What Does the Future Hold for the Law Librarian in the Advent of Artificial Intelligence? / Que réserve l’avenir pour le bibliothécaire de droit avec la venue de l’intelligence artificielle?

Abstract

Technology is transforming the work of information professionals as the methods for retrieving information continue to evolve. Artificial intelligence (AI) is being incorporated into many legal practices for the purpose of research, e-discovery, analysis, and documentation. While the legal profession is in the early stages of incorporating AI technology, it is uncertain whether an AI-enhanced application will be able to conduct legal research intelligently, eliminating the need for lawyers and law librarians to engage in research activities themselves.

Résumé

La technologie transforme le travail des professionnels de l’information, les méthodes de repérage d’informations évoluant continuellement. L’intelligence artificielle est de plus en plus intégrée dans de nombreuses pratiques juridiques pour la recherche, l’e-discovery, l’analyse et la documentation. Alors que la profession juridique est dans les premiers stades d’intégration de l’intelligence artificielle, une incertitude demeure quant à la capacité des applications utilisant l’IA à exécuter une recherche juridique intelligemment, éliminant la nécessité pour les avocats et les bibliothécaires de droit de se livrer eux-mêmes à des activités de recherche.

Introduction

Artificial intelligence (AI) has become a topic of heated debate for many professions. The technology’s benefits are substantial as it has the ability to reach reasoned conclusions, which can outpace the human mind’s ability, at a significantly cheaper cost and with increased speed, accuracy, and consistency (Chang 2016). For decades, law librarians have been on the front lines of training law firm associates in the art of legal research. From reference services to formal training programs, firm librarians have been fundamental to bridging the gap between law school courses and the realities of research in practice. The introduction of AI software in many law firms worldwide questions the continued relevance of the role of the law librarian. AI will impact legal research, practice, and librarianship; however, it will not replace the expertise of the law firm librarian who is skilled not only at research but also at working alongside the changes to information structure and systems.

Recently, a group of University of Toronto students developed Ross Intelligence, an “artificially intelligent attorney” that is built on top of IBM Watson, a cognitive computer that competed and won the game of Jeopardy in 2011. Ross has already been introduced in several large firms worldwide and seems to produce better results than traditional legal research platforms that require Boolean searches for best results, despite a natural language option. It has the capability of answering legal questions by “reading the entire body of law and returning a cited answer and topical readings from legislation, case law, and secondary sources” (Goodman 2016).

Like many AI systems, Ross Intelligence is marketed to appeal to those who find research to be “mundane” and would prefer to spend more time with clients while AI takes on the more “mechanical” aspects. This reasoning suggests that Ross will not replace lawyers but, rather, only enhance their work since lawyers will be able to spend more time on “the legal tasks that computers are lousy at and that humans can perform well—communicating with clients, counselling them, watching out for their interests, [and] advocating for them in the court-room” (Kroh 2016). Although this rationale may safeguard lawyers, it fails to consider the future of information professionals. To comprehend what the impact of AI could mean for both the future of legal research and information professionals, this analysis will first provide a brief introduction to AI. It will then evaluate how AI has been adopted in legal research, practice, and librarianship thus far from both the perspective of lawyers and librarians. Next it will examine the possible consequences of implementing AI and will survey the qualities and skills that might allow law librarians to continue working in firms despite the evolving force of technology.

General introduction to AI

Before diving into the discussion, it is necessary to identify the concepts that frequently describe AI in the literature. Often AI is described as cognitive computing when it is in fact a subcomponent. To clarify, cognitive computing refers to a set of comprehensive capabilities based on technologies that not only include AI but also go far beyond it. By definition, cognitive computing contains “the fields of machine learning, reasoning and decision making technologies, language, speech and vision recognition and processing technologies, human interface technologies, and high performance computing” (Rossi 2016, 2). It is designed to foster new discoveries, solve a range of practical problems, and boost productivity.

