Project Spotlight: Legal Automation Toolkit
Published: May 4, 2023
As winners of the 2023 Digital DNA Digital Innovation of the Year Award, we wanted to share some more detail on what made the Legal Automation Toolkit carried out by our Data & Intelligent Systems (D&IS) ML R&D Team the worthy recipients of this prestigious award!
Business Need
The 18-month long project was in partnership with Allstate's US Department of Law & Regulation.
Inspiration for this project came about after meeting with legal teams on the ground who highlighted their frustration at some of the repetitive and manual tasks they had to complete, often taking time away from more complex legal matters.
After identifying and capturing ideas the team prioritised them based on these criteria to help streamline processes:
- Impact on our legal team's job satisfaction
- Potential time savings to allow time to be spent on more valuable tasks
- Complexity (can it be solved by existing vendors/technology)
From research the team knew a major challenge is that each jurisdiction/state in the US has different legal processes. From the outset, the team engaged heavily with our legal partners throughout the project to ensure they had full understanding of the requirements. The technical solution involved building an initial proof of concept in a way where most of the components were reusable as it was rolled out countrywide. The team used an array of tools, including AWS Textract for OCR, NN NLP classifiers from Huggingface and custom NLP classifiers. The tool is deployed on Cloud and integrates with RPA bots and databases.
The Solution
The completed toolkit automates the gathering of relevant data, the generation and completion of legal documents and the correct placement of the documents in the legal departments existing document management platform.
It leverages multiple tools and techniques including custom AI/ML models, RPA and Cloud.
The toolkit has two main components:
Claim File Summary: It is a requirement of the court to provide a claim summary. Traditionally this was done by an attorney or paralegal who read and searched 100s of pages of file notes, police reports, witness statements and more. This data was spread over multiple systems and in varying formats (which could differ by jurisdiction). This process can take minutes to days but on average around 1 hour. The goal of this project was to automate the gathering of this data in a central location and provide a summary in the form of a word document that consisted of around 30 fields in a standardized format across every jurisdiction.
Legal Answers Generation: At the beginning of a case Allstate is served with a summons & complaints petition. This outlines the case being made and has a list of allegations in the form of paragraphs. In the traditional process the legal team has to address each allegation and format a response document to submit to the court. Using AI/ML tools the team deployed a solution that pulls in the relevant documents and uses ML to generate the final correctly formatted response document on which the ML/AI model predicts the attorney's responses. This reduced the end-to-end processing time from 50 to 5 minutes per case.
Project Success
To date the team have rolled out the Claim File Summary Component countrywide and launched the Legal Answers Component in 9 jurisdictions.
- Time Saving: Analysis shows that the tool saves around 75k hours per annum which equates to millions of dollars in time savings.
- Improved Job Satisfaction: The feedback the team has received so far has also been extremely positive with it having a huge impact on job satisfaction of the 1000's of legal professionals that Allstate employs.
- Accuracy: Continuous testing and validation of the models have shown them to have high accuracy & the documents generated are at human level.
