00:00 Clip audio We know what to do and we do it well. Examples abound around our organization.

00:04 Paul Brandus The commanding voice of authority and certitude, or so it seemed, when that man extolled the virtues of his company. His name was Jeffrey Skilly, and the company at the time was known as Enron Gas Services, later shortened to just Enron, a Houston-based energy, commodities, and services company. In the late 1990s, Enron was a Wall Street darling. Revenues soared, and by the year 2000, Enron was ranked as one of the biggest companies in the world. That year, according to Forbes, it claimed revenue of more than $100 billion, four times software Goliath, Microsoft, another business publication, Fortune, rated Enron the most innovative large company in America. But beneath these numbers, the gusher of profits, the glittering image, something was amiss. Enron was less than transparent with investors, a fact noted in a now infamous earnings call with analysts in April 2001, after a presentation, Skilling, by then chairman and CEO, took questions. What kind of company, a publicly traded company, doesn't share its most basic financial information with investors? The answer, it turned out, was one that had something to hide, the truth. Of course, there's another word for this deliberate and malicious deception, disinformation. I'm Paul Brandus, and that's the name of this award-winning podcast series, Disinformation. As usual, I'll be joined by Meredith Wilson, Chief Executive Officer of Emergent Risk International, a global risk advisory firm. She'll offer her insights into this crucial topic.

The analyst who challenged Jeffrey Skilling, his name was Richard Grubman, died in 2019. His accusation that Enron was not being transparent about its finances turned out to be painfully true. What financial information Enron did share was overly complex. It soon became apparent that Enron was cooking the books, misrepresenting its earnings, and altering its balance sheet to make the company look healthier than it was. Skilling's chief financial officer and other executives misled Enron's board, its audit committee, and pressured its accounting firm, Arthur Anderson, to ignore most of this. Since the definition of disinformation is the deliberate manufacturing and dissemination of false information with the intent to deceive, Enron's actions clearly fit the bill. Here's Meredith Wilson. It wasn't just Enron, of course. Other huge cases of financial disinformation have rocked the markets and investing public over the years. You might remember a company called WorldCom, once America's second biggest long-distance phone company. Then there was Tyco, a New Jersey conglomerate, and Health South, an Alabama medical firm. That last one, Health South, is explained by Craig Lewis, once chief economist at the Securities and Exchange Commission, now a professor of finance at Vanderbilt University in Tennessee. I spoke with him at a recent disinformation conference at Cambridge University in England.

04:24 Craig Lewis They got in trouble because they initially were attempting to position their earnings numbers so that they could beat the analyst street estimates. In the first quarter they did that, they needed a relatively small amount of accruals to take a negative earnings surprise and turn it into a positive earnings surprise. I think the CFO at that time certainly thought that that would reverse out the next quarter or so and that it would all work. As time went on, it got harder and harder and harder to reveal the fact that they weren't performing nearly as well as the street wanted them to. There was an active example of trying to manipulate the accruals that they were booking to reflect performance that was inaccurate. In that sense, it's disinformation. And so what we were trying to do at the commission was learn from that type of behavior and see if you could identify other firms that were trying to do similar things with their accounting performance. In other words, trying to prevent, trying to project an inaccurate representation of what is actually going on in an attempt to deceive investors.

05:45 Paul Brandus Now Health South could have been honest about their numbers, but a poor quarter would have driven the stock down. We can't have that, now can we? So executives chose to cook the books instead. To Meredith Wilson's earlier point, it's all about the money. Then, of course, there was one of the biggest crooks ever, Bernie Madoff, who made off with tens of billions of dollars over decades before his arrest in December 2008.

06:09 Clip Audio The FBI arrested him this morning after he told senior employees yesterday that his business was a giant Ponzi scheme.

06:22 Paul Brandus An old-fashioned Ponzi scheme among this swindler's many victims, Juliet Pfeffer. For about 20 years we were with Madoff and one day I went to sleep a very comfortable lady and I woke up. It was a catastrophe. It was all over. All my money was gone. She and legions of others were fooled by Madoff's charm and his phony investment statements. He told them that they were doing well right up to the moment when it all came crashing down. Bernie Madoff, the villain, the bespoke suited peddler of disinformation, died in prison two years ago, one of the most reviled characters in American history. So we've seen the destructive power of disinformation here, its ability to wipe people out and ruin lives. It's a good time to take a break, but when we come back, since everyone's talking these days about artificial intelligence and things like Chad GPT, we'll explore whether these things can help root out future attempts at financial disinformation.

07:33 Ad-read This series on disinformation is a co-production of Evergreen Podcasts and Emergent Risk International, a global risk advisory firm. Emergent Risk International, we build intelligent solutions that find opportunities in a world of risk.

07:50 Paul Brandus Gentlemen, we all know that first impressions matter and if you're not taking care of your skin, that's going to be a huge problem for you. We're going to have to do a lot of research to find out what's going on. I've been using their skincare products for a couple of weeks, now just twice a day, and I can see the difference. The first thing I do when I wake up, even before that first cup of coffee, I use Caldera's Clean Slate. It's a facial wash. They told me it would be refreshing and they were right, it is. Then I put on a base layer, which hydrates the skin, really important to do that. And in the evening, the good, that's what it's called, the good. It makes my skin look tighter and smoother, makes wrinkles and fine lines less visible. Every drop of the good contains millions of antioxidants that help cleanse and protect my skin. Feels great, looks great. In fact, clinical trials have found that 94% of men's skin looked younger and healthier after just a few weeks. You can add me to that group. Want to look better and feel better? Make sure those first impressions are perfect with Caldera Labs, the leader in men's skincare. Here's an exclusive offer available only to listeners of this podcast. 20% off, just go to calderalab.com. That's C-A-L-D-E-R-A, calderalab.com, and use the promo code DISINFORMATION. Make an unforgettable first impression. People are going to say, you look younger. What's your secret? It's no secret, Caldera Labs, the leader in men's skincare, made only with top-tier ingredients. Your skin deserves only the very best. CalderaLabs.com, again, use the promo code DISINFORMATION. In the Enron case, I played the soundbite of the analyst who challenged Jeffrey Skilling about Enron's lack of transparency. That should have been a red flag to more people. In the Bernie Madoff case, there were also red flags. As early as 1999, nearly a decade before his arrest, at least one sharp financial executive smelled a rat. That executive was Harry Markopolos. From 1999 to 2008, he and his team of financial sleuths uncovered disturbing information about Madoff. They took it to the Feds, the Securities and Exchange Commission, on multiple occasions.

