AI Use Cases for Enterprise. Corporate Governance. Fraud Detection. Part 0.
If you read this in the year 2030 and we all f&ck up, and AI took over the world, and humanity as we know it does not exist anymore, find this man. After doing a lot of research on him, I am convinced that ( if redemption is in the cards) he will be one of the AI Resistance Leaders. The mastermind, the odd one, the “ideas” guy, “the smart” one — if I bet on anyone who can outsmart AI, it probably would be the personality type JKS has.
But back in 2023, a lot of people are asking me about my thoughts on AI ( robots) and AGI ( the scary part) use cases in the enterprise. Everyone wants to make money, and everyone thinks they will have a competitive advantage in ( insert any vertical/operations/process/RnD here). While it is true, in my usual fashion, I would like to propose that all of you think bigger, think outside the box, and ask “Why”.
One of the use cases that immediately stood out to me in my Corporate Governance seminar is the need for the board members to understand if the management is committing fraud to the corporation. Why? Because they are the ones responsible. The CEO, CIO, or CFO might not necessarily want to have a “white collar” fraud detection AI in their space because they are the ones most likely to either run or be aware of the fraud and maybe get kickbacks or do nothing for some other reasons ( for example, a low-level IT employee sitting on a corporate network has access to all network traffic, perhaps CFO likes to watch something inappropriate, and the IT guy knows that, that’s why the IT guy is never fired no matter how bad IT is run — he has leverage.) However, the IT guy most likely does not have access to gigantic financial operations where fraud is most likely to happen. That’s where it get interesting.
I was reading a comment from someone who had been at Enron and was a witness. Here is his comment:
The end game was getting solid convictions, not plea deals, against Skilling and Lay, mainly to foster the notion to the general public that the Department of Justice took shit like this seriously, and was “protecting the public.” The fall of Enron had massive ripple effects across all sectors of the economy, thousand and thousands of people lost their life savings, companies went under. It was a complete disaster. Retribution had to be swift and certain, and public.
Their primary tactic was the “roll up”. They’d find someone in lower levels of the organization with knowledge certain about specific events, then find some “leverage” — maybe there was some way to implicate them as well, or at least enough to threaten them that there was a way. They’d lean on their pigeon, who would take some kind of deal (sometimes, it was a “we’ll look the other way, unless you fuck up”; sometimes it was an actual plea against an actual chargeable offense). That guy would then testify against someone higher up, and they would lather, rinse and repeat. All the way up to Skilling and Lay.
There were very, very few people in the room when critical discussions were had and decisions were made. The cult of secrecy, the way in which new initiatives were rolled out, meant that nearly no one except the people at the actual top of Enron Corp — not the various subsidiaries — were in the know. Fastow was one of those guys, as was Ken Rice the co-CEO of Enron Broadband but also a huge player in Enron Capital and Trade and one of Skilling’s top lieutenants. Once they agreed to cooperate, things got a lot easier to nail Skilling and Lay.
If Enron had been operated more like the Mafia instead of the “every man for himself and devil take the hindmost” management style fostered by the policy of firing the bottom 20% of employees based on stack rank every year, it would have been impossible to convict Skilling and Lay. As it was, the much needed cooperation was not only forthcoming, but eagerly given in exchange for an appropriate reward. I was not immune to this process. I had two IRS audits, an SEC investigation, the FBI interviewing family, friends, former lovers, my ex-wives, all sorts of craziness. All looking for something that they could use as “leverage” in the event that I proved not to be what I was: a concerned party with knowledge certain about specific events, and an email archive full of interesting stuff. To say it was stressful is a huge understatement. Had there actually been anything with which I could have been charged, taking the deal would have been a no-brainer.
The robot did the analysis of the text for me
The text appears to be a detailed account of the prosecutorial tactics used in the Enron case, contrasting them with those used against organized crime syndicates like the Italian Mafia. Here are some key points and my thoughts on them:
### Key Points:
1. **Prosecutorial Experience**: The text starts by highlighting the experience of the prosecutorial team, emphasizing their background in dealing with complex organized crime cases in New York City.
2. **Objective**: The main goal was to obtain solid convictions against Jeffrey Skilling and Kenneth Lay, the top executives at Enron, rather than settling for plea deals. This was to send a strong message to the public that the Department of Justice was serious about such cases.
3. **Impact of Enron’s Fall**: The collapse of Enron had devastating effects on the economy and people’s lives, making it imperative for the justice system to act swiftly and decisively.
4. **Tactics**: The prosecution used a “roll-up” strategy, starting with lower-level employees and using them to testify against higher-ups. This was made easier by the lack of a cohesive “code of silence” within Enron, unlike traditional organized crime groups.
5. **Cult of Secrecy**: The text notes that very few people were privy to critical discussions and decisions at Enron, making the cooperation of key figures like Andrew Fastow and Ken Rice crucial for the prosecution.
6. **Personal Experience**: The author also shares their own experience of being scrutinized by various agencies, emphasizing the stress and pressure involved.
