CCBJ: With legal teams facing increasing pressure to do more with less, how does artificial intelligence (AI) – specifically generative AI – help alleviate those challenges and enhance efficiency?
Matt Lung: With generative AI, the phrase I like to use is “force multiplier.” It can help you get started and generate momentum, and it has parallel workflows that allow the lawyers to work in tandem with the generative AI tools, which helps them be more efficient. Automated document drafting is an example. You can have the AI draft the initial versions of your templates, so that you have a head start and can spend less time on manual drafting. It can also look at the previous language you’ve used to ensure that you’re being consistent across all your templates.
Let’s say I have about 200 cases that I’m handling simultaneously. How can I possibly draft anything that summarizes what’s going on in all of them without spending two weeks working on it? Well, you get a head start with generative AI. It can take that massive set of data and turn it into a concise summary, highlighting the key points. Then you’re not spending all that time trying to pull the data together.
Or let’s say you have a meeting with your internal stakeholders. You’re not an expert on everything, but you need to have an informed basis for the conversation, and you can get that with legal research tools and AI tools. It’s doesn’t just tell you that a case is relevant. It drafts an initial memo for you, almost like a first-year lawyer would.
All these things can help an in-house lawyer who’s facing challenges from a variety of perspectives, with everybody wanting a piece of their time. Leveraging generative AI to give you that first draft, to give you that summarization, to give you that legal research – it really increases efficiency.
Talk to us about which legal tasks benefit the most from generative AI, and how does technology transform the speed and effectiveness of legal operations overall?
You want to start with what generative AI is really, really strong at, which is summarization and analysis and comparisons of large amounts of text, but also drafting, categorizing and brainstorming. These are all things that generative AI is great at because it can take a massive set of information and consolidate it into a smaller piece of information. The lawyer benefits from the AI’s speed, effectively multiplying themselves.
Document review is an obvious one. LexisNexis, for example, has a tool called Create+ that will review the database of your existing contracts and tell you, “This is your standard template, this is your standard language, this is what’s been agreed upon.” It can process an incoming contract and say, “This is inconsistent with your internal position. Here’s the clause that you always use.” So, as a lawyer, particularly if you’re a new lawyer, rather than coming in with a blank slate and trying to understand, “What is our corporate position?” you’re immediately leveraging all that knowledge from every lawyer that’s been there before you, every deal that you’ve ever done.
Legal research, which I mentioned earlier, is another important one. You can’t possibly know everything about every legal issue that you’re going to face. When you are asked a question by your client, you usually have very little time to give them an answer. But now you have the ability to get up to speed rapidly, and not just by having the research done, but even an initial memo, in a format you like, to get you started in terms of understanding the issue or having an initial conversation with your client, or as you get prepared to have an informed conversation with you outside counsel or an outside expert.
These are the things that in the past would’ve taken a lot of time, or maybe you would have had an outside counsel draft the memo for you and you’d get it back a couple of days later. Now you have generative AI behind you, giving you that push, that head start. You can react much quicker than you would have been able to in the past. If you have a litigation-heavy practice, you can summarize and chronologize the documents as they come in. You can compare deposition transcripts to what the evidence shows. You can get a handle on what your legal exposure to some issue really is, not based on your gut feeling, but based on an analysis of what the company has done for the past 20 years in cases like this. And you can be specific: What is it like when you’re in this jurisdiction or in front of this judge, with this plaintiff’s lawyer, with this defense lawyer?
Being able to pull that information together, from such a variety of sources, and then consolidate it into something understandable in a short time frame will help you be more efficient in determining your litigation strategy and expenditure.
When you revise your contracts with generative AI, you may find that now you’re signing 60 or 70 percent of your contracts unchanged.
Excellent. So, how does AI enhance the drafting process, and what measurable improvements does it bring to legal department workflows?
Perhaps the biggest thing it does to enhance the drafting process is help you be less lawyerly, if that makes sense. Many lawyers repeat the same clauses as they write, but if you look at a contract 20 years later, it’s kind of this cut-and-paste mishmash of the past 20 years of legalese. Sometimes it doesn’t really make a whole lot of sense, because it was written at different times, and definitions have started to shift a bit and it doesn’t flow very well.
Because generative AI does a good job of reviewing and revising and suggesting changes, both for compliance with changes in the law but also in terms of style, it’s a wonderful place to start with the drafting process – whether it’s “make this simpler, make this shorter, address this issue that I’ve been hearing about from my customers for the past six months” or whatever. Let’s just say a clause is not well-written for some reason. You can have it make a suggestion that fixes the issue, and as you fix the issues in your contracts, as you make them clearer and get fewer redlines back, you’re being more efficient in your contracting. The more contracts you can send out and have signed without having to negotiate, without having to get on the phone to work through some issue, the better off you are. That’s one way it can enhance the drafting process – by looking not just at your positions but also at what feedback you’re getting from your counterparties when you’re negotiating.
Another thing AI can do to enhance the drafting process is leverage your current knowledge base across those various silos that we’ve all entered into in one way or another, particularly as we’ve become more geographically spread out in this new world of remote or hybrid working. How do I know what my lawyer in Los Angeles has done to resolve this clause? How do I know what my lawyer in the United Kingdom has done to resolve this clause? How do I know what someone who was here five years ago but has since left did to resolve this issue? The amount of volume is unimaginable. But with AI, you can investigate those contracts. You can reference them as a grounding source of truth, knowing that this is what we do, these are our templates. This is a solution that Janice in Los Angeles had just last week. I haven’t had a chance to talk to her, but I can still benefit from that language because the AI tool can look at the contracts she’s done. It can pull that clause out and suggest it to me as a solution to my problem. Those are the kinds of measurable improvements that you’re going to see more and more of down the road.
