Inside Risk: How Paul Boynton is Redefining Global Due Diligence with AI and Deep Data | Ep 1106

Summary

---

The Digital Executive Podcast

Episode: Inside Risk: How Paul Boynton is Redefining Global Due Diligence with AI and Deep Data
Guest: Paul Boyington, Co-founder & COO, Company Search Inc. (CSI)
Host: Brian (Coruzant Technologies)
Release Date: August 30, 2025
Duration: ~12 minutes

---

Episode Overview

In this episode, host Brian interviews Paul Boyington, an industry veteran and co-founder of CSI, a next-generation business intelligence platform. Paul shares insights from his extensive risk management and insurance background, and how CSI leverages AI, deep data, and proprietary tools to transform global due diligence, address compliance complexity, and support companies navigating fast-shifting regulatory environments like the new BIS 50% rule. The discussion elucidates CSI’s unique approach to data integrity, scalable AI, and entity mapping in international trade and compliance.

---

Key Discussion Points & Insights

1. Paul's Career Origins and Vision for CSI

[01:20 – 05:03]

• Paul describes his career trajectory from underwriter at SGI Management Co. to negotiating reinsurance treaties and then managing risk for Fortune 100 companies at Johnson & Higgins.
• His early international and operational experience highlighted the crucial role of in-depth, boots-on-the-ground due diligence, especially for challenging or emerging markets.
• Quote [04:27]:
  • “That is the passion of that next-level due diligence...that’s behind what we’re trying to do here with this CSI platform—provide that level of due diligence, but you don’t have to get on a plane and you don’t have to get the information yourself.”
    — Paul Boyington

2. Human Connections & Data-Driven Analytics

[05:05 – 06:30]

• CSI aims to bridge the gap between pure data analytics and the irreplaceable insights that come from local human relationships.
• Their platform aggregates data on backgrounds, ownership, media, operations, performance, and even fills coverage gaps where major credit agencies like Moody’s and S&P don’t operate.
• Quote [05:45]:
  • “We’re not replacing a person, but we’re trying to provide gaps of info...the insight that you’re not looking at, maybe you should be looking at as you ascertain and analyze risk.”
    — Paul Boyington
• CSI’s analytics can let clients replace arduous travel and site visits with smarter, data-augmented remote decisions.

3. AI as a Tool—Powered by Quality Data

[06:54 – 09:16]

• Paul notes that true value in AI depends on data quality, not just algorithms:
  • CSI’s strong data quality index underpins every new AI feature.
  • The in-house large language model 'Zaba' sources solely from proprietary, cleaned datasets—unlike general LLMs scraping the entire internet, Zaba gives context-rich, industry-specific outputs.
  • Example: For a query on Turkish textile mills, Zaba immediately surfaces financially validated, detailed candidate companies from CSI’s deep database (vs. generic web search links).
• Despite challenges scaling with a small team, CSI continues to improve anomaly detection, analytics, and LLM-powered functions, always balancing domain expertise and technical resources.
• Quote [07:39]:
  • “Zaba is our AI mode...extracting info just from our data pool...Whereas our database and Zaba have every textile mill in Turkey...and it will come back and say based on your criteria, these are some things you need to think about and these are five good ones you may want to do business with.”
    — Paul Boyington

4. Navigating New Compliance Rules: The BIS 50% Example

[10:07 – 11:45]

• The introduction of the US BIS 50% rule raises stakes for ownership transparency in global supply chains (e.g., Vietnamese firms with Chinese control now subject to tariffs).
• CSI’s proprietary tools for entity mapping and ownership penetration research are designed precisely to reveal these hidden links (“ownership of the owner, subsidiaries, affiliates”) much more efficiently than traditional methods.
  • Coverage spans extensive customs data from 33+ countries and comprehensive trade information well beyond official reports.
  • CSI is positioned as an industry leader in this niche.
• Quote [10:51]:
  • “We are probably the industry leader in ownership penetration research...every company in China that does business in the US—we know who they’re selling to and we know who they’re buying from.”
    — Paul Boyington

---

Notable Quotes (with Timestamps)

• On next-level due diligence
  “That is the passion of that next level due diligence. That is behind what we’re trying to do here…”
  — Paul Boyington [04:27]
• On the role of human vs. machine
  “We’re not replacing a person, but we’re trying to provide gaps of info...”
  — Paul Boyington [05:45]
• On AI and proprietary data
  “Zaba is our AI mode that is extracting info just from our data pool...whereas our database and Zaba has every textile mill in Turkey…”
  — Paul Boyington [07:39]
• On regulatory shifts and CSI’s edge
  “We are probably the industry leader in ownership penetration research...every company, China that does business in the US—we know who they’re selling to and we know who they’re buying from.”
  — Paul Boyington [10:51]

---

Key Segments & Timestamps

| Segment | Timestamp | |--------------------------------------------|-------------| | Career Path & Origins of CSI | 01:20–05:03 | | Human Element in Data-Driven Analytics | 05:05–06:30 | | AI, Data Quality & CSI’s Proprietary LLM | 06:54–09:16 | | Regulatory Landscape: BIS 50% & Compliance | 10:07–11:45 |

---

Memorable Moments

• The candid discussion on the limits of current LLMs vs. domain-specific tools like Zaba [07:39].
• Illustration of CSI’s real-world impact by detailing how their tools uncover hidden trade/ownership relationships now vital under new export rules [10:51].

