sufficient income and need a dedicated employee for 40 hours a week, there are fantastic outsourced uh uh virtual assistant companies that work with uh people in the Philippines. So I have I work through one of those companies. I have two resources dedicated to me full-time. They're 40-hour a week employees, although they're not my employees, so I don't have to pay any taxes on them. And so depending on the level of income you are, you can always find uh the resources and then uh to help you. And like I said, I just went to the top computer repair store in Kerrville and asked that that question. Um and then um there's so much information now on teaching yourself, and the the tools have almost gotten idiot proof, almost.
SPEAKER_00Yeah.
SPEAKER_01Uh so you know, uh, if you don't have to put a credit card in to do the free trial, I would say go to any of these services. There's some emails that you and I probably get daily on uh new uh uh generative AI tools, and I check those out uh every time. I've still got bookmarked. There was one from Friday that uh Apple has eight generative AI tools now, and I'm working through all eight of those to see what uh Apple may bring that I don't currently have. Any discoveries so far on that? Well, here's the I where I think the next generation's going, and or I'm trying to figure out how I can be a part of this as an entrepreneur. Um right now, if you entered a compliance topic into Chat GPT, you'll you know, like we discussed, you would get something. Well, I have been writing about compliance for 15 years. Sure. I've been podcasting about compliance for 12 years. I have every one of those stored in the cloud. And if I could take, so what if I could sell the Tom Fox library as a self-contained unit? Now, uh any those are all in the public record. Anyone could go get those, but I've got them all on iCloud. Yep. You know, and so if I could figure out a way to ring fence that information and create a generative AI compliance tool, uh I I would have something. And I think that would be true in any area, any vertical, any silo, any industry. Uh, you know, you name it, supply chain, operations, sales, BD. Great point. See uh marketing, finance, etc. So that's really that's what I'm trying to see if Apple's gonna give me the ability to create something like that or allow me to pay somebody to do that.
SPEAKER_02You know what? One of our uh former podcast guests, I need to get you introduced. I think that the solution is right under your nose. So please do that offline. Yeah, for sure. So um, and that's an interesting I uh idea. You mentioned a couple things I want to unpack there. The tools, almost idiot proof. Totally agree. And I think that there are people that are out there, like when ChatGPT first came out. I mean, I've been using, been a digital marketer for 20-something years, very comfortable with tech. But when it first came out, I thought, oh, that's somebody else's opportunity. That's it's technical, right? Uh turns out it's not. And that what I've what kind of our main thesis at my my main business is chiefaiofficer.com. And one of the things that I teach people is that listen, the hard part, you're already doing it. The domain expertise is the hard part. The tools aren't. Now, you can find that tool expert, but if they don't know your domain, they're gonna have limited impact. But you, having been in your industry, whatever that industry may be, for X number of years, you're bringing with you an instant under if you can understand what the tools do, then you can instantly go, oh, I can use that here, I can use that here, I can use that here. So for those of you listening that think that there's some sort of impediment to you uh being as capable as Tom is with these stuff, the stuff in your own business, it's not. Please start, like he said, take the weekend, use the free examples. Uh I think you'll be surprised that the learning curve is a lot uh more compressed than you would expect.
SPEAKER_01Uh it is it's almost either cut and paste or drop and drag. Almost just that. And it just dawned on me, and and when you were describing learning the prompts, I graduated from law school in 1983. There were two computer search services, Westlaw and LexusNexis. And guess what? You had to enter a prompt to find the cases that you needed. So this technology has been around forever. And it when we say, Oh, you must refine your prompts, that's not some black box magical thing that you need a degree to do. It's literally you and I sitting down for a weekend, starting with Chat GPT, tell me all the Sherlock Holmes stories that deal with ethics, and just go from there. They're um in the conference I was at, the virtual conference last week, a guy in 20 minutes showed us how to write a 200-page book. And it was just refining prompts, refining prompts, refining prompts. And he would get uh a table of contents, and then chapter one, he would refine that prompt, and then he would refine it down. Now it's a process, and you have to sit through it, but it's just refining what we used to call in uh the legal world your queer query. Now it's refining your prompts.
SPEAKER_02So one of the one of the things that I guess first got me um FOMO about AI was I heard a quote from Peter Diamandis, and that was that by the end of this decade, there's going to be two types of companies, those that are AI enabled and those that are extinct. What do you think? Do you do you think that that timeline is accurate? Do you hold a similar sentiment?