According to the creators of IBM Watson, in an artificially intelligent system, the system can tell a user what course of action to take based on its analysis; whereas, in cognitive computing, a system provides information to help the user decide what course of action to take. For example, the capabilities of cognitive computing are highly beneficial when assisting doctors in identifying the correct diagnosis for a patient. AI systems such as IBM Watson use a patients’ individual health data—including genetic information—with the wealth of material available in public databases, textbooks, and journals to help come up with more personalized treatments. Doctors can then use the recommendations to decide on how they will pursue their patient’s treatment (Hernandez 2014).

While the field of AI is extremely broad when it comes to providing a concrete definition, many scholars describe AI as “everything from relatively simple computer programs and conversational agents to sophisticated robots” (Talley 2016, 387). AI attempts to replace human intelligence with something synthetic. However, IBM suggests that AI is best explained as “augmented intelligence,” meaning that these systems are built to enhance human expertise and skills rather than replace them. Many companies have attempted to replicate patterns of human thought, yet most of these systems seek only to provide outputs or perform tasks drawing on human intelligence (Thompson 2015, 11). Essentially, AI systems can be used as practical tools as opposed to autonomous systems that engage in legal decision making on advanced or philosophical levels (Schafer 2007, 386).

In addition to ROSS Intelligence, there are many AI-based products with legal applications. While the products are not yet widely adopted, they have generated a lot of attention in the legal industry. Some of the companies that currently specialize in AI-based products for the legal industry include Ravel Law, Neota Logic, RAVN systems, Lex Machina, and NextLaw Labs (Gediman 2016, 36). The majority of these products are able to use AI technology to prepare standard contracts, assist with trial preparation, and identify internal experts. It is unclear whether these applications qualify as being truly intelligent; however, by using algorithms, they are able to mimic certain human interpretative and judgment-related processes, which are then applied to the queries and data. Users can enter natural language questions into the application, which will then search the database for the answer that it determines to be a close match to the question that was asked (37). It is the workhorse behind intelligent technology—the “intelligent agent”—that helps to predict the correct or best results based on the variables that have been received and any prior knowledge (Talley 2016, 387).

On a similar note, machine learning is an intrinsic aspect of AI that enables algorithms to improve through self-learning from data without any human intervention. Specifically, it is an advanced learning paradigm that “learns continuously, accumulates the knowledge learned in previous tasks, and uses it to help future learning” (Chen and Liu 2016, 1). The more data that the system is exposed to, the more it learns, and the more accurate it becomes over time. Ultimately, the “neural network” becomes a complex set of decisions that the computer can make to arrive at an answer (Gediman 2016, 38).

The adoption of AI in legal practice

With the possibility of embracing AI into legal practice, this next section will explore both the perspectives of lawyers and librarians to understand how AI has, and could be, adopted into legal research and practice.

A lawyer’s perspective

Richard Susskind is one of the legal industry’s leading prognosticators, whose works have sparked much debate within the profession. In an era where machines can out-perform human beings in many tasks, he questions the prospects for employment and considers the tasks that should be reserved exclusively for people. It is through his work that the industry has started to question whether the adoption of AI in law firms will signal what Susskind calls “the end of lawyers.” He suggests that lawyers, among other professionals, face a future in which “increasingly capable machines, autonomously or with non-specialist users, will take on many of the tasks that are currently the realm of the professions” (Susskind and Susskind 2015, 231). However, as evolutionary beings, we are constantly searching for ways to upscale our abilities, and AI is becoming a large aspect of this modern reality.

The advantages of implementing AI are clear, with the main argument being that they “could give the lawyer more time to devote to complex work that requires judgment and expertise and will expand further into the heart of the clients’ business” (Tearle 2008, 10). Other supportive arguments have been that the software can be used for due diligence by sifting through large amounts of data, short-list relevant facts, and enhance and augment human reasoning. It can also navigate the complex web of regulatory law in seconds (Bundro 2016). As mentioned earlier, platforms such as ROSS Intelligence have a natural language-processing capability that allows practitioners to ask questions in plain language. For example, asking a question like: “can an automatic stay be lifted if a plaintiff in another case requests it?” would retrieve a response that has assessed legal precedents and has suggested readings (Cutler 2015; Taal et al. 2016, 361). It also attaches more information about the cases paired with the citation and contains confidence ratings (Taal et al. 2016, 361).