---

Conclusion

Paul Boyington’s interview offers a deep dive into the transformation underway in global risk and due diligence. Through stories from his insurance career and technical details about CSI, listeners gain a nuanced understanding of the interplay between human insight, AI, and proprietary deep data in navigating today’s complex compliance environment.

For those managing risk, compliance, or international supply chains, this episode is a concise, insightful resource on the next era of business intelligence.

Transcript

[Brian]

Foreign welcome to Coruscant Technologies, home of.

[Host]

The Digital Executive Podcast. Welcome to the Digital Executive. Today's guest is Paul Boyington. Paul Boyington is co founder and chief operating officer of Company Search Incorporated, AKA csi, a next generation platform for business intelligence and business analytics. Paul's expertise as a developer, implementer and provider of information emerged from decades of utilizing data driven decisions, servicing industry leaders, multinationals and subject matter experts in the corporate insurance space, driving programs for operational and risk management. Results are career hallmarks, facilitating emerging market sustainability and growth for the likes of Dell, McDonald's, Walmart and leaders in the commodities and chemicals industries. A testament to a career that secures risk sharing partnerships to facilitate large capital investments. Paul has worked on all sides of the insurance matrix, learning the trade as an underwriter and analyst with fcia, negotiating reinsurance treaties with Willis Faber and retail corporate relationships with Johnson and Higgins. Well, good afternoon Paul. Welcome to the show.

[Paul Boyington]

Hey, thanks Brian. It's great to see you again. I really appreciate you having me back.

[Brian]

Absolutely, my friend, I appreciate it. And you're still in Dallas, I'm still in Kansas City. And we probably talk about barbecue rivalry or football, but let's get into something more exciting. The stuff that you're doing right now. If I could, I'm gonna jump into your first question. Your career path spans underwriting at fcia, negotiating reinsurance at Willis Faber, and direct client relationships at Johnson and Higgins, culminating in the creation of csi. Could you walk us through how these experiences shape your vision for building a next generation business intelligence platform?

[Paul Boyington]

So, you know, my first job out of college was at the SGI Management Company, which is a work credit insurance company. And that's really where it started for me. And I got involved with international business. But you know, I was a green visor underwriter. Right. So it was my first look at, you know, operations financials of foreign companies. And I did that for a few years and you know, understood that process and I jumped over into the reinsurance side of the business, which is a step removed. Reinsurers, you know, insurance, drugs companies. Right. So reinsurance treaties are the backline of insurance risk takers. And yeah, so my background in the trade credit insurance business played a big role in our team that was placed in reinsurance treaties for various underwriters of export credit and political risk insurance. So did that for a few years and then this is all in New York, ended up working at Johnson and Higgins, which has become Marsh McLennan quite a while ago. But you know, this is now the, the late 90s and I was part of a national resource team in the trade political risk insurance area. So you know, kind of think of it like in those days we were sort of like a modern day platform instead of looking on the CSI platform, but you do a 200q5 in those days. You would pick up the phone and he called actual people to come in and provide expertise on lanes of business that they didn't have the expertise to do. So we would come in and help them get their clients taken care of. And what we did, and ultimately I did work for Fortune 100 companies. My gig was to help our clients do more business, do more trade, do boots on the ground operations in those countries with the support of insurance capital. By going through those experiences, the insurance companies didn't want to get involved with it and those you know, sort of surplus level of due diligence, you do have counterparties doesn't really cut it when you're talking about these different Latin America or Asian countries that we would have to go the extra mile and bring these capital providers on site to see these kind of parties, learn all about them, do deep dive behind the curtain, look at the owners operations at the business cycle, all that stuff. So we could, you know, develop the trust you needed to get a build on to get a new partner. And that's really the passion of that next level due diligence. That is the behind what we're trying to do here with this PSI platform, provide that level of due diligence but you don't have to get on a claim and you don't have to get information yourself that we can provide for you.

[Brian]

Thank you, I appreciate that, that backstory there, you've got just a plethora of experience in this space, insurance, insurtech, et cetera. But I like the fact that you, you know, right out of college you were an underwriter, right. And you moved into reinsurance and then you have national, international experience, a lot of work in the extensively in the due diligence space. So I really appreciate that and I know all that contributes to what you.

[Host]

Do today with your CSI platform.

[Brian]

Paul, you mentioned that strategic insights often stem from local partners and human relationships. How do personal connections and on the ground knowledge complement CSI's data driven analytics?

