Shannon Capone Kirk runs the e-discovery practice at Ropes & Gray, a prestigious global law firm with panoramic views of the Boston skyline. In order to understand what she does these days, she insists you need to understand what life was like when she entered the profession in the late ’90s; her first job was “document review.”

copyright by www.wbur.org

“What that meant was literally spending weeks upon weeks in either a warehouse or a conference room flipping through bankers boxes and reading documents, paper documents,” said Kirk. Back then, every large corporate law firm used an army of first-year law grads for this sort of paperwork. “And if we found something that was relevant to the litigation we would tag it with Post-it notes, and that was it,” said Kirk. “That is how archaic it was.” Document review was time-consuming and expensive but essential for the discovery phase of any legal trial.

The Software Revolution

In the early-to-mid 2000s, technology began to nibble away at certain legal tasks, as law firms experimented with software that could streamline this document review process.Law firms would collect gigabytes of data, put it into a review platform, and then ask lawyers to run search terms.

Kirk says this system was advanced, but still “inefficient,” and relied heavily on lawyers to sift through the results. But in the last couple of years, e-discovery platforms, such as Relativity, have become more popular, and the algorithmsAn algorithm is a fixed set of instructions for a computer. It can be very simple like "as long as the incoming number is smaller than 10, print "Hello World!". It can also be very complicated such as the algorithms behind self-driving cars. they use have become much more sophisticated — they’re no longer solely reliant on search terms. It’s the machine learning how to prioritize what documents a lawyer finds — essentially, artificial intelligenceArtificial Intelligence knows many different definitions, but in general it can be defined as a machine completing complex tasks intelligently, meaning that it mirrors human intelligence and evolves with time..

Humans still set the parameters. But computers whittle down those millions of documents by using predictive coding. Kirk explains that predictive coding is essentially akin to the “thumbs up” button on Pandora — the lawyer trains the software to find what it’s looking for. Lawyers can now sift through gigabytes of data and find “300 key documents” within a week, according to Kirk. “Gone are the days when you would staff 50 to 75 first- and second-year associates to a document review, that just does not happen anymore,” she said. “You could never do that before. Ever.” […]

read more – copyright by www.wbur.org