Interestingly, many argue that these expert systems also have the potential to help with access to justice, as the high cost of legal services is out of reach for many (Brescia et al. 2014, 554). AI can perform tasks such as providing information about legal issues, explanations about basic legal concepts, referral to legal forms, and provide advice about when to hire a lawyer (570). However, there is the conundrum when one considers whether or not these services are an “unauthorized practice of law” since they have the ability to provide legal advice without consulting the assistance of a lawyer (579). For example, in the United States, legal battles for and against the intelligent software LegalZoom have been well documented in regard to whether this service constitutes the unauthorized practice of law. Through the use of automated technology, LegalZoom helps users create legal documents for personal and business purposes. The “consumer services relate to wills, divorces, prenuptial agreements, personal bankruptcy, immigration, disability benefits, personal injury and real estate documents” as well as other small business related activities (Solomon and Moyse 2016, 1). While the site may market itself as being a service that helps people create legal documents, it also advertises itself as being a “personalized, affordable legal protection service” that has the capability of providing assistance on what information should be on those forms (Brescia et al. 2014, 574).

In the case of Unauthorized Practice of Law Committee v. Parsons Technology Inc., the court found that a computer program that tailored legal forms to individuals’ particular needs was considered an unauthorized practice of law.¹ The program created “an air of reliability . . . which increase[d] the likelihood that an individual user w[ould] be misled into relying on them.”¹ As such, there is the potential for intelligence software to go far beyond clerical services by providing personalized guidance that could be held to constitute the practice of law. AI software should only provide patrons with possible resources that they could consult for further information as well as an explanation on how to access these resources (Talley 2016, 401).

Much of the literature jumps to the conclusion that computers are changing the way law is practised; however, they do not prove the imminent and wide-spread displacement of lawyers by computers alone. Such a rash conclusion foregoes the opportunity for a more nuanced and careful analysis. AI is not capable of the type of creative and independent thought that lawyers have gained through years of practice, nor is the software capable of unaided bias. Further, it cannot uphold the rule of law or dispense justice and is unable to weigh the significance of evidence (Bundro 2016). Ultimately, to suggest that human lawyers will become obsolete due to automation is, arguably, premature.

A librarian’s perspective

Librarians thrive on information, and, yet, they can also “choke on it,” so to speak. Information overload is a problem that affects law librarians every workday. It occurs when an individual receives so much data, making them unable to engage in higher levels of processing (Hensiak 2003, 85). Today, law librarians are taking on more tasks and responsibilities then ever before, which is likely to result in information overload. For example, they are answering extensive reference questions, creating and maintaining webpages and intranets for patron use, using push technology for current awareness, and are engaging in knowledge management initiatives (91). During the average workday, they are also faced with many interruptions such as e-mail notifications, patron requests, last minute meetings, and phone calls (92). Further, they are expected to be experts in finding news and business information as well as statutory and administrative law. It is no wonder why law librarians are feeling the constant pressure to multi-task in an effort to meet the increased expectations of patrons. This, in turn, leads to the influx of additional information in quantities too large to digest (91).

A solution to information overload could be to incorporate AI software. This could increase the amount of expertise and information that can be brought to bear on the problem without increasing the burden of overload on other individuals (Hensiak 2003, 95). For example, automation could help answer basic reference questions, leaving more time for librarians to focus on complex reference questions that require more of their expertise. If an intelligent reference assistant was unable to adequately answer the question, the system could refer the individual directly to a librarian for further assistance (Talley 2016, 395). By having the database keep the conversation logs between the AI assistant and the patron, the logs would allow librarians to identify questions that AI assistants failed to answer correctly. Not only would this allow a librarian to follow up with the patron, necessary modifications could be made to the software to better assist patrons in the future (Balleste 2007, 53). AI assistants could also help to extend the hours of operation of the library so that additional service is available 24 hours a day, seven days a week (54). In terms of library materials, AI technology could be used to check in and out materials, assist with inter-library loan requests, and recommend other materials that might be of interest. Essentially, automating some of these functions would help free up time for librarians to work on more labour-intensive projects (Talley 2016, 399). Some scholars have suggested that, if AI software is capable of performing these “core librarian tasks,” one must question whether librarians will still be needed in organizations such as law firms. This question will be explored later on in the discussion.