SPEAKER_01Uh I don't think the other companies will be extinct. They will be lesser and different. Uh, it's a difference in virtual and brick and mortar. Sometimes you want a brick and mortar experience and sometimes you want a virtual experience. Um but uh I do agree with the the sentiment, except it may not take that long. It may be 2025, uh, at the speed of which change. And and uh go back to that Apple example, Apple's most sophisticated generative AI tool, that's the one I'm looking at. And if if I can figure out how to ring fence the information I have and make a minimum viable product, uh every company can do that. And if you can have your own AI resource tool that you can sell or market or license, uh that's gonna be a huge change.
SPEAKER_02Yep. I like the idea of where you're going with that. With 15 years of compliance content, that would be a very powerful uh case-specific application of AI. Uh I can't wait to see. You're gonna make it. I know you are, so I can't wait to see it and start using it.
SPEAKER_01Well, I guess, Chris, what I I would maybe change also that in 2030 there will be those who have fully embraced and can use generative AI and those who can't. And it'll be almost as stark as those who can't use a computer now, or maybe even can't even add. Uh, because those of us who can will be going at light speed or even, you know, hyper hyper speed.
SPEAKER_02Yeah. Uber humans, uber cognition for sure. So let's talk about. Um, we don't exclusively always kind of go into the tool stack, but I'm curious. I got a podcast, we repurpose the content using generative AI. I'm curious what your uh tool, your AI-powered tool stack looks like for uh the work that you do with the the network.
SPEAKER_01Let me start with the guy that uh the college student, he's at BYU, he graduated this week, he started a company called Podflow, and literally out of his dorm room. Uh and uh it is uh basic, but it it gives you all of the things I described. Um in terms of the tool that changed everyone's life in podcasting, it's called D Script. And uh DScript is more sophisticated than um PodFlow, um, but I use both. Um there's another tool called Cap Show, and they have uh more more sophisticated um written product out of your podcast than either D script or podcast AI, and I use them all. So I really focus my part of the podcast world is the written word because of my legal background. The uh the team in the Philippines do the uh graphics and animation, etc.
SPEAKER_02Are they using the AI tools to do the graphics and animation?
SPEAKER_01Absolutely.
SPEAKER_02Yeah, absolutely. Do you know are they using D script or they have their own little stack?
SPEAKER_01Well, they have my they have access to whatever I use because I buy a number of like the number of licenses, but they have uh other tools. Um Canva is a huge, hugely popular tool. Um DALI, I think is a video tool. Um and there's a couple of audio tools, but I can't remember what their names are.
SPEAKER_02So you guys are doing full spectrum, full uh full sense, all the senses are being addressed by AI in your process.
SPEAKER_01Absolutely. And uh these are not expensive. The the individual audio video graphics tools, those are $10 a month.
unknownYeah.
SPEAKER_01Uh DScript is $250 a year. Uh so these are affordable tools uh that you can use. I think Podflow is $25 or $30 a month. Uh so these are all things that uh you can afford, uh, and uh it will give you a lot more information a lot more quickly.
SPEAKER_02So yes, it's cheap. It's inexpensive, let's say, but oh, I've still got that learning curve. No, you can probably go to YouTube and anything you want to learn with a generative AI tool, there's a 22-year-old who's gonna teach you how to be a power user on that within a you know 30-minute video. So it's uh the barrier's down.
SPEAKER_01Yeah, any established company, they're gonna have webinars at least once a month, sometimes twice a month, because it's in their interest to teach us how to use their tool. And exactly like you say, there's a five-minute tutorial literally on everything because it's in their interest to do so.
SPEAKER_02One of the things is for us is the repurposing of that content. So once you've done the episode, you are creating social posts from it.
SPEAKER_01Do we just go through the list? Yeah, that'd be great. Okay. Uh yes, I tell people you give me a word, I'll give you a sentence. You give me a sentence, I'll give you a paragraph. You give me a paragraph, I'll give you a blog post, you give me a blog post, I'll give you a white paper. You give me a white paper, I'll give you a chapter of my next book. So I'm very much process-oriented, and I see everything in that light. So I would, if I was, if this was my podcast, I would take it, dump it into D script, and get uh show notes out of it. Uh, there would be a paragraph description and then some key highlights. You can also pull quotes uh and other information. You can do it in outline form, uh, you can do it in summary form, and I would put the resources, I would put your contact information uh and anything about you you wanted me to put in there. I would take that transcript and I would get a first draft of a blog post on it. I would then take that, edit that, add to it, make it my writing, and then I would post it on my blog. What I did, uh I write an annual I have a compliance handbook, I'm in my fifth edition. Every year I make a uh I change it by 20%. Last year, the biggest change was in data-driven compliance, which was analytics for compliance officers. Well, six months into it, I see next year is going to be AI and compliance. So every podcast I do around AI, I collect, and then I'll re-edit all of that into a chapter for the book. Nice. Back to the podcast, I create um four videograms for social media sharing. Three at 60 seconds or less, because that's a limit for Twitter, Reels, and shorts, one at two minutes or less for LinkedIn. You can post a LinkedIn video of up to 10 minutes, but most people won't listen much past two or three. So I get four uh videograms for social media sharing content. I get all of my social media posts, tweets LinkedIn, Facebook, Instagram, and I post all of those. Part of my team does that all day, every day. I post six podcasts a day, so it's a lot of stuff.
