I have always considered myself to be fairly accepting of interesting new technology, perhaps even to the point of early adoption. Although the financial commitment of law school has significantly reduced my disposable income, I still enjoy reading about the “next big thing” and following developments in information technology (“IT”) and general computing. In the context of the law, one of the most interesting recent developments is the creation of ROSS Intelligence, an online research tool for legal practitioners that utilizes IBM’s Watson interface.1 ROSS is designed to help lawyers complete mundane legal research and other associated tasks efficiently, saving their clients’ money and promoting greater access to legal services in the process. Hiring a lawyer can be incredibly expensive,2 so it is difficult to imagine a convincing argument against a tool like ROSS from an access to justice perspective. At the same time, I feel it is prudent to exercise some caution with regard to resources like ROSS, however effective and helpful they may be.3 This sort of sentiment on my part extends to other well-established services like CanLII,4 although it should be noted that CanLII serves a different purpose than ROSS.5
The main issue is that widespread adoption of online legal research tools, which will only become more advanced and accessible in the future, has the potential to reduce the value of flesh-and-blood lawyers and law students like myself.6 For someone who has (so far) spent over two years, hundreds of hours, and many thousands of dollars on an education that should support a healthy living, this is a bit of a frightening proposition, and is not unlike the anxiety other professionals may feel as automated processes threaten to remove more and more jobs from our economy.7 ROSS is designed for legal professionals, so they should be able to control who has access to the service, but who can say that ROSS will not be successful enough to warrant the eventual creation of a public access version sometime in the future? While I applaud and agree with ROSS co-creator Andrew Arruda’s sentiments vis-à-vis the utility and importance of ROSS as a concept,8 I do not think Arruda would deny that ROSS Intelligence is also a corporate entity that exists, however indirectly, to secure monetary profit.9
Of course, any apprehension I feel toward ROSS is largely dependent on an assumption that artificial intelligence will indeed progress far enough to accurately replicate the thought processes and considerations that go into effective legal research. Not having access to ROSS myself, I am not in a position to discuss its sophistication in detail. The only thing I can say with certainty is that strong legal research requires an ability to isolate the root of a problem as well as the flexibility to change approaches if necessary.10 I question whether artificial intelligence will ever be able to reproduce the technique of an outstanding legal researcher. Individuals approach research problems with the benefit of years and decades of lived experience. I do not subscribe to the notion that any research, whether legal or not, is conducted in a vacuum. With the benefit of previous research, people can understand and adapt to new developments in ways that artificial intelligence simply cannot, at least not yet. People can recognize issues with research questions and other foundational aspects of projects and correct them. To my understanding, current artificial intelligence can only answer questions that have been posed to it; it has no way to evaluate the utility of those questions themselves. To avoid sounding unfair, I want to clarify that I am not trying to paint ROSS as either an inept tool or some sort of harbinger of the end of legal practice as we know it. My aim is merely to suggest that it may be a good idea for the legal profession to ensure that it retains some degree of control over the use of services and tools like ROSS. Perhaps ROSS is best suited to simpler, “black and white” legal issues and tasks, and that is where its focus will remain. Apart from issues surrounding their own efficiency, lawyers have a professional responsibility to ensure the public receives the best legal services possible.
The ways legal services are delivered are certainly changing.11 ROSS is only one example of a general trend toward greater digitization of legal service delivery. In addition, many lawyers in North America are eschewing the traditional idea of a brick-and-mortar law office in favour of more flexible, open-ended office structures.12 In my view, attempting to halt or reverse these trends is a project that is doomed to fail. Today’s legal clients want services delivered quickly and efficiently.13 This is partly a product of the high price of legal services, but it is also attributable to the information age we live in. The internet, for better or for worse, gives many people the confidence to say they understand cases and legal texts about as well as their lawyers do. Perhaps because I am a law student, I feel compelled to refute such an idea as a matter of principle. My future clients will not have the hundreds of hours of training that I do when it comes to interpreting jurisprudence and legislation. They may not have the benefit of a broader education in the humanities, which goes a long way in developing effective argumentation skills and critical thinking. Self-representation is on the rise in Canadian courts, especially in the civil context,14 but I think people would do well to interpret the cost of legal services in light of the experience and expertise that legal representation offers. The work lawyers do has value, and it is the responsibility of the legal profession to protect that value.
