One of the key takeaways from the Legal Services Corporation’s Innovations in Technology Conference earlier this month is that legal services organizations (LSOs) are not wasting any time in applying new technologies like AI and machine learning to access to justice issues.
Whereas LSOs have found past success in reaching clients through basic tools like texting, they are now moving to more advanced platforms like document automation to better streamline internal processes. Some are even going one step further by embarking on artificial intelligence (AI) and machine learning (ML) projects to determine how they can help address the 86% of civil legal problems reported by low-income Americans that aren’t fully resolved.
Access to justice starts with literal access: figuring out how clients best receive, digest, and act on legal information. On the lower-tech end, text messaging has proven to be a successful tool for reaching those in need. Hanna Kaufman, Counsel for Innovation and Technology at the Lawyers’ Trust Fund of Illinois, explains that for LSOs, “most of the people those organizations are aiming to serve have access to text messaging even when they have limited access to other technology. Texting meets people where they already are, using modes of communication with which they’re already familiar, and typically at a very low cost.”
Kaufman helped launched Rentervention, which automates intake and responses to clients seeking help with housing issues via a chatbot. In a similar vein, New York City housing tech nonprofit JustFix recently launched an online learning center to centralize tenant advice. This work builds upon a number of studies detailing how texting information like court reminders can decrease court failure to appear (FTA) rates by 26% and reduce open warrants by 32%, as well as reducing the no-contact rate for online legal help applicants by 6.5%.
“If the legal industry could adopt the use of text messaging in a widespread way, it would enable people without access to other forms of technology or abundant free time during traditional business hours to be able to get help,” Kaufman says. “It would also streamline the way people across socioeconomic statuses receive legal guidance.”
Using Legal Tech Internally
In addition to connecting with clients using apps, LSOs are increasingly turning to new technology internally to produce higher volumes of work. In particular, document automation is becoming the standard for generating repetitive documents. Dorna Moini, the Founder of Documate, explains that “document automation breaks down the economic, geographic, and temporal barriers to justice. Because it takes fewer lawyer-hours to generate the same documents, automation like Documate’s no-code platform expands the affordable, flat fee services that are accessible to those who can’t afford hourly rates.”
Document automation also enhances how LSOs like Bet Tzedek and Equal Justice Wyoming can collaborate by sharing workflows with each other to modify them for their own jurisdiction. For example, a domestic violence platform for Los Angeles can easily be modified by a Northern California legal aid organization to meet the local rules of San Francisco and Oakland and add additional forms of relief, Moini explains. “This type of open collaboration is the future of both document automation and the delivery of legal services.”
Community.lawyer is another automation platform that assists LSOs like New Mexico Legal Aid intake clients, draft documents, and share legal knowledge. Scott Kelly, Community.lawyer’s co-founder, has found that automation has helped LSOs “dramatically scale the impact of their legal expertise. We’ve seen organizations use document assembly apps to go from serving dozens of people to thousands of people in a few short years.”
Adapting to the Technology
As more advanced tools like AI and machine learning become more pervasive, LSOs like ILAO, Pine Tree Legal Assistance, and Massachusetts Legal Aid are now incorporating tools like AI classifiers in their intake flow to direct inbound clients to the most relevant resources. The concept is that by essentially leveraging a massive, dynamic logic tree as a triage tool, clients can quickly and easily find the best resources for their issues.
While the idea is exciting, there are significant challenges in helping clients navigate the world of AI. For instance, AI classifiers have to match clients’ understanding of their own problems in order to be effective, and oftentimes, a client doesn’t know the associated legal term, says Jay Hancock from Pine Tree, adding that if a problem encompasses more than one legal problem, a linear tool simply will not suffice. The logic tree must have the ability to dynamically move between AI branches to address issues across multiple practice areas.
Gwen Daniels of ILAO points out that for maximum efficiency, the classifiers must be the same throughout all of the potential LSOs available to address a specific issue, and given the hundreds of potential criteria organizations use to intake clients, navigating those differences might prove difficult.
One technology that might help address AI’s challenges is machine learning. Specifically, ML has the potential to improve the way clients and LSOs use technology to identify legal issues — one of the biggest barriers to clients seeking help at the outset.
“Generally speaking, machine learning is in the business of making predictions,” explains David Colarusso, the Director of the Legal Innovation & Technology (LIT) Lab at Suffolk Law School. “If you can frame a problem as a prediction problem and you have the right data, it can give you repeatable and scalable solutions. For access to justice, this might mean getting resources to those in need more quickly or freeing up professionals’ time by minimizing the time spent on a mechanical task, allowing them to do more human work.” He added that this would allow us to expand the reach and capacity of existing solutions while making practical new approaches to mitigate the access to justice crisis.
Colarusso created Learned Hands, an online game that “aims to crowdsource the labeling of laypeople’s legal questions for the training of machine learning classifiers and issue spotters.” In Learned Hands — created in partnership between the Suffolk LIT Lab and Stanford Law School’s Legal Design Lab — lawyers and law students review real-world legal questions and identify the legal issues they see, producing a labeled set of texts. A classifier is trained to recognize common patterns in this data and use them to identify similar legal issues in new texts.
So, does all this mean that the future of access to justice look like robots analyzing your issue, routing you to the right chatbot to help, and then following up as the case progresses?
The practice of law is so context- and judgment-dependent that it may be doubtful that the immediate future looks so automated, but there is no doubt that these tools, when applied correctly, will make meaningful strides in the way clients actually access justice.