SPEAKER_02So what are you doing with the extra time that you now have because instead of a 20-hour effort, you can do it in the morning.
SPEAKER_01I have a long and unfortunately storied history of addiction. And part of that means I got one switch. It's on or off. Got it. When that switch goes on, I can't stop. And so it's just, I just work more, I do more, I create more stuff. I on our uh Sunday night call with my team, the question, half laughingly, was how many podcasts are we going to start this week, Tom? And I said, uh, don't worry, it's only three. So, you know, I'll start three new podcasts. So are these yours or are you doing this for clients? So uh my network has 60 podcasts in it, and it's divided into two categories. Those that I host uh or produce are called Tom Fox Podcasts, those that somebody else hosts or produces are called non-Tom Fox Podcasts. So I have 50 active podcasts, and I have 10 non-Tom Fox podcasts. So that's in what actually I have another network too, but that's another story. Um, so that's my biggest network. Wow.
SPEAKER_02Okay, so everybody listening to this, Tom, how big is your team? Three. Okay. And like the other two are really just following your lead. They're not necessarily like thought leaders in the space, they're just doing what you tell them. They're doing all of the hard work. Yeah. Okay. So everybody listening to this, please understand that Tom has got 50 podcasts that he's doing with a team of three and a handful of AI-powered tools. And this is like, are you working 120 hours a week? 80. Okay. But I would imagine that, like you said, with with your personality, it's almost like you go into a flow state. So it doesn't feel like work.
SPEAKER_01Well, it's actually that I love it. Yeah. Because someone asked me at a podcast conference, what's the hardest thing about podcasting? And I thought about it a minute. I said, there's nothing hard about podcasting. Here's what I do all day. I talk to people like you. And one of the reasons I'm the one of the top experts in compliance is I talk to every thought leader. And so, and in AI, I'm doing the same thing. I'm talking to the top people. And even though I'm being interviewed, guess what? I'm learning more from your questions than you're getting from me. Believe me. So I view this as just having fun. And I just love what I'm doing. And I see no reason to stop doing it. And I'm pretty successful at it. So I'm going to keep going. I would love it.
SPEAKER_02So, Tom, we're coming up to the uh top of the hour. What would be the best next step for somebody who, first of all, they just wanted to maybe follow what you're doing, but in particular, your your discussions on compliance. Well, how can they plug in?
SPEAKER_01Sure. So my website is uh compliancepodcastnetwork.net. Uh my blog posts are on that site. I blog daily as well. All of my podcasts are listed on that site. I'm on LinkedIn at Thomas R. Fox. Uh, every post goes there. Uh, YouTube is uh the Compliance Podcast Network. Uh, you can email me at tfox at tfoxlaw.com. If you're new to podcasting, if you're new to compliance, I've got a ton of materials I will share with anyone anytime for any reason. Uh happy to chat with you going forward.
SPEAKER_02Fantastic. Thank you so much. And everybody, if you'll check the show notes, we'll have the links to all of those uh destinations that Tom mentioned. Tom, thank you so much. This was a fantastic inaugural uh episode on compliance. And um, I'm glad that we didn't find somebody who was uh, you know, too buttoned up and dry on the compliance side. You you happen to have the compliance expertise plus the enthusiasm and expertise when it comes to using AI in the process, which is always a wonderful match for our guests. So uh with that, thank you, everybody. Tom, you got any closing remarks? Uh Chris, I hope we can continue this conversation. I think that we will. Take care, everybody. See you on the next episode. Thanks for tuning in to using AI at work. Don't forget to subscribe for more conversations about how to use AI at work. And a special thank you to our sponsor, Chief AI Officer, for empowering businesses with AI education and training. Visit their website for a free AI readiness assessment and AI strategy guide to help you get started using AI at work. That's www.chiefaiofficer.com. Well, thanks to our producer, Evan Desolnier, for making this episode possible. Follow us on Twitter at the handle UsingAI at Work and visit www.usingaiatwork.com for free resources to help you harness AI in your role.