Drawbacks of AI software

There are many drawbacks of using AI software, which should be considered before their implementation. Some of these issues include masking the inadequate research skills of law students and young professionals, the costs of AI, and the associated privacy concerns of machine learning.

Information literacy

A common refrain in the literature is that law students are graduating with a lack of advanced legal research skills (Margolis and Murray 2012, 121). This is surprising considering that legal research skills are considered a fundamental and core skill for law students and budding lawyers. Examining research resources and conducting quality analyses facilitates clarity of thinking and helps to achieve demonstrated competence. The LexisNexis (2015, 4) “White Paper: Hiring Partners Reveal New Attorney Readiness for Real World Practice” demonstrates that “young lawyers often lack the ability to research more complex legal issues in cases, statutes and regulations, [as well as] determining the strength of validity in primary law, and legislative/administrative content.” Further, employers, particularly those with more years in practice, expect new lawyers to be research experts; they have high expectations when it comes to their research skills as “they should be able to adequately and effectively find everything that’s up to the minute” (Wawrose 2013, 532).

In a recent survey conducted by the American Association of Law Libraries, law firm associates indicated that new lawyers spend more than 30% of their time doing legal research. As a result, approximately 50% of associates think legal research should be a larger part of the law school curriculum (Mart et al. 2013, 8). Today’s legal research curriculum tends to focus on writing and the use of computer-assisted research, where students continue to conduct superficial research and fail to consult standard sources and indexes (Osborne 2016, 404). Teaching writing and teaching research require different knowledge bases and approaches. Even though many legal writing programs incorporate law librarians into legal research instruction, this is not always the case (Bintliff 2009, 2). There is an immense amount of pressure on faculty to bring new law students’ writing skills up to a professional level, and as a response to this pressure, greater emphasis is placed on writing in their classes, causing research instruction to receive less attention (1). This points to several areas of concern with the research skills of recent graduates, including inefficiency, problems distinguishing among sources, preference for easy access sources, and just doing “enough research to get by” (Margolis and Murray 2012, 131).

Frequently, students are inclined to regard Google as their first, last, and best research solution. Unfortunately, search engines like Google tend to neglect one’s ability to go through the effort of defining one’s information need, as many have given up on clearly articulating the reason for the query or gap in information (Gallacher 2005, 27). By just typing a word or two into the search box, the search engine is expected to disambiguate the query and provide an answer. Often users will then scroll the first page of the results for the “right site” and do not clearly know what they are searching for until they see it (28). Nonetheless, the ability of law students to find information through Google is not the issue; rather, the issue is students’ inability to “dig deep, to think critically, to evaluate the information they are finding for fit, and to engage in the legal analysis required for legal practice” (Osborne 2016, 407). The potential therefore exists for students to transfer their inferior research knowledge and practice into the workplace.

Several scholars claim that AI could help to improve young lawyers research skills by engaging them more fully in the process of finding legal research materials, while others are not fully convinced that this is the case (Talley 2016, 396). The performance of legal research is a skill, and like any other skill, it is learned by doing and engaging in critical thinking as well as by problem solving. It is questionable whether this could be achieved by relying on machine intelligence to do the “heavy lifting.” While the usage of AI may be convenient, there are many downfalls of AI that should be considered, as this technology does not always guarantee a high-quality search. The software will not be able to determine if a question is too broad or too narrowly focused, nor will it be able to consider what else might have been available had the question been phrased differently (Badke 2015, 72). For instance, humans have the ability to make assumptions and draw conclusions about common knowledge; however, AI does not have the ability to deliver an answer that draws upon sound judgment if it is not outlined in the question. Similarly, a searcher will still have to examine the search results to determine their quality, relevancy, and importance. It will also be unable to explain the provenance of the information in terms of its background, purpose, the reasons the resource exists, and the biases that shape the information (Badke 2015, 72–73).