10:49 Harry Markopolos Markopolos later testified before Congress. Inexplicably, the SEC never acted upon those repeated and multiple warnings over a nine-year time span. And my formal written testimony makes clear the SEC is over-lawyered, has too few staff with relevant industry experience and professional credentials to find fraud, even when a multi-billion-dollar case is handed to them on a silver platter.

11:08 Paul Brandus In retrospect, this is one of American capitalism's most tragic I-Told-You-So stories. There were warnings, lots of them, and nothing was done. Markopolos also accused the SEC of being unqualified

11:28 Harry Markopolos to do its job. Unfortunately, the SEC staff lacks the financial expertise and is incapable of understanding the complex financial instruments being traded in the 21st century.

11:40 Paul Brandus He also charged that the SEC was afraid to take on Wall Street, fearful of the powerful industry that it was tasked with regulating. So with all of that in mind, here's my question. If the watchdogs didn't understand what Madoff was doing, or as Markopolos charged, chose not to get involved, what might this say about something far more complex than a mere Ponzi scam, like algorithms, bots, high-frequency trading, and the possibility of bad guys combining all of this with AI to spew financial disinformation? Or, to be more optimistic, could AI play a role in rooting all of that out? One area in which AI could help concerns the 10K forums that publicly traded firms are required to file with the SEC. A 10K is where you can find just about anything concerning a company. It's a goldmine for savvy investors. You can learn about a company's revenues, debts, management, operations, and plans, areas of concern and competition. A 10K, in short, is a great way to examine a company from the inside out, and the government makes all of this information available free to you. Just go to sec.gov forward slash edgar, E-D-G-A-R, edgar. Most of the stuff you see in a 10K, by the way, is text. Craig Lewis, the SEC economist turned finance professor, says this stuff tails with the potential of artificial intelligence. You see, AI uses something called large language models, or LLMs, that can sift through huge chunks of text and perhaps ferret out hints of possible malfeasance.

13:56 Craig Lewis So in addition to having these quantitative metrics based on ratios, for example, you can also try to take unstructured data, like text, and put structure around it and analyze the structured data itself. And so we, I've been doing a lot of that as well, and it's really interesting. I think the natural language processing component to fraud detection and financial results is kind of right in its infancy.

14:25 Paul Brandus This could be useful because while publicly traded companies are required to file 10Ks with the feds, as I mentioned, there's no requirement that the text has to be clear as a bell and easily discernible to the human brain. Even when your intentions are good, writing clearly can be no easy task. Thus, while artificial intelligence and its ability to sift through huge blocks of text holds promise, as Professor Lewis says, it could also hold peril. Once again, here's Meredith Wilson.

14:58 Meredith WIlson One of my worries is when we think about the markets and using AI is not even necessarily intentional disinformation, but that, you know, that misinformed large language models might be used to automatically make financial decisions that could, you know, in the worst case scenario, that could make a decision that pulls a whole bunch of money out of one place at once, right? Where you have suddenly, you know, I think about the Silicon Valley bank collapse, for example, right? But what if that were just an automatic response to, you know, some detection of weakness in the financial system, right? And suddenly all of this money is being pulled out of the system by an automatic process designed, you know, possibly designed not for that purpose, but without a human in the loop to say, hey, wait, don't do this, right? Because as AI gets faster, it makes those decisions faster. In a lot of cases, what people worry about is that those decisions will be made so fast that we don't have time to react before the damage is done.

16:09 Paul Brandus Well, isn't that already the case? I mean, you look at sort of the flash crash back in what, 2010, I mean, most trading even back then was done without humans, done by algorithms that are looking for certain trends and patterns and so forth. So it's already done automatically and faster than humans can react. I suppose now it'll just be all of that on steroids.

16:39 Meredith WIlson Right. Well, and that's the concern, right? I kind of feel like sometimes when we look at, you know, at some of the things that are going on with AI on the negative side and, you know, acknowledging that there's a lot of positive trends here too. But on the negative side, are we standing on the edge of a cliff here? Are we just waiting for that one moment to happen? And, you know, most people who are looking at AI right now will tell you, yeah, we're definitely going to have some missteps. We're definitely going to have some hard lessons. I just hope that's not one of them.

17:10 Paul Brandus As I speak, someone somewhere is no doubt using AI to manufacture and spread disinformation about something, like pictures, audio, video, text. If regulators ignored warnings about the scams that were Enron, Worldcom, and others, if they allowed a giant Ponzi scheme to operate for years in plain sight, why should we think that AI-powered disinformation can and will be properly controlled? Have a tip, idea, or example of disinformation you'd like us to check out? Contact me personally at [email protected]. Thanks to Professor Craig Lewis of Vanderbilt University for his insights, sound from CNBC, ABC, and C-SPAN, our sound designer and editor, Noah Foutz, audio engineer, Nathan Corson, executive producers, Michael Dealoia, and Gerardo Orlando. And on behalf of Meredith Wilson, I'm Paul Brandus. Thanks so much for listening.