In complex litigation, legal and financial investigation teams have historically operated in siloes. eDiscovery professionals sift through mountains of unstructured communication data (emails, chat logs, and documents) while forensic accountants independently trace structured transactional data through spreadsheets and bank records.
At RelFest Chicago 2026, a panel moderated by Maribel Rivera, VP of Strategy and Client Engagement at ACEDS, brought together top industry experts to discuss how modern artificial intelligence bridges this gap:
- Sam Taylor, Partner at Mintz Group
- Joey Marks, Senior Director at Morae
- James Sherer, Partner at BakerHostetler
By drawing lessons from historic matters like the Bernie Madoff Ponzi scheme litigation, the panel highlighted how integrated AI platforms are accelerating fact-finding, unifying structured and unstructured data, and fundamentally reshaping case strategy.
The Historical Divide: Lessons from the Madoff Case
To understand the power of modern AI tools, one must first look at the traditional, highly manual workflows of the past. The liquidation and recovery proceedings following the 2008 collapse of Bernie Madoff’s firm stand as a defining example of large-scale financial investigations.
James Sherer reflected on the early physical and technological constraints of managing such massive volumes of evidence. During his work on the Madoff case, he spent a great deal of time sifting through boxes in a warehouse in Queens, searching through legacy systems, and analyzing information that was meant to be fictitious to throw investigators off.
Because funds from the scheme were continuously reinvested, moved abroad, and transferred across accounts, tracing those assets required painstaking effort. Sam Taylor recalled the manual labor required to translate financial findings into actionable legal strategies. “We were running traditional keyword searches, but then you’re still looking at individual documents one by one, tracing patterns and dates as you find them, and reiterating with new search terms.”
Also, a big part of their role was illustrating and diagramming what they found, which included sitting with pencil and paper, drawing lines and circles to illustrate the money trail, and using different tools to turn those into usable graphics. Taylor explained that modern technology has transformed this part of the work. “Now, there are much more advanced tools that allow you to quickly iterate, find the facts you want to find firstly, and secondly translate it into a clear pattern so a clear legal argument can be made.”
The sheer volume meant that case teams faced immense risks. Historically, document review relied on large teams of people performing keyword searches and manual analysis under tight deadlines, creating a high risk of missing important evidence. If a human reviewer missed a single critical document or context clue, a vital piece of the case theory could be lost.
Merging Structured and Unstructured Data with AI
Today, modern AI tools enable rapid data ingestion, pattern recognition, and the ability to link structured financial data with unstructured communications (such as emails and messages) in eDiscovery review environments. Rather than running parallel investigations, legal teams can now correlate financial movements directly with the conversations behind them.
Joey Marks emphasized that today’s specialized AI tools allow case teams to quickly connect intent with financial activity. “Having an ecosystem where you have different AI solutions that you can utilize at different stages in different phases of a project ultimately allows you to put the full story together, present it, and win your arguments.” Additionally, querying structured financial records alongside email and chat threads allows investigators to instantly test hypotheses, which can save a lot of time that would have been wasted on dead ends.
From Production Obligations to Case Strategy
A central theme of the panel was the shift in how legal technology is deployed. Traditional eDiscovery often focuses on meeting court production deadlines and filtering out non-responsive documents. Modern financial eDiscovery AI, by contrast, acts as an offensive tool designed to build a superior case strategy early in the process.
Sherer highlighted the high return on investment (ROI) that early, targeted intelligence offers in financial matters. “The tools can be pricey, but they are a drop in the bucket compared to the billions of dollars that were recovered in this matter.” Furthermore, AI-driven financial investigation tools can help identify hidden assets and greatly expand the scope of the case, leading to much higher recovery.
Sherer added that getting answers faster changes the entire dynamic of depositions and cross-examinations. “You’re never supposed to ask a question you don’t know the answer to,” he said. Using the right solutions lets legal teams become better informed sooner so they can develop the deposition testimony.
Key Takeaways for Legal and Investigative Teams
As law firms and corporate legal departments continue to evaluate generative AI and specialized investigative tools, the panel shared several foundational principles for implementing technology effectively:
- Keep the ‘Human in the Loop’: While AI drastically reduces the time required for data ingestion, pattern recognition, and initial review, human expertise remains essential for verifying findings, checking for hallucinations, and establishing evidentiary proof.
- Focus on Proportionality and Fit: Not every matter involves a multi-billion-dollar fraud. Case teams must evaluate when to use broad enterprise models versus hyper-specific point solutions designed for complex financial node mapping and transaction tracing.
- Reallocate Resources to High-Level Strategy: Shifting away from massive armies of manual document reviewers allows legal teams to stretch budgets and focus capital on top-tier strategy. As Marks summarized, “When you can reallocate budget to the highest-level thinkers and strategists on a team, that provides a tremendous amount of value to the client and helps you actually find the truth.”
By combining modern AI-driven financial analysis with robust eDiscovery workflows, investigative teams no longer have to spend months connecting dots manually. Instead, they can surface critical facts, establish clear narratives, and drive faster outcomes.