For some time now there has been a battle in the legal tech press, at conferences and on social media, over the question, Will robots replace lawyers? As the battle has progressed, the opinions have coalesced into two diametrically opposed sides.
The first believes that lawyers are destined for an artificial intelligence–empowered nirvana. According to these pundits, intelligent machines will arise soon to handle all of the drudgeries of the practice of law, leaving the better, creative parts of the law to be enjoyed by lawyers. The general tone of most law department leaders toward technology is hopeful. In the LexisNexis 2018 Insights report, titled Legal Technology: Looking Past the Hype, 57 percent of the general counsel surveyed believes that technology investments have already increased productivity. Moreover, 60 percent of general counsel believes that technology will help improve the accuracy of legal work in the next three to five years.
The other side of the argument paints the stark picture of a jobless robot hellscape. As the landmark law journal article “Lola v. Skadden and the Automation of the Legal Profession,” in the Yale Journal of Law and Technology, states: “Technological innovation has accelerated at an exponential pace in the last few decades, ushering in an era of unprecedented advancements in algorithms and artificial intelligence technologies . ... [T]o survive the rise of technology in the legal field, lawyers will need to adapt to a new ‘practice of law.’”
The reality is that both sides in this debate are wrong—if only because both sides are right.
The Robots Are Already Here
So let’s look into the three general categories of robotic process automation (RPA).
First, there are rules-based processes that mimic human actions (such as predefined workflows) that automate highly repetitive tasks like accounts payable and event-driven processes like client intake.
Second, there is cognitive automation, which interprets meaning in context using machine-learning technologies and natural language algorithms to classify and extract meaning from semistructured and unstructured content.
Finally, there is artificial intelligence (AI), which is really a new frontier that models human intelligence to augment the legal decision-making processes such as dispute resolution. As was pointed out in an ABA Journal article, “How Artificial Intelligence Is Transforming the Legal Profession”: “Artificial intelligence is changing the way lawyers think, the way they do business and the way they interact with clients. Artificial intelligence is more than legal technology. It is the next great hope that will revolutionize the legal profession.”
How RPA Is Your Friend
So how could RPA work in practice for a small firm or solo lawyer?
Let’s start with something basic: the client intake form. Every law firm needs to do this, and it seems simple, right? But, even then, it is a tedious and error-prone process.
Your first step should be digitization of all incoming client-related documents and forms. By using OCR technology, you can transform documents into business value by capturing and validating information in any format as soon as such information enters your firm by turning that information into a machine-understandable format. Fortunately, there are RPA systems that can do just that, by building upon the workflows that you already use to extract the information from your forms and enter it into your operational systems. Or, in the alternative, these RPA systems could build your forms for you as extensions of those systems, to automatically enter that data into them.
Once documents are transformed into machine-readable text, they can also be converted into searchable PDFs. Information from these documents may then be extracted, such as client-related metadata, which may be populated into your matter management system and you may then apply the business rules associated with your work product. Converting your PDFs into searchable documents ensures that you can quickly find documents in your content repositories.
Moreover, with searchable PDFs you may redact portions of documents that may be sensitive in nature or that may contain personally identifiable information before sharing them internally or externally. In many cases sensitive data is visible in plain sight within the document, but sometimes such data can also be “hidden” within metadata, attached files and comments. Instead of going through all these areas manually and redacting or removing sensitive information piece by piece, with the appropriate software tools these can be removed quickly and easily. This way you can “sanitize” PDF documents by removing hidden data with just a few clicks.
Furthermore, searchable PDFs are extremely helpful during litigation, when documents are collected and reviewed by attorneys before presenting them to opposing counsel during the discovery process. The ability to perform text searches to identify potentially responsive documents is critical. If not done thoroughly, opposing counsel can challenge the process used to determine what documents were provided to them, which could result in possible delays, sanctions or potentially losing the case.
But why stop there? What if you could use RPA systems that go through all of your old client intake forms, extract not just the information but the meaning from the words therein? You could create a database of all of your client matters, to populate, confirm or correct the fields of all of your operational systems. You could power market to those clients. You could analyze trends, such as learning where your clients are most likely to come from, to make that marketing more effective. You could potentially determine which clients and cases are likely to be the most lucrative—and which ones are most likely to stiff you on your bills—all before you agree to represent them.
Assessing the financial risks to law firms of taking on new cases is a highly labor-intensive and time-consuming task. Research by McKinsey Global Institute found that 56 percent of a firm’s client-facing activities, such as conflicts management and client credit ratings, may be fully automated. “Instead of bookkeepers and legal aides digging around for sources, the data can be automatically found, gathered and placed into the appropriate risk registers by robots. Partners or their financial controllers will have more time to assess the risk at the client and transactional level. It also makes the law firm more self-reliant, saving costs and effort in contacting consultants.”