Arguably, these downfalls may not matter to those who value simplicity over sophistication, as the common mentality is that this computer software is “good enough at research” (Brooks 2009, 295). After all, in many cases some people tend to choose the simple, inferior solution that they can achieve at a cheaper and faster pace, over a better solution that requires more time and effort. Examining this issue from a librarian perspective highlights that information literacy skills (that is, the ability to recognize when information is needed and to find, evaluate, and use the information effectively) will continue to be of the utmost importance for law students as future legal workers (American Library Association 2000). Essentially, students need to understand the information cycle, how to express clear research problems, and know how to evaluate information (Badke 2015, 3). For example, effective legal research means being able to organize and make sense of the information that has been located. By applying evaluative criteria to judge the credibility and relevance of the retrieved information, one should be able to ask oneself questions such as:

• • is the information current;

• • are there any gaps in the information;

• • how well do the search results answer the original query;

• • do the search results raise new queries all together; and

• • how can this information be applied effectively to resolve a specific issue or need.

Asking these questions provides a knowledge base for users to understand what they are looking at, whether the information adequately answers their questions, and how they can effectively use their research findings in their work (Hutchinson 2014, 591).

It is evident that the literature is full of negative commentary about the skills and abilities of the millennial generation, including the concern that they have not learned “with sufficient rigor the skills necessary for complex and in-depth research” (Margolis and Murray 2012, 131). Since the new generation of law students are digital natives, incorporating information literacy into legal research instruction will go a long way. Finding information is no longer an issue, as there is no “right” process that all researchers should follow; rather, students need less instruction on how to find the law and more instruction on assessing and evaluating the sources that they find (156). Not only are librarians an asset to the development of this knowledge within the legal research curriculum, they are also vital to its continued practice in law firms. After all, employers recognize that research is not just about finding results, they want associates to think strategically about putting together the best combination of sources for the task by considering: “what’s the problem I’m being asked, what’s the resolution being required, what are my tools to get there?” (Wawrose 2013, 533). As such, librarians are a fundamental resource to consult to help find answers to these questions. Thus, regardless of the constant change to how research is being conducted, one aspect of legal research will never change—that is, the ability to carefully read, interpret, and analyse the information that has been found to provide a solid answer to the original research question (553).

Costs

There has been a lot of discussion about the cost savings that AI can bring to legal practice, but there has been little conversation about the resources needed to implement the technology. Connie Brenton, senior director of legal operations at NetApp, a data management company, cautioned against overlooking these challenges at “Legalweek: The Experience 2017 Conference.” She noted that while many are attracted to the novelty of AI, the technology requires a significant amount of resources to get up and running, which should not be overlooked (quoted in Dipshan 2017). In explaining the financial reality of AI, other experts have suggested that lawyers have to account for the costs behind licensing and purchasing. According to Jennifer McCarron, a technology program manger at Cisco, implementing a virtual legal assistant such as “Riverview Law’s Kim,” is estimated to start off at a fixed rate of $30,000 for 10 users. However, additional upgrades add to this base price. For example, to add a single application such as “templates for auto-generated documents” could cost between $22,000 and $37,000 per enterprise. Furthermore, firms also need to factor in additional costs for set-up as well as the training and maintenance of the software. Given that AI technology has to actually “learn” the information, there will need to be personnel devoted to helping the machine accumulate the specific legal knowledge needed to perform its work (Dipshan 2017).

While Riverview Law’s Kim is basic AI software, it is difficult to generalize these costs to include all types of legal AI software, especially those that are more sophisticated. Nevertheless, when researching the types of companies that are investing in their development, it is reasonable to believe that the cost will be great. In an era of tight budgets, where research database subscriptions are being slashed and every dollar is scrutinized, using this technology may be cost prohibitive, especially in smaller firms (Talley 2016, 399).