This brings me back to ROSS and lawyers’ use of artificial intelligence in their day-to-day work. To summarize, ROSS is an incredible tool and will likely go a long way in promoting access to justice by increasing the efficiency of routine legal research. At the same time, the ways future services like ROSS are packaged and promoted should not suggest that an artificially intelligent “lawyer” is in any way a worthy replacement for a real human being. Open and equal access to information is important and will likely do more good than harm in the long run, but technology simply cannot equate with human experience. As legal professionals, it is up to us to ensure that our profession remains ours in the face of the constant march of computer science.
* BA Hons (University of Saskatchewan), JD Candidate (University of Saskatchewan). Any errors are mine.
1 See online: ROSS Intelligence, <https://rossintelligence.com>, archived: <https://perma.cc/8UP4-MAXP>; Drew Hasselback, “Meet ‘ROSS,’ the Bankruptcy Robo-Lawyer Employed by Some of the World’s Largest Law Firms”, (9 August 2016), Financial Post, online: <http://business.financialpost.com/executive/smart-shift/meet-ross-the-bankruptcy-robo-lawyer-employed-by-some-of-the-worlds-largest-law-firms>, archived: <https://perma.cc/UM3H-SUFR>. ROSS was co-developed by an alumnus of the University of Saskatchewan College of Law, Andrew Arruda. Arruda has spoken at length about the impetus behind ROSS: “to democratize the law” and improve access to justice for those who cannot afford traditional legal services (TED Institute, “The World’s First AI Legal Assistant | Andrew Arruda | TED Institute” (21 December 2016), online: YouTube <https://www.youtube.com/watch?v=wwbr0fombFs>, archived: <https://perma.cc/6QRH-BD4X> at 00h:04m:50s).
2 See Mallory Hendry, “Priced for Value”, Canadian Lawyer 41:6 (June 2017) 32, online: <http://www.canadianlawyermag.com/staticcontent/AttachedDocs/CL_June_17_PricedValue.pdf>, archived <https://perma.cc/V6MK-8G9Z>.
3 In this, I suppose that I fit the stereotype of my eventual profession. Lawyers are typically averse to change and often “sneer at the idea that their work could ever be done by a website or app” (Michael Skapinker, “Technology: Breaking the Law”, Financial Times (11 April 2016), online: <https://www.ft.com/content/c3a9347e-fdb4-11e5-b5f5-070dca6d0a0d?mhq5j=e3>, archived: <https://perma.cc/8AXH-6UFQ>. See also Pierre Noreau, “The Old…and the New? Elements for a General Theory of Institutional Change: The Case of Paperless Justice,” translated by Mary Baker and Emily Grant in Karim Benyekhlef et al, eds, eAccess to Justice (Ottawa: University of Ottawa Press, 2016) 263 (“law is one of the only spheres that still resists the integration of new technologies even though these technologies have completely changed the landscape in a wide range of public service sectors” at 263). Noreau suggests the changes most likely to take hold are those that digitally replicate existing physical ways of doing things, instead of reimagining them completely (ibid at 291).
4 Online: The Canadian Legal Information Institute, <https://www.canlii.org>.
5 As a student, CanLII is extremely useful as a repository of case law and legislation. It is particularly attractive because of its intuitive interface (although working at the Law Review has given me a greater appreciation of the physical library at the College of Law). However, CanLII primarily exists to provide access to legal resources to the general public, while ROSS is a tool designed for lawyers. If CanLII did not exist, individuals looking for jurisprudence, whether for personal interest or to help in comprehending their own legal issues, would have to rely on other online databases that are less robust. In Saskatchewan, the Law Society hosts recent judgments from the province’s courts online (see “Public Resources”, Law Society of Saskatchewan Library, online: <https://www.lawsociety.sk.ca/library/public-resources.aspx>), but the documents are linked from CanLII. The Supreme Court of Canada publishes its decisions online (see online: Judgments of the Supreme Court of Canada, <https://scc-csc.lexum.com/scc-csc/en/nav.do>), but there is no convenient way to track case history or search for citing authorities.
6 A strong affirmation of the value of lawyers in engaging with the Canadian legal system is found in s. 10(b) of the Canadian Charter of Rights and Freedoms (Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11), which affirms the right to legal counsel “without delay” following detainment or imprisonment by authorities (ibid). According to the Supreme Court of Canada, the purpose of s. 10(b) is to promote “adjudicative fairness” in legal proceedings (Clarkson v The Queen,  1 SCR 383 at 394, 26 DLR (4th) 493). Commentators have warned of the difficulties of representing oneself in court, largely related to the comparative “powerless[ness]” of self-represented litigants (see Diane Grant, “Representing Yourself in Court is Popular but Costly and Risky”, CBC News (31 December 2015), online: <http://www.cbc.ca/news/canada/representing-self-court-lawyers-1.3375609>, archived: <https://perma.cc/7VDN-LJWL>). In the criminal context, the common image is of the entire power of the state arrayed against a single individual. The dangers of self-representation may not be present to the same degree in the civil context, but it is difficult to suggest that lawyers are not extremely helpful in any case, given their training and experience.