Another measurable improvement is just in terms of contract turnaround time. As you look at your contracts and see that a particular clause is a constant point of negotiation or presents some type of issue that your counterparties are consistently raising, you can ask generative AI to revise that clause. Once it knows that the questions or concerns that have been raised about the clause revolve around X – this issue or that – it can revise the clause to address the concerns in compliance with your policies. It can solve those issues.
Once you do that, the next time you send that contract out, that clause doesn’t get redlined anymore. So, maybe in the past it took four days to be signed, and now it takes two or three days. As you leverage that across all your contract templates, across your company, it adds up. Every 60 minutes that a lawyer isn’t looking at the contract on the other side or redlining it and sending it back – that adds up across the entire legal team and makes the turnaround time for your contracts much, much quicker.
It all adds up. Even five minutes a day per lawyer across a legal team of 20 people, over a period of a year, adds up. Then also, consider those deals that didn’t get closed because you ran out of time because your contract wasn’t negotiable: When you revise your contracts with generative AI, even if you’re just explaining things better and not necessarily changing the risk positions at all, you may find that all of a sudden you’re signing 60 or 70 percent of your contracts unchanged, as opposed to just 30 or 40 percent. That’s a big difference because now you’re processing more and more contracts. Your revenues go up. It is a win-win for everybody.
That’s great. You mentioned that AI doesn’t necessarily need to even address risk positions to be valuable, but it can help with that too. So, what are some effective ways generative AI can help identify and mitigate legal risks within a corporate setting?
Many companies have vast databases of executed contracts that they have simply been renewing for years without any visibility into what they truly say. In the past, that felt like a hole that was too big get out of if you ever did jump into it. So, companies just internalized that risk and moved forward. But with generative AI, instead of having to hire a team of lawyers to go through every contract line by line, you can use generative AI on it. You can have it review and summarize this massive set of your contracts, with a set of ground rules – like look for contracts that have this provision or are in this jurisdiction or have unlimited liability, whatever it is you’re trying to isolate or account for.
You can now get that first cut done by generative AI, which can flag the outliers for a lawyer to review, and now your legal team is looking at hundreds of contracts, instead of tens of thousands of contracts. Now you’re in a position to really ask, “What is the issue with this contract? What can we do to renegotiate it or amend it, or do we need to terminate it?” You’re mitigating corporate risk that had always existed but you just hadn’t stepped on it yet. But now you have a chance to avoid that risk or close that compliance gap, because now you know it’s there and you were actually able to find it.
That’s something that you would not have had the time to do in the past, because you would have had to sit there and go through everything, page by page. But generative AI does a much better job of that, and obviously much quicker. Then you add a human lawyer to the loop, but they’re looking at a much smaller data set.
Let’s keep in mind that, yes, identifying and mitigating these kinds of existing contractual risks is important in a corporate setting – but the risk of not closing deals to begin with is also real, and I think it’s common sometimes in the in-house world for people to forget that. This gets back to making sure that your contracts are ones that your counterparties will sign, and formatting your contracts in the best possible way, using language that helps explain why a clause is the way it is.
You can also ask generative AI to think about bigger picture risks. Give it your persona. Who are you? Where do you work? What does your business do? Say, for instance, that you are a senior corporate counsel for a legal technology company that sells products in the United States and Canada. Give that info to the AI and ask it what business risks it thinks you might be facing in the next 10 or 15 years.
Say you’re operating in six different jurisdictions. The AI can help you identify potential risks that you should do more research on. Literally, you can ask the generative AI to brainstorm with you, and it will go back and forth with you and generate ideas about things you can chase down, things you can focus on, things you can direct your ops team or your compliance team to consider. It can help you look at the big picture and see, “Do we have gaps here?”
Every little generative AI push helps free up your time to be a strategic business partner.
How can AI empower in-house counsel to become stronger strategic partners in the business, driving alignment, efficiency and value?
That’s a simple one. It allows in-house counsel to spend fewer hours on time-consuming yet necessary work such as summarizing all of your open matters in the state of California or reviewing and amending all of your existing contracts in a jurisdiction that just changed its laws. Every head start you get, every little generative AI push helps free up your time to be that strategic business partner, that value-added partner, where you can really leverage those creative thinking and problem-solving skills that in-house counsel bring to the business team.
AI can help you position to be viewed as part of the business team, rather than just a legal silo. That’s the key to becoming that stronger strategic partner — having the time to align with the business, to work with them and understand what they’re doing. You can become that partner when you have the time to spend on it and you’re not stuck in routine reviews of thousands of contracts.
Last question. What shifts will AI bring to the relationship between in-house counsel and outside law firms?
I know that some people think that generative AI is going to be the end of lawyering as we know it – or maybe the end of the need for outside counsel. I don’t believe either of those things are true. First off, I believe outside counsel will actually benefit from the use of generative AI. In terms of the work outside law firms do for their clients, generative AI is going to move their lawyers up the value chain, which helps them become more strategically important to their corporate clients. Also, internally, AI is going to drive more complicated work toward outside counsel. It’s going to free up legal spend for the complicated matters, because it’s not being used up on the routine and mundane things that are lower risk, which means the businesses can now get a better handle on things.
Businesses will get a better understanding their existing risk profile because they’ll finally be able to truly see the risks that are there – and get a head start on addressing them. Understanding the historical risks that they already carry will allow them to better evaluate legal issues and more accurately scope benefits and risks and determine the need to engage outside counsel. And they’ll have the bandwidth to do it. So, I think the generative AI shift will actually benefit both the in-house counsel and their outside law firms by moving everybody up that strategic value chain.