[Paul Boyington]

We do a lot of the data points of due diligence, right. So background operations, ownership, performance, media. We have a GMU at Dow Jones, so we have the media name, what's going on the public space with whatever kind of party you're working for, credit opinions, we do credit Ratings that we provide in countries that Moody's and The S&Ps don't provide, but they actually acknowledge our credit opinions and ratings and use them. And you know, we do industry and market analytics, so we can't really, we're not replacing a person, but we're trying to provide gaps of info that maybe we're going to have that you can't get elsewhere. But the insight that you're not looking at, maybe you should be looking at as you ascertain and analyze risk. So we're not going to be that person that takes you around the factory or takes you around a customer. But what we want to be is a tool that can help you get far enough to maybe you can get it done with a couple of zoom calls with your local partners and you're going to have to jump on that plane or, you know, we have enough information that gives you enough insight that you trust that you're making the right decision.

[Brian]

Thank you, I appreciate that. Absolutely. And there's just so much data these days to glean through when you're doing due diligence, as an example. But you know, humans can make mistakes and machines can too, but machines are.

[Host]

Less likely to if they follow their.

[Brian]

Proper programming and algorithms, et cetera. But I like how you are leveraging business intelligence and, and those insights to fill the gaps, to help again that.

[Host]

Partnership, as you would say.

[Brian]

I really appreciate that. Paul, you've stated that AI is only as powerful as the data behind it. And CSI emphasizes a strong data quality index. How do you maintain data integrity while scaling AI driven features? And what's next for CSI's AI roadmap?

[Paul Boyington]

Thanks for prompting that one because that's an exciting thing about us. So you know, we look at that trim AI little because we've been in a machine learning business for almost 10 years now, so we've been hearing AI for a long time and we kind of look at that phrase and go, eh, you know, it's super overused and all that. Right. But yeah, it is different. It's totally different now that we were doing machine learning models 10 years ago and we have a large language model module on our site now. It's called Zaba. And Zaba is our AI mode that is extracting info just from our data pool. Right. So when you go on Grok or ChatGPT or something, it's extracting data from the whole Internet except our data pool. So if you ask, you know, Grok about textile mills in Turkey that, you know, make genes like Levi Strauss and they've been around 10 years and they're profitable to stuff like that. It's not really going to have info on the different textile metals in Turkey. It'll give you different websites to look at and research, whereas our database and Zaba has every textile mail in Turkey in our database and it will come back to stay based on your criteria. These are some things you need to be thinking about and these are five good ones that you may want to be doing business with. And so we'll give our subscriber everything, financials, performance, contact info, the whole gig. So it's very cool, very insightful tool that we're excited about. It's new, it's going to improve, but for now the results have been very encouraging. And these are things that are difficult to manage because you know, we're a little company, you know, we're going to need to get some more investment and all that stuff. We're always doing that balance between any good domain expertise with skilled data scientists. So we can scale in a way that, you know, anomaly detection and analytics feels along with our data. We have many terabytes of data so it's always a challenge. But we're going to firmly be in that AI space in the business intelligence world.

[Brian]

Thank you. I appreciate that. The last time we talked about this.

[Host]

You did talk about the work you've.

[Brian]

Been doing in machine learning the past 10 years. But so much has changed as you highlighted and your LLM behind your platform. Zaba is really powerful around providing sound financial data behind the research that you're doing. But you did mention you're still going through some growing pains, which is a good thing, right? It's a good problem to have. You're figuring out your way and how to improve and having to add some data scientists, et cetera, to really make your platform even more robust. So I appreciate the insights there. Paul, the last question I have for you today, there are many real world examples of hidden ownership links that could trigger regulatory violations under the new rule called BIS 50%. How is CSI's proprietary data and entity mapping tools help compliance teams get ahead of regulatory complexity?

[Paul Boyington]

Thanks for asking that one. Because that is a new rule that's implemented by the new administration and it's a very important rule and it's something that really emphasizes the need to do that next level of due diligence, particularly in ownership penetration. So the common example is Vietnamese companies that are ultimately owned by Chinese companies or Chinese individuals. So ultimately if that ownership is over 50%, Vietnamese company will be subject to Chinese terrorists. So you have to do the due diligence to not only understand who owns your supplier, but who owns the owner and who their subsidiaries are, who their affiliates are, do one of those companies. And that is an area of hey, that we are probably the industry leader in ownership penetration research. And along with that we're very strong in the importance portrait. So when you're looking at information that's provided about customs houses, we obviously have for I think it's 33 countries now, but we also have trade information on countries that don't provide customs customs involved because we have so many data points. So every company, China that does business in the US we know who they're selling to and we know who they're buying from. So we'll have that info from that customs agency. So those are areas where our stack that's just one or two little examples of the data on our site that is easy to use and provides a lot of value.

[Brian]

Thank you, I appreciate that. And I didn't know that you had that much knowledge and depth and experience in ownership penetration research, trade information international, obviously. But this new BIS 50% rule is important. It does help highlight, you know, those companies that are maybe owned by another entity. And your example here was a Vietnamese company that had maybe more than 50% ownership by China. That's obviously going to change how the tariffs work in and with your research, your platform, you're able to provide that level of information more quickly and more accurately. So I appreciate that and Paul, it was such a pleasure having you on today and I look forward to speaking with you real soon.

[Paul Boyington]

I appreciate it, Brian.

[Brian]

Bye for now.