Andrew Arruda (pictured) co-founded IBM Watson-backed legal research provider ROSS Intelligence in 2014 with fellow University of Toronto student Jimoh Ovbiagele, since when the company, which relocated to Palo Alto in 2015, has signed eight major US law firms, the most recent of which was Bryan Cave.
With ‘real’ AI still something of a rarity in the legal market, Legal IT Insider caught up with Arruda to find out how ROSS is developing and whether its use cases measure up to initial expectations.
What practice areas is ROSS being used in now and how is that expected to develop over the next, say, year to 18 months?
ROSS is live in US bankruptcy and intellectual property law. We are also exploring other legal AIs with our partners such as a ROSS KM AI. As for the next year to 18 months, we will be expanding ROSS’s breadth and depth meaning additional AI capabilities in areas ROSS is already in and widening the scope of ROSS’ capabilities into many more practice areas.
When you first launched ROSS, what were your expectations as to how the software would be applied by law firms and has the reality mirrored that or not?
We first saw ROSS as technology that would be quickly adopted by solo-practitioners and lawyers at small law firms as our AI has the power to level the playing field for those with less resources. While we certainly had Big Law and in-house teams in mind, we thought adoption would weigh heavily with solos and small law firms at the beginning.
In reality, ROSS was quickly adopted by large and medium sized law firms as well as in-house teams in addition to small firms and solos. We learned more about the extreme shift in the legal-market post-2008 and how the pressure from clients to be efficient is not something that is constrained to any one slice of the legal industry.
What are the key challenges you have faced using ROSS in practice and how have you overcome them?
Adoption of any technology within an organization can be a challenge as adopting new technology means breaking old established habits. This being said, one of the challenges we faced early was being the “new kid on the block” and having to establish ourselves in lawyer’s workflows. As time went on and our profile continued to raise we began to see lawyers at firms we are working with jump at the opportunity to use ROSS, which has been great.
Within the firms already using the technology, who is it that gets the most benefit from it and uses it most frequently?
ROSS is a frontline tool for practitioners. Associates do most of the legal research tasks so I’d say associates gain the most out of using ROSS and use it most frequently, not just in that it allows them to work more effectively, but it also allows them to reach their professional potential sooner. By enabling associates to focus their energy on high-value work, they are able to learn much more about the practice of law rather than spend their time searching for information. ROSS provides them with the answers they need, rather than search results.
The thing is though, ROSS has a few AIs which do different powerful things so while associates benefit the most from our core search AI capabilities, partners seem to benefit the most from ROSS’ law monitor AI, which keeps them up to date on any changes to the law for all the cases they are working on.
What do law firms struggle with most when it comes to using this kind of sophisticated AI tool?
The beauty of AI is that, while in the background the technology is incredibly sophisticated, AI allows for extreme simplicity when it comes to the user experience. Since ROSS understands how humans communicate, filtering by date/jurisdiction etc. is done away with, as lawyers simply ask the question they are looking to answer in its entirety – the way they would ask a lawyer at their office with experience in the matter.
What this translates to is very sophisticated technology being tied behind a simple user experience that is extremely intuitive.
What firms do struggle with is wrapping their minds around what AI can do at present and where it is going. This is why, in addition to providing our AI software, ROSS Intelligence also works with law firm leadership to establish the firm’s AI strategy and keep the firm leadership up to date with advances in AI as they occur, so they can always stay on the cuttingedge.
Do lawyers feel threatened and does that inhibit the way they interact with the software?
Lawyers only feel threatened before they use the technology as the idea of artificial intelligence sometimes strikes the imagination and ignites thoughts from sci-fi movies from the 1980’s and 1990’s.
Once lawyers use ROSS they end up seeing just how much more they can accomplish which for the team at ROSS Intelligence, there’s nothing really better than that!
I always say there are three stages when it comes to introducing AI technology: 1) the fear stage, where folks make decisions based on very little data and past (usually fictitious) information, then 2) the information stage: this is where folks actually read up on what artificial intelligence is and what it can do today and where the technology is going tomorrow and lastly 3) full-adoption: this is when the organization is educated on AI, and makes the decision to bring AI into the organization.
Some people, firms and organizations go through these stages faster than others, and it’s the early adopters who stand to gain the most as they can reap the benefits before the technology becomes status quo and everyone is using it.
AI tools will almost certainly take out lower level legal work. What jobs do you envisage may be created that don’t exist now as the adoption of AI technology becomes commonplace?
I do not believe that AI tools will “almost certainly take out lower level legal work” because the legal market at present is not addressing everyone who needs a lawyer. Time and time again studies show that the majority of people who need a lawyer cannot afford one.
This means there is a massive untapped market of folks who 1) have insufficient funds and 2) want a lawyer to help them.
I think that AI technology will allow for this untapped market to be addressed by savvy lawyers who use AI technology, in combination with nonAI technology, to create a practice which is efficient making serving this untapped market, what Richard Susskind has labelled the “latent legal market” possible.
Do you envisage a time when ROSS is used by non-lawyers or will it always augment legal practice?
We build all of our AIs with human lawyers at the center of them, and this is a conscious decision because the right legal training is essential to being able to navigate the law.
In my opinion, the issue at present isn’t that we do not have enough lawyers, as we see report after report of the overabundance of lawyers in the market who cannot find work, who also happen to be in record numbers of debt, but that the tools lawyers have available to them essentially price them out of being able to address most of the people who need a lawyer.
What does 2017 look like for you and what can we expect from ROSS?
2017 looks like another record year at ROSS Intelligence. We can expect to see ROSS supercharging lawyers’ abilities at even more organizations, being used at more law schools and also driving access to justice through a variety of new programs we are launching which allow deserving organizations to have access to ROSS technology free of charge. 2017 here we go!