Other Law Firm Uses of RPA
A recent Deloitte survey found that 91 percent of the finance executives surveyed consider accounts payable automation as a top priority. Even though the accounts payable automation technology is at a mature stage, only 23 percent of the companies surveyed implemented straight-through touchless processing of vendor invoices. And, according to Ardent Partners, invoice-processing transaction costs may be as high as $13.47 per invoice and can take as long as 11.4 days to process.
Moreover, incoming invoices tend to be nonstandard, arriving in multiple formats and channels, such as paper, fax, email with PDF attachments and electronic data exchange. Automating invoice receipt through the application of intelligent capture technology, which automates the extraction of invoice data and posts to the enterprise resource planning system, is proven to reduce transaction costs by as much as 32 percent.
What about billing? Like accounts payable, it also tends to be a labor-intensive process that can be easily automated. Using RPA, a “bot” would automatically run without any human intervention to prepare and generate bills from your matter management system, resulting in as much as a 50-percent reduction in labor costs, improved cash flow and visibility of working capital.
Compliance checks is another area where RPA can deliver significant efficiency improvements as well as help your firm mitigate legal risks. For example, specific bots may be set up to search external databases and data sources, extract relevant information and generate a suspicious activity report that may alert your firm to potential risks.
How much of your time is wasted in routine client communications? RPA can also be applied to automate highly repetitive correspondence functions with your client. For example, by setting up a client portal, documents may be more efficiently exchanged; upon receipt or sending of documents, an automatic confirmation is sent that may include a review of the documents and, in the event of missing documents, alert the sender to forward the required documentation. Received and/or exchanged documents may then be automatically profiled into your matter management systems with the requisite metadata added for traceability and for chain of custody.
Applying AI to Contracts
Finally, there is a new frontier for improving the nature of legal work: the use of AI to automate what is universally recognized as highly tedious and repetitive contract analytics. We should probably start with the most critical pain points lawyers face with mounting pressures to manage an increasingly heavy case load in managing commercial agreements, merger and acquisitions transactions and compliance obligations.
With the proliferation of digital content, the first step of any contract analytics project is simply finding all of the contracts located on local drives, shared drives, SharePoint sites and even email servers (because so many just get left as attachments to emails). After finding all of the contracts—which, even for some of the best-organized organizations, can take days—the next step is to figure out which ones are actually active and how the mass of agreements, addenda, appendices, purchase orders and more all fit together.
The next step is to start working through the contract language. This should be easy when that language comes from your own standardized contracts, assuming of course that those contracts are standardized. Unfortunately, they so often aren’t and instead have inconsistent naming, inconsistent or no metadata tagging, not to mention a lack of harmonized clause libraries or playbooks. Of course, making sense of standard terms doesn’t work when you don’t have standard terms, which is, in a nutshell, the dilemma of third-party paper—and why, by the way, the experts say that value leakage on a third-party contract can be from 5 to 15 percent of the contract’s value.
Contract analytics enables you to apply advanced AI technology to automate the identification, extraction and analysis of contract terms, surface business critical information such as contract entities and evaluate them against standard clause libraries. You can then perform clause comparison and contract reviews based on algorithms to automate in a way that is implicitly understood by humans. Once you can get to this point, you can start driving efficiencies and creating business value. As stated in a February 2018 Harvard Business Review article titled “How AI Is Changing Contracts”: “AI software, however, can easily extract data and clarify the content of contracts ... It can let companies review contracts more rapidly, organize and locate large amounts of contract data more easily, decrease the potential for contract disputes ... and increase the volume of contracts it is able to negotiate and execute.”
To sum up, RPA is not science fiction; it’s available in the here and now. This new kind of power won’t just eliminate hours of drudgery, it would make your practice more successful. And isn’t that the point?
Let’s Hire the Robots Now
Lawyers and the legal profession stand at a crossroads. The technology is now available to make our professional lives so much better and productive. That same technology can also replace much of what lawyers do. For some lawyers, technology will present that first, fantastic option. For all too many, technology will instead leave them with the short end of the stick.
It’s up to you how you react and whether you hire the robots to work for you. As the ABA Journal article put it so poignantly: “One thing is certain—there will be winners and losers among lawyers who do and do not uptake AI, respectively. As one senior lawyer recently remarked, ‘Unless private practice lawyers start to engage with new technology, they are not going to be relevant even to their clients. The AI train is leaving the station—it is time to jump on board.’”