Privacy concerns

Lastly, AI software may be useful, but there is no guarantee any information collected from previous users will be kept private. AI systems feed on data and learn from one task and apply it to another. As result, by collecting massive amounts of data from different sources, it makes it difficult to properly have informed consent and control for big data uses, when there is no guarantee how the information will be used and what it will be combined with. As such, these decision-making machines are an “all seeing, all remembering in-house guest” that can manipulate the trust of the user to make the service more targeted and efficient. Some scholars have raised the question of what could happen to confidential client information that falls into the hands of these systems (Dickson 2017). For example, when personal information is combined with external data sets, this could draw inferences about new facts about clients—facts that are meant to stay private and that should not be disclosed to others (Mehmood et al. 2016, 1822). This is a concern because clients may not be informed that their information is being stored and processed by the software. In addition, they might also be unaware that their information could be circulated to other parties for further use. It is unclear whether clients can be offered a way to opt out of this form of data sharing (1825).

If firms are going to be using AI technology, necessary steps should be taken to regulate information collection, sharing practices, and the disclosure of user data. While many view the upside of AI as being so transformative, there should not be a moral obligation to feed AI unlimited data because of the amazing things it can do. Users should not have to “entertain the death of privacy” to gain the benefits of AI. Furthermore, it is evident that there are several drawbacks that should be considered before the implementation of AI software. First, information literacy skills should not be neglected, as evaluating information will continue to be a fundamental skill for legal practice. Likewise, AI technology is still young and the market is evolving, while AI can modernize the legal industry, it is likely to be tempered by the reality of getting up and running, which will come at a cost (Dipshan 2017). Lastly, AI and machine learning feed on data. The bigger these data sets are, the smarter the intelligence software will be since every query is retained by the engine for analysis to uncover patterns in the mountains of information that have been received. As a result, AI poses privacy challenges since a users’ personal information can be represented in the data sets and recycled for further use.

The future of law librarianship

In addition to the generic skills of managing information and services, the role of the law librarian has always been, and always will be, complex. For instance,

law librarians need an understanding of how the law is made and amended. They need to know the process whereby law is enacted. They need to know how courts make law. They need to know the jurisdiction of the courts. They must know the appeal process. They must know how to read a judgment. They must understand legal citation. They must know how to determine the current state of the law on any issue and they must know how to locate what the law was on that issue at any relevant time. They should know the most authoritative sources for information on all areas of the law. In addition they must be aware of how to access this information from both print and digital sources and have an awareness of the strengths and limitations of particular sources. They must have the ability to teach library clients how to use the materials in the library.

(Johanson and Brown 2009, 18–19)

While the primary role of the law librarian will continue to revolve around finding specific information for their patrons, this role may take less time with increasingly faster technology. As a result, this shift could mean that librarians will be busier than ever before because they are taking on other tasks to increase their value to the organizations that employ them. The expanding roles of today’s legal information professionals tend to focus on knowledge management, transformative client support, and business development work. Some specific examples of law librarians’ evolving roles include process improvement, knowledge management, and competitive intelligence. Of course, the role of law librarians is not limited to these activities.

Process improvement

Legal Lean Sigma is becoming increasingly popular in law firms as a method for improving a firm’s performance in delivering legal service. It combines techniques from both Six Sigma and Lean Processing by focusing on improving quality and efficiency by eliminating waste (Crosby 2017, 43). Specifically, Six Sigma focuses on the measurement and reduction of errors, and Lean Processing methodology focuses on creating efficiencies and value for the customer (43). One of Lean Six Sigma’s core frameworks is the data-driven improvement cycle DMAIC, which is an acronym that stands for define, measure, analyse, improve, and control (Tjaden 2010, 5). Ultimately, the goal of this philosophy is to identify and eliminate non-essential and non-value-added steps to streamline productivity, improve quality, and gain client loyalty (Mazzeo 2016).

