The Legal Data Intelligence Podcast with David Cowen (Episode 20)

In this episode, Robert Keeling places ediscovery—marked by linear, document-heavy processes—as a precursor to the emergence of Legal Data Intelligence (LDI). He talks to host David Cowen about LDI as a framework and guiding philosophy that empowers legal professionals to navigate complex digital ecosystems, enabling nimble firms to collaborate across disciplines and deliver smarter and more scalable client service.

Keeling focuses on how modern communication formats—like Slack, emojis, and GIFs—pose fresh challenges for discovery, privacy, and privilege, and how generative AI models can be used to solve them.

Listen to the full episode and read a partial transcript below.

David Cowen: How do you see Legal Data Intelligence? What value can it bring?

Robert Keeling: LDI, I think, is a great idea to get actual structure around and give meaning to and definitions to something that’s always existed, right. To actually give meaning to it; it helps define the concept, and it helps leverage and use it for our clients, but also internally.

And so, I think, when we're thinking about Legal Data Intelligence and how it can be used, I think of it in two ways. I think of it from a client perspective with things like information governance or how they bring data to bear. And then I think of it from a firm perspective—how a boutique firm like Redgrave can use LDI to be more efficient and more effective.

Looking backwards on the evolution of ediscovery, are there any lessons that you can draw that are applicable today?

There was a time where the amount of email just kept increasing exponentially every year. It seemed like this intractable problem: How to collect, preserve, review, and produce all this email? But now, when you think of email, it’s the least of your worries, right?

Now, we have a lot of technology that helps us go through all the different steps very efficiently and effectively. And we're developing more of it—you know with large language models—to make the process even more efficient.

But I think as data gets more complicated with emojis, GIFs and videos, there will be a fundamental need to do things more efficiently, and cost-effectively. I think the advances we are seeing with generative AI, in particular, will make it easier to navigate new and emerging data sources more effectively.

Can you drill down a little bit on how you will be using LDI?

As an example, if you go to the LDI website, there's a section on internal investigations and they have an example: second requests. I do a lot of second requests; I do a lot of antitrust work. We have a whole playbook for a second request. But it doesn't just exist as a Word document; it's in Asana, a different platform that helps us effectuate and bring order to the chaos of data productions. And so being able to analyze over time our work on different second requests; to pull in, not just best practices, but best precedents; thinking about how we are going to effectuate it, and the tools we are going to use. Creating a team and a particular process around — I think that paradigm of thinking will have a direct correlation to success.