In the Hot Seat With Our Testifying Experts:

Serena Morones, CPA, ASA, ABV, CFE

 

How is the firm using custom AI tools in litigation that’s been a game changer?

I think on some level everyone is saying they’re using AI now. And the question is whether someone’s more or less giving a nod to it, meaning casually using it, or whether or not we are really digging deep and trying to understand how it can change the way we do cases.

We have devoted ourselves to understanding how AI can change how we do cases. And the way I’ve done that is I have broken down what are some of the greatest challenges in working up a case. What are the pain points? What are the things that take us the most time?

One of those is understanding what the production is. We’ll get a giant production and we need to understand what’s in that production and how is it relevant to the case we’re building. So I’ve taken this process and these pain points and I’ve imagined, well, what could AI do that would make it better? And we’ve developed tools that get at some of those specific pain points and make that process go a lot faster.

We’ve developed, for example, a document index, a document insight tool using AI and some Python code that gives us very quick insight into giant sets of production, identifying key documents, describing documents, helping us to narrow down what we have very quickly.

We’re working on taking that a step further, and if we have common sets of key documents, that we’ll be able to extract complete data sets out of those documents as a sort of a second part or a subset of that same tool. So that’s an example of just really looking at what is it that we’re doing in working up a case and what might AI be able to do better.

And then we have a full-time person whose job is to develop these custom tools using AI, using his programming skills, using our input of how we work up cases.

So I think the way we’re doing things differently is we’re developing very specific applications that we think bring us quicker insight, a deeper understanding of the evidence, and where we’ve identified, you know, some of the greatest challenges that we face as experts. And we are developing tools that very specifically address those challenges.

 

 

 

What’s an example of how you’ve used AI on a case?

So a great practical example of the power of AI came a few months back on a case where we got a production of documents of 30,000 emails with attachments in two weeks of time to tell the attorney any insights that we had out of that.

The attorney thought it was going to be impossible. They apologized. They said, you know, it’s okay if you can’t do this. And we said, bring it on. Let’s try it.

We used our document, our smart document index tool, to summarize 30,000 emails and identify using logic what would potentially be key documents based on the case. The document index tool narrowed 30,000 documents down to about 1,500 documents. And then I narrowed it down according to timeline at issue, more to a fewer number, and came up with 300 key documents that I reviewed visually, manually reviewed them, and then identified probably about 15 or 20 within those that were absolute smoking gun documents for this case.

We gave that to the client. He was completely blown away that we were able to do that. We did all of that in about a week from the time that we got the 30,000 emails.

So that’s an example of the power of AI and the power of these tools. When I had that experience on that case, I knew our industry would never be the same. I had this vision that this is the future of AI and the power of it to get to insights very quickly is going to be a game changer in cases from here on out.

 

 

How has AI changed what attorneys should expect from a forensic accountant?

I think attorneys need to be looking at whether their forensic accountants are learning the tools and using them in the deepest way possible. Because I can tell you from our own case work that those forensic accountants who are working to master the tools are able to bring meaningful answers in a fraction of the time that we used to be able to. That translates into cost savings.

So I feel like attorneys need to be evaluating their experts a little bit differently. There have been traditional ways to evaluate an expert. Is this expert credible? Are they qualified? Can they do the things that we expect them to be able to do?

But today there’s a new question. How are they doing with their technology adoption? Are they taking too long to work up a case? Are they using manual methods when today’s methods can provide instant summaries and instant insight.

 

 

 

How do we ensure reliability when using AI in our analysis?

A lot of people want to know what do we think about the reliability of AI tools right now and expert work needs to be reliable. And the way I think about it was significantly impacted by a judge, a Ninth Circuit senior judge I heard speak at a conference who said that nothing has changed about how experts or attorneys need to present in court because of AI. That there was 100 years ago in the United States, there’s a concern about the reliability of photographs. And photographs started coming into evidence, and there’s a lot of debate about whether photographs could be manipulated. But what happened is that the parties were responsible to demonstrate that the photos were reliable, and that everybody had an opportunity to try to prove otherwise and the process would vet the reliability of any type of evidence.

And there’s nothing different between AI and this photograph issue in that everybody’s responsibility is, everyone is responsible to show that the evidence is reliable. So as experts, that means we’re responsible to use AI in a way that’s reliable. We’re responsible to understand where reliability breaks down and not rely on AI in a way that we shouldn’t rely on it.

So for example, if we ask AI to do something in a quantitative way, we have an associate review it from top to bottom. We replicate it another way. We’re not taking an output from AI ever and relying on that because that wouldn’t be appropriate. That would be like giving it to a brand new staff and not reviewing it, not having somebody with more seniority review it.

So yes, right now, AI is not fully reliable for quantitative outputs, but it does give us so much power and so much insight and so much computational capacity when we give it very narrow parameters. So what we do is we give it very strict guidelines, very strict prompts and definitions. We don’t ask too much of it. And then we have associates review all of the work and make sure that what we are giving to our clients is accurate.

 

 

 

What’s the advantage attorneys get working with our tech-forward team?

We can identify issues in the case so much faster. The human mind can only process so much information and it takes months and months and months.

So AI is giving us insight that we might have taken six months or a year in a really big complex case to be able to bring to light to attorneys. And I think that helps attorneys strategize and know whether they have a case, whether they need to keep investing in the case.

I am also seeing truly that AI is bringing the cost down because what used to take us six months is taking us weeks now.

 

———————————————————————-

Serena Morones, CPA, ASA, ABV, CFE is an independent damages and business valuation expert. Top commercial and IP litigators turn to her again and again to resolve high-stakes disputes through expert testimony or settlement.

503-906-1579  |  [email protected]