The US Legal Services Corporation is now ‘ready to announce’ its ‘Legal Navigator’ project ‘to the world’, according to a presentation to the Self Represented Litigants Network by senior LSC executive Glenn Rawdon. This rather underestimates previous coverage of the project. The LSC has, as Mr Rawdon also said, been working on this for some time with a range of collaborating organisations. It has been trailed at several earlier LSC technology conferences. Those interested have been waiting for word of progress. And now we have it.
The project has a number of interesting – and potentially replicable – components. One relates to the use of Artificial Intelligence. This is the coverage in an earlier post: ‘In the US, Microsoft has been a co-funder of a ‘Legal Navigator’ project designed to have an AI element in problem identification ad referral. And, linked to that there remains considerable interest in using Natural Language Processing and Machine Learning to help identifying and responding to legal questions. A project between Stanford University and Suffolk Law School has developed a game called Learned Hands to assemble some of the necessary data. Learned Hands is a game in which you spot possible legal issues in real people’s stories about their problems. You read the stories, and then say whether you see a certain legal issue — family law issues, consumer law issues, criminal law issues, etc. The game is also a research project. Each time you play, you are training a machine learning model to be able to spot people’s legal issues. This model will be used to develop access to justice technologies that connect people with public legal help resources.’
If you are interested in this aspect, then a video featuring David Colarusso of Suffolk University Law School Legal Innovation and Technology Lab is available on the net as the opening part of a session of an LSC 2020 conference. In the course of time, Learned Hands became Spot, ‘an issue spotter. Give Spot a non-lawyer’s description of a situation, and it returns a list of likely issues from’ a national list now in its second incarnation, also covered in a previous post.
The portal presents us with a good example of collaboration in the legal services field. We have universities like Suffolk. We have funders like the LSC and the Pew Charitable Trusts which have backed the project. We have Microsoft which promised $1m in services. We have probono.net and the National Centre for State Courts which has also been involved. This is considerable combined firepower. We also have two different jurisdictions – Hawaii and Alaska – that have been willing to act as pilots.
As I understand the idea, it is essentially to provide a wrapper that can be replicated by any state or organisation within which a user can have assistance in identifying their problem; be given guided assistance in resolving it; or referred on. The package allows original material but also the incorporation of material that exists elsewhere. The LSC is interested in it as a prototype for statewide services but you could think of using it – to take the example of West Hampstead Law Centre where I was once employed and which, alas, no longer exists – to provide help to the people living in the Kilburn area of London which links to a mix of national information (eg from Citizens Advice) and local sources.
A further refinement is that the site can lead into guided interviews (the Americans are good at these) which lead you through how to resolve a problem and allows the user to print off a document which will assist them.
The two pilot states now have a website which allows you to explore the Legal Navigator further. This still needs populating with more content but it helps you to get a sense of the idea. This is its promise: ‘Need a plan of action for your legal issue? Our Guided Assistant can help. Answer a few questions to help us understand your needs. You will be provided with an Action Plan that details the steps you can take to proceed with your legal issue.’ As a matter of policy, the site has strong privacy criteria. It does not require an account; it is free; it does not share information; it does not store location. Identify an issue and it will list available information from relevant sources. You can use Google translation facilities for most languages. I tried Czech. It looked OK: I could at least see that it got the right script.
To an outside observer, this potentially flagship project seems to be running a bit slow. But, that may be because it involves some complexity and has so many moving parts. Its value for legal service providers outside the US is in its example of what can be done to add value to basic legal information. And in the collaboration that lies behind it. We need the preparation work to continue a bit further before we will be able to see precisely how helpful users in Hawaii and Alaska find their sites. And the extent to which other states and organisations take up the model which has been designed with open source software to facilitate copying. The LSC wants to facilitate replication within the US. Mr Rawdon even produced a helpful price list from the firm involved in building the site. Thus, it would cost $26,750 ‘to get it up and running’. Maintenance cost are around $13,000. You have to produce the material. For foreigners, the value is more in the example of what can be done.