The most fundamental step to applying Lean Six Sigma in the legal setting is to identify the waste that a firm encounters. Some examples of waste could include giving a client legal advice without performing sufficient research; par-taking in over-production by having too many lawyers on a single client matter; or engaging in extra-processing by creating multiple document drafts and over-researching issues (Mazzeo 2016). This identification is meant to uncover exactly what tasks are happening/occurring in an organization and finding a more effective way to accomplish the same thing. This often entails interviewing people, researching, and gathering data to capture a fuller picture of what is working and what is not (Crosby 2017, 44). To provide an example of how a firm could approach one of the issues above, a process map or flowchart could be a useful visual in identifying the steps of a process and guarding against any errors that could potentially occur. Process mapping is important to legal professionals as it provides an opportunity to see a workflow from start to finish as well as the connection to other players that could contribute to the process (that is, lawyers, paralegals, and other professionals) (Mazzeo 2016). Ultimately, this helps to label the various responsibilities of those involved and the many steps that must be taken to achieve the desired outcome without duplication of efforts and overuse of resources. Essentially, process improvement will help law firms to look at the bigger picture and understand that any improvement needs to be continuous. Whether it is in hiring, training, billing, or case handling, improvements add value to clients in a market where competition is at an all-time high. After all, if a client does not feel that they are being taken care of, there are many other qualified firms out there that they could consult to fulfil their needs (Mazzeo 2016).

So what role can information professionals play in making process improvements? Law librarians consider process improvement in their normal workflow already, leaving a huge opportunity for information professionals to lead the way (Crosby 2017, 46). For example, process improvement is used in the library field to maximize users’ satisfaction. Librarians tend to have a clear understanding of their patrons’ needs and demands and are constantly thinking about how they can be a better service to specific user groups (Al-Zubi and Basha 2010, 86). By using Lean Six Sigma principles and methodology, librarians can be the ones to help their firm move forward on these fronts. Convincing lawyers to examine their own work is a challenge as “it is very difficult for lawyers to see what they do as a process and to think in terms of standardizing that process” (Crosby 2017, 43). Information professionals could take the lead role in capturing and monitoring data on changing task flow by examining questions such as:

• • how can processes be standardized;

• • what could be done to reduce or eliminate repetition;

• • how can communication be improved within the team and/or with clients; and

• • could delegating tasks reduce costs?

It takes time to make improvements, but information professionals can help an organization to establish best practices for any given process to facilitate transformational change (Crosby 2013).

Interestingly, some law librarians are choosing to do formal training in Lean Six Sigma since an employee with this unique skillset is viewed as a competitive advantage. There are different levels of certification that can be acquired, such as white, yellow, green, and black belt; each with their own specific tools and experience. In addition to formal training, there are also a variety of online courses available through most community colleges.

Knowledge management

Knowledge management refers to the set of practices that facilitate the creation, capture, organization, and dissemination of knowledge (Fraser-Arnott 2014, 1). Similar to process improvement, the goal of knowledge management is to reduce the amount of time spent repeating work or locating information that should be easily accessible (Tjaden 2010, 2). There are many activities that law librarians engage in to further knowledge management and move beyond traditional library and research responsibilities. For instance, in an effort to capture explicit (that is, documented knowledge) and tacit knowledge (that is, experiential knowledge that is difficult to transcribe), librarians are using tools and methods such as content management, document management, intranets, and taxonomies to organize internal and external knowledge (Crosby 2012, 1–2). In some firms, librarians are involved in practice groups, participate in meetings, and organize knowledge artefacts such as minutes, notes, and audio recordings that come out of the meetings (7). Other knowledge management responsibilities that librarians take on include the creation and maintenance of blogs, wikis, and detailed directories of expertise; developing and indexing case briefs for future reference; as well as documentation of social networks, knowledge of intricacies of the law, client matters, letters, factums, agreements, comments, and, ultimately, any information that individuals are willing to share (2). Law librarians are transforming themselves into value-adding knowledge professionals since they are able to use their expertise to strengthen knowledge flow within their organizations. It is through the implementation of knowledge management that firms will be able to make better use of internally developed knowledge. This is accomplished by information professionals’ ability to create, record, and store information effectively (Lastres 2012, 3).

