Lexoo speaks to Richard Tromans, founder of ArtificialLawyer.com, to find out what New Wave legal technology is, how in-house lawyers can get internal buy-in for legal tech adoption, why London is the epicentre of the legal AI revolution, his predictions for the future, and the best advice he’s ever been given.
Thanks for speaking with us, Richard. Can you tell us a little bit about your background and why you set up Artificial Lawyer?
About 10 years ago, I started working at a large US consultancy as a strategy consultant to law firms. I then worked for a smaller boutique consultancy in the City of London. Two years ago, I set up my own business, Tromans Consulting, and I've been advising law firms on strategy and business decisions since.
It became quite clear to me about 18 months ago that AI was going to have a profound impact as it evolves and spreads through the market, and I needed to take this into account when advising my clients. I decided to launch a professional blog a year ago, called Artificial Lawyer, as nobody was writing on the topic consistently and in any level of detail. The blog became for me a way of capturing all the relevant information in this area, in a way that was beneficial to myself and my clients.
I wasn’t expecting it, but the blog has now turned into a global website. There are news articles, readers from around the world, sponsored thought leadership, advertising, and it's now taking up a lot more of my time! I cover it day in, day out, follow all the little developments and bring it all together, which hasn’t really been done before.
Your website is focused on "New Wave legal technology". What do you mean by that?
For me, New Wave legal tech is technology that performs work. Up until very recently, a lot of legal tech has effectively been what you might call "librarian technology" - collecting data, labelling it and finding ways of sharing it.
Now AI has come along offering a way to actually do the work of a junior lawyer or paralegal. Rules-based systems like Neota Logic can crystallise the knowledge of a group of lawyers in a particular practice area. Once set up, a person can ask the system questions and it will provide pretty solid, reliable answers. You may then want check this with a real lawyer but it's the first baby step into a true industrialisation of cognition.
Law is effectively a cognitive business. It's about understanding, processing and then feeding back that knowledge to clients. A lot of what lawyers do is simply asking: "What are in these legal documents? What do they mean? How should I approach this problem? What do I need to do?", and so forth. That doesn’t really call on great legal minds to engage. So the first wave of legal tech affects several areas. For example, you can now use AI to better understand court cases and predict the outcomes of similar cases; you can also use natural language processing (NLP) to analyse the narratives in legal bills to determine whether it fits with the billing guidelines of a client. The uses just keep expanding.
We're about two years into broader commercial adoption of AI in the legal sector and significant uptake has only really occurred in the last 12 months. AI systems, automated systems and expert systems will increasingly do more work, but I don't personally believe that this is going to take away the jobs of very experienced lawyers.
We at Lexoo speak to a lot of in-house lawyers, and in the last 12 months, many are now seriously looking into adopting legal tech. How do you think in-house lawyers can make use of legal tech? If they’re hesitating or facing internal resistance, for example, on the basis that the technology is not ready yet, what do you say to that?
If in-house lawyers are getting internal pushback on the basis that legal tech is not mature yet, that doesn't make any sense to me. It's like saying the internet doesn't have enough websites so let's not use it.
Take document review - which is probably the most prevalent use of what I would describe as AI technology - the technology is now very mature and reliable. Say you work for a large company that’s particularly concerned about change of control clauses in your contracts. Once the document review system has learned the limited number of variations to the clause, it will be able to pick them out without a problem.
In-house legal teams can use technology to build internal services. You could use a process automation system like Neota Logic to create very valuable internal resources. There are also contract automation tools like synergist.io and Juro that can be used across the company, for example, by allowing a sales or procurement team to create legal documents with clients, which then gets checked by the in-house team.
I also think in-house teams will need to think through their own internal resourcing. Do you adopt legal process outsourcing, use an alternative legal service provider, engage a service like Lexoo or work with certain law firms? The answer probably is that you need a mix of all of them.
I can’t express it strongly enough; in-house lawyers should just start using legal tech.
What legal tech innovation do you think will be most widely adopted in the next five to 10 years?
AI document review. There used to be a big technological barrier in developing the NLP so that systems can read legal clauses, but we're on the other side of that barrier now and the systems have become a lot easier to use. Kira, for example, is a quick study facility that makes it very easy to train up a system using just a small number of documents. It's quite intuitive to use and it'll probably become more intuitive as the years go by.
Lawyers will realise it's not rocket science, and clients will see that this technology is available and reconsider how much they pay to the actual lawyers to do the work. It is absolutely inevitable that it will happen, and I think every major law firm will be using AI for their document review within the next five years.
The key task for legal tech is to not be seen as legal tech, but just a part of your desktop in the same way that email is. Fundamentally, the real change is behavioural. The lawyers just have to stop doing things manually and inefficiently. The driver for inefficiency is lawyers being paid a lot of money to be this way, but until recently, there simply was no alternative. Now there is another option.
What legal tech innovation do you want to be more widely adopted and want more people to get excited about?
I would like to see expert systems become more complex. Expert systems are essentially very complex logic trees that use probabilistic inference to guide you through a series of questions and answers to lead to a conclusion.
Let's say, for example, you're using a Q&A expert system about employment law and you ask: "I want to fire Bob. Bob's 66. He's not at the age of retirement but we really want him to leave. Can we do this?" There are a lot of issues involved here, but if you’re not asking for an absolute answer and just want some key points, the system can already do this for you.
At the next level, the system will need to learn to engage more, start to absorb documents, use NLP to read them, feed that information back into the system, and produce actual advice. A lawyer can then do a final check of the recommendation. I don't think we're that far off this.
There’s a tiny gap between producing useful information and giving advice. Closing this will, I think, require a combination of the technology that is used in document review and expert systems, as well as some process automation. I think that’s where we’re going.
You travel around the world giving talks on legal tech and you have a deep understanding of the industry. Which city is at the forefront of this?
London. Maybe I'm saying this because I live here, but London feels like it's the epicentre of the legal AI revolution.
It's interesting because the UK has always had a unique position in the legal industry, despite it being a relatively small market. There's this incredible concentration of talent, with big law firms, large clients and developed legal AI companies all operating here, and I think that fosters a lot of activity. Some of the biggest legal AI companies in North America have a lot of clients in the UK.
There is also some really good stuff coming out of Toronto, California and New York. You can go to Silicon Valley and see a lot of legal tech companies like Lex Machina, Ravel, etc, but California is not really the legal centre of America, so you don't get quite the same connectivity.
And finally, what is the best advice you've ever been given?
Probably from a family member a long time ago: youth is wasted on the young, and so you should enjoy yourself when you're young because it’ll be a lot harder to when you’re older.
It’s very difficult now because after university, you're expected to go straight into some kind of training or apprenticeship, particularly if you want to be a lawyer or another professional. There seems to be so little time to explore the world.
Law firms tend to want to hire people who fit a cookie cutter mould, and I’m not sure that’s the best strategy. As the basic process work that a lot of junior lawyers do is sucked up into the world of AI and automation, the work that's left over will be more and more dependent on human skills: empathy, being able to think broadly, understanding humans and how the broader economy works, and not thinking that everything is based in Tunbridge Wells.
My dream would be that law firms start recruiting people say four or five years after university, after they've gone and explored the world a bit but, being realistic, I know they're not going to do that. It would be nice if that did happen.