A.I. Helps Take Automated Contract Analysis to the Next Level
By: Erçin Aslan | Law Technology Today
Digitally augmenting the work of lawyers and in-house counsels could be the next step in the vast field of contract review. In complex fast-paced economies, diverging interests collide and are balanced out by the law and the tools it provides for that purpose, such as the use of contracts.
Accordingly, contracts can become as thick as books. An entire industry of attorneys, counsels, paralegals, and consultants are busy trying to sort out the world, often accused of a dark art with complex terms and an impenetrable language of its own.
Courts are notoriously overwhelmed and expensive to the point where relying on a good contract can make the difference in getting along and ahead of the crowd. Hence the importance of being earnest and efficient when it comes down to drafting and reviewing your contracts.
Yet still, contract review processes have been petrified by the concept of billable hours—the traditional operating model of external counsel—and by the lack of innovation in in-house counseling. In short, the legal industry has left the door wide open for technology disruption.
Expect the Unexpected
Early players that walked in that door in the past have been providers of product and lifecycle management solutions that attempted to get a better grip on contracts. Soon thereafter, solutions for compiling customized contracts popped up that used standardized building blocks that the end user selects while navigating decision tree schemes. Other solutions facilitate the process on its fringes as shown by the wide use of electronic signature tools.
Now, new players are swooping in and tapping into the potential that advancements in computer science have made in artificial intelligence. Their cutting-edge solutions offer prediction technology, legal analytics, document automation, contract and IP portfolio management, online dispute resolution, e-discovery, online dispute resolution, and litigation funding.
The providers often turn to their clients in bringing alive the tools and algorithms they offer. These tools accelerate current processes and free up resources for focus and new terrain. It turns out that the clients simply know best how to repurpose generic tools for their proper ends. Their transversal expert teams propel the technology’s potential to levels of exploitation which even sector-specific tech providers hardly ever achieve on their own.
This holds especially true for contracts, which—no matter their level of standardization—are as diverse and stretched in scope as their users are. Consequently, customizing automation tools becomes an intrinsic act of successful deployment of a viable solution for expediting contract drafting and review.
Transform Legal Data into Actionable Information
The goal of contract review tools is to transform legal data and content into relevant and contextualized information. In other words, squeezing ambiguous groups of words into the pristine certainty of the binary code: 0 and 1.
But such goals are often wrongheaded. Contracts are part of the law which in turn is a discipline of the social sciences that are based on anything but categories of black and white. It’s all greyish. For the foreseeable future, automation in the field of contract review will need humans for what we are good at: Associating ideas, thinking out of the box, and giving sense where there is ambiguity.
However, and on the same note, reviewing contracts isn’t a dark art either. Most contracts, especially in business, drift to repeatedly use terms and a language that stand for ideas and concepts fixed by the law or judicial precedents.
If you add to this a familiar context—let’s say a specific business sector—you will have the proper grounds for automating parts of the work on a grand scale by:
- Flagging documents as relevant to a particular case (e-discovery)
- Recognizing and tagging clauses and legal concepts (contextualized search)
- Detecting anomalies in documents by unusual language or undue omissions or additions (analysis)
- Attributing a file to a contract type, to a region, a currency, a product (categorization)
All of the above can help to reduce the amount of time reviewing from top to bottom massive amounts of standard contracts.
Enable the Information-Driven Legal Organization
How is this possible? Let’s put it very simply: “It’s the search, stupid.”
The better your search tool is equipped to recognize relevant information, the better the outcome of the automated process will be. For that purpose, the automation tool must be able to understand context.
Unless your contract uses a specific or outlandish vernacular, it’s possible to define markers that are commonly used in contract language and script often mockingly referred to as “legalese” for expressing recurrent legal concepts. On top of that, and once your search tool is fit to find these concepts, you’ll need to tell if the findings correlate with what is the agreed standard, be it a preset standard or customized standard.
This is basically what the legal industry is mostly doing by hand. Instead, you can go further in helping your stakeholders by equipping the automation tools with the ability to compare the findings with the mass of similar files. If in doing that the tool were able to continuously and automatically recalibrate, it wouldn’t be much of an effort to have it also inferring recommendations or taking actions on its own, such as pushing forward a detected anomaly to a human reviewer.
At this point, you will have found a tool that combines basic search with machine learning algorithms at its core. From there on it’ll be up to you to unleash your creativity and facilitate the lives of lawyers, legal departments and all other stakeholders who’d benefit from reanimating deadlocked business models and impermeable structures.
I am expecting the unexpected.