Competitive intelligence

Lastly, competitive intelligence also continues to grow as a major focus for law librarians given their extensive research skills. Competitive intelligence entails the process of gathering information about an organization, industry, or client to capture highlights and summaries of significant facts from news, industry reports, and specialty databases and resources (American Association of Law Libraries 2011, 5). The process not only enhances client relations but also assists with strategic planning and helps to increase productivity and profitability (5). Although many are able to find information through mainstream search engines, such as Google, librarians are skilled at being able to locate the right information in a timely and efficient manner. As a result, they are considered an invaluable asset in a firm’s goal to identify, attract, and retain clients (11). For instance, they can use their skills to investigate how clients view themselves, and they can learn about the current projects clients are involved in and the status of their work industries (Fisher and Bender 2006, 1). In addition, librarians can reconnect the firm with inactive clients and contribute to target marketing campaigns (2). Essentially, once a firm starts using competitive intelligence, there are many different ways it can be applied: finding clients, researching companies, developing existing clients, market research, industry research, and current awareness (American Association of Law Libraries 2011, 8). It is through this work that librarians are considered information gatekeepers since they are able to determine what content will get passed along and how (Cohen 2008, 2). Since partners and firm management do not have the time to sift through multiple pages of printouts, librarians can read the information, synthesize it to establish connections, and make the information more comprehensible (Cohen 2008, 1). Fundamentally, librarians help to “connect the dots” in these many instances, turning raw data into analysed information that can be helpful to both the firm and clients (American Association of Law Libraries 2011, 8).

Interestingly, there are several online tools and AI systems that can be used to map and synthesize competitive information. For example, these tools have the ability to screen thousands of pages on a specific topic by extracting keywords and concepts. The information can also be automatically summarized into an easy-to-read format based on common results that were frequent in the search. Despite these tools being able to extract information in less time, they are far from perfect. According to the 2015 “Marketing Automation Performance Report,” the top concerns for implementing competitive intelligence automation software included lack of budget, lack of in-house skills, lack of time, uncertainty of results, poor integration with other systems, and uncertainty that methods for data collection are sufficient (Marion 2015, 19). Ultimately, these factors have a huge impact on the adoption of automated competitive intelligence software since firms are not fully convinced that they are worth the time and investment (5). This suggests that automation cannot replace the quality of the highly detailed analyses conducted by research teams or information professionals, at least for the time being.

So what does this all mean?

While the rate of change in the legal industry seems to be accelerating, law librarians will continue to be an instrumental asset to law firms. Despite the trend toward self-service research through smart delivery systems, these systems still need to be customized and maintained according to a firm’s legal practice areas. In addition, legal professionals will need to be trained on how to use these smart research tools, and librarians will be best situated to conduct such training. Moreover, information professionals will continue to play a role in delivering service through advanced research platforms, on-demand research, providing alerts, and other services to support end-users (Lambert 2016). It is evident that the research skills of librarians are increasingly being used within other areas where their knowledge and familiarity with a wide range of resources is viewed as a competitive advantage. They are highly influential retrieval and analysis specialists that impact the vast knowledge base of a law firm, both internally and externally, online, and in print. As such, the risk of law librarians being replaced by intelligent research systems seems unlikely; librarians are adapting to, and are embracing, change by taking on advanced roles and will continue to do so as long as firms continue to have information and research needs.

Conclusion

The rise of AI software being incorporated into legal practice has elevated the fear of job cuts in the sector, as the technology is able to perform tasks that are currently performed by humans. Despite AI being in the early stages of implementation, it is here to stay. Since lawyers are embracing these technologies out of belief that they can make their practices more efficient, librarians should do the same. After all, law librarians have nothing to fear since their traditional roles are, and will continue, to evolve to keep up with technology. As such, the data-driven technology will assist human work rather than replace it. The work that consumes both lawyers and information professionals involves strategy, creativity, judgment, and empathy, and these are efforts that cannot be automated, at least, for the time being.