As a data-intensive profession, the legal trade relies on in-depth literature review. The long hours spent poring over documents make the practice a candidate for efficiency improvements.
McKinsey & Company, the global management consulting firm, predicts more than a fifth of a lawyer’s job and about a third of a law clerk’s can be automated.
With the aid of machine and artificial intelligence, tech companies have set their sights on augmenting, if not overhauling, the legal grunt work behind the scenes.
Tech startups claim they can speed up the review process and analyze the facts better thanks to their super computers.
Some have introduced document review platforms that examine contracts and case filings, which would otherwise take lawyers and paralegals hundreds of hours to scrutinize.
Powerful software can also predict the outcome of a case – by evaluating arguments and counter-arguments through AI – even before they are presented in court. This level of sophistication is designed to improve research.
Legal foresight for HR?
Toronto-based Blue J Legal is one example of an AI-powered case review and prediction platform.
“Our methodology converts unstructured data from thousands of court decisions into structured data,” said Benjamin Alarie, CEO of Blue J Legal.
The company’s proprietary deep learning software is said to have been refined by computer scientists and law professors from the University of Toronto. It purportedly identifies, in seconds, patterns hidden in datasets to predict how a court will rule in new cases.
The startup’s Employment Foresight tool is the first ever outcome predictor specializing in employment law. The most common issues addressed by Employment Foresight include:
- Reasonable notice – what is the reasonable length of notice that an employer must provide to a worker upon dismissal?
- Worker classification – is an employee an independent contractor for employment law purposes?
- Cause for dismissal – is there just cause for dismissing a worker?
- Overtime eligibility – is an employee subject to overtime pay requirements?
“In the hands of an HR professional, the software works by first asking for a series of inputs through a short, plain language questionnaire,” Alarie told HRTechNews.
The software compares the input to previous cases then pulls up data on the possible ruling. A corresponding confidence level, expressed as a percentage, predicts the likelihood of the outcome. The report includes an explanation of the predicted ruling along with similar decisions.
“When tested against cases that the system has never seen before, Employment Foresight is able to achieve 90% or greater accuracy,” Alarie said. “The software is also updated with new decisions as they are published, enabling the system to improve its predictions and provide up-to-date outputs.”