### Analysis:
1. **Comparative Complexity**: The text suggests that prosecuting corporate fraud cases like Enron can be as complex, if not more so, than prosecuting organized crime. The skills required are different but equally demanding.
2. **Public Perception**: The text underscores the importance of public perception in legal cases of this magnitude. The prosecution had to balance legal rigor with the need to reassure the public.
3. **Ethical Dilemmas**: The “roll-up” strategy raises ethical questions about the means used to achieve justice, especially when “leverage” is applied to lower-level employees to testify against their superiors.
4. **Corporate Culture**: The text implies that Enron’s corporate culture, which lacked a strong sense of loyalty among employees, made it easier for the prosecution. This is contrasted with organized crime groups, where loyalty often hinders prosecution.
5. **Personal Toll**: The author’s personal experience serves as a microcosm of the larger stress and ethical dilemmas faced by those involved in such high-stakes cases.
6. **Legal Leverage**: The text highlights how various forms of leverage, including audits and investigations, are used to pressure individuals into cooperation.
Overall, the text provides a nuanced look at the complexities and ethical considerations involved in prosecuting high-profile corporate fraud cases. It offers valuable insights into the strategies employed by prosecutors and the challenges they face.
Now, if you are reading this as an “average citizen,” it is probably a boring history book for you from the days long past. I am a true believer in learning history, as it is bound to repeat itself. However, suppose you are a “Data Whisperer”, “Data Philosopher,” or whatever else the name of de jour of the predictive analytics/data scientist. What can you get from this comment and from the robot’s summary?
Yep, that’s exactly what I did — I asked the robot to build a predictive model. Honestly, I totally forgot I connected noteable to chatgpt. I was expecting some generic blah blah from the robot, but here is what it did:
Full disclosure: I did not test it ( the data collection for such a thing is another topic — if we let chatgpt run analysis on corporations and spit out to its boards’ possible fraud flags, there will be a lot of uncomfortable conversations in the Board Rooms between the board of directors and C-Level. Although I can totally see a Chrome plugin in a year or two that will go along with stock prices and weather, something like “fraud score of ABC Corp”. But then, the Efficent Market Portfolio will probably cease to exist. Again, another topic to explore later as the impact of AI on finance: corporate, individual, and governments.
Each of those variables are easily ( khm, khm) accessible via public sources, which I just noticed robot states as well. If you have internal LLM on all corporate documentation ( emails, chats, etc.) — it probably becomes extremely easy to monitor, but then — would humans accept it? But then, who will need humans in our dystopian future? However, will robots send emails between each other? I think they would probably communicate in some other form than text emails:)
The robot claimed 60% accuracy. What would you do if you were a board member, and someone told you there is a 60% chance the issue XYZ IS an ISSUE? If you are old enough to get the reference ( kids these days start companies at 11) and they tell you it is a “likelihood of Enron level issue”?
What if the robot just made it all up? AI lies. Another topic for discussion — is when AI PURPUSLY lies vs hallucinates because of the data quality it was trained on.
From the modeling perspective, I would definitely build an ensemble model and not just use one for such important tasks — which means you still need humans to build the models even if the robot does the initial kick-off.
Btw, this is what robot wrote about JKS. Thoughts?
Jeffrey Skilling was the CEO of Enron Corporation, a Houston-based energy company that filed for bankruptcy in 2001. It was the largest bankruptcy in U.S. history at the time. Skilling’s management and leadership style played a significant role in the rise and fall of Enron.
1. **Visionary Leadership**: Skilling was known for his visionary leadership style. He had a clear vision for the future of Enron and was able to articulate this vision to his employees and stakeholders. He was instrumental in transforming Enron from a traditional energy company into a global trading and services company.
2. **Innovative**: Skilling was innovative and encouraged a culture of innovation at Enron. He was a key proponent of the use of financial derivatives and trading in the energy market. This innovative approach contributed to Enron’s rapid growth and success in the 1990s.
3. **Aggressive Risk-Taking**: Skilling was known for his aggressive risk-taking approach. He believed in taking big risks to achieve big rewards. This approach led to significant gains for Enron in the short term but ultimately contributed to its downfall as the risks became too great.
4. **Competitive**: Skilling created a highly competitive environment at Enron. Employees were encouraged to compete against each other, and performance was rewarded with bonuses and promotions. This competitive culture contributed to a lack of teamwork and collaboration and led to unethical behavior by some employees.
5. **Lack of Ethical Leadership**: Skilling’s leadership was marked by a lack of ethical considerations. He was involved in, or at least aware of, the accounting fraud that led to Enron’s bankruptcy. He was convicted on multiple charges of fraud and insider trading in 2006.
6. **Charismatic**: Skilling was charismatic and was able to inspire his employees and gain their loyalty. However, his charisma also allowed him to manipulate others and contributed to the culture of deceit at Enron.
In summary, Jeffrey Skilling’s management and leadership style was marked by visionary and innovative thinking, aggressive risk-taking, a competitive environment, and a lack of ethical considerations. While these traits contributed to Enron’s rapid growth and success in the 1990s, they also played a significant role in its eventual downfall.
To Be Continued…