7 See e.g. The Canadian Press, “Ottawa Warned about Job Losses that Could Stem from Automation”, CBC News (18 March 2017), online: <http://www.cbc.ca/news/business/job-automation-federal-government-1.4031206>, archived: <https://perma.cc/AF4X-7E76>. While it is intuitive that “[t]he jobs at the most risk are those that require repetitive activities like an automotive assembly line,” there is some concern that “high-skilled workers, such as financial advisers, are already being replaced by software programs” (ibid). See also Andy Blatchford, “Bank of Canada Warns Automation Will Lead to Job Losses”, The Globe and Mail (18 April 2017), online: <https://www.theglobeandmail.com/report-on-business/economy/bank-of-canada-warns-automation-will-lead-to-job-losses/article34733216>, archived: <https://perma.cc/N3QS-Q5ZQ>.
8 TED Institute, supra note 1.
9 On the ROSS Intelligence website (supra note 1), the existence of a “[s]ales” department is clearly indicated. The pay model for ROSS is likely subscription-based.
10 This sort of ability falls within the scope of chapter 3.1-2 of the Law Society of Saskatchewan’s Code of Professional Conduct (Regina: Law Society of Saskatchewan, 2016), which applies to all practising lawyers in Saskatchewan. The rule states that “[a] lawyer must perform all legal services undertaken on a client’s behalf to the standard of a competent lawyer” (ibid, ch 3.1-2). One of the main purposes of the rule is to minimize expense to the client, just as ROSS aims to do. Commentary 5 of chapter 3.1-2 says that “[a] lawyer should not undertake a matter without honestly feeling competent to handle it, or being able to become competent without undue delay, risk or expense to the client” (ibid).
11 See generally Canadian Bar Association, Futures: Transforming the Delivery of Legal Services in Canada, CBA Legal Futures Initiative (Ottawa: Canadian Bar Association, 2014), online: <http://www.cba.org/CBAMediaLibrary/cba_na/PDFs/CBA%20Legal%20Futures%20PDFS/Futures-Final-eng.pdf>, archived: <https://perma.cc/SV26-9FVR>. The CBA suggests the current climate of change in the legal profession represents an “outstanding opportunity” for innovators (ibid at 7), but cautions that “[i]t is critical for the Canadian legal profession to catch up with the forces that are radically changing the delivery of legal services and, by extension, the role and participation of lawyers in those activities, without further delay” (ibid at 4). The CBA, as a national body representing thousands of lawyers, “must…transform itself into a constant and active advocate for change” to address the changing nature of being a lawyer in Canada (ibid at 5). See also Karim Benyekhlef, “Introduction” in Benyekhlef et al, supra note 3 at 1, 9ff.
12 This evolution, if it continues, may also help to reduce the cost of legal services. See e.g. Mamie Joeveer, “How Two Stay-at-Home Moms are Changing the Legal Industry”, Forbes (27 August 2014), online: <https://www.forbes.com/sites/mamiejoeveer/2014/08/27/how-2-stay-at-home-moms-are-changing-the-legal-industry/#d0a717e1d345>, archived: <https://perma.cc/5548-H4BU>; Leigh McMullan Abramson, “Do Lawyers Need Offices Anymore?”, The Atlantic (9 October 2015), online: <https://www.theatlantic.com/business/archive/2015/10/do-lawyers-need-offices-anymore/409417>, archived: <https://perma.cc/8ACY-SLMH>. In Canada, the Law Society of Upper Canada (Ontario) allows lawyers to create “home offices” as long as they are able to provide legal services in a professional manner, in line with the expectations of the profession (The Law Society of Upper Canada, “Home Office”, online: <https://www.lsuc.on.ca/HomeOffice>, archived: <https://perma.cc/XFS6-3B6L>).
13 According to the CBA, in recent years “there has been a shift in market power to the client side” of the lawyer-client relationship (supra note 11 at 4). Different clients “have varying demands and expectations” (ibid), but it appears the high cost of legal services (and the corresponding inability of many to afford representation) has prompted a general increase in expectations vis-à-vis the quality of provided services. Today, clients “want services to be quicker, cheaper, and smarter” (ibid at 6). One might (perhaps rightly) regard such a set of characteristics with suspicion; a higher quality of service seems to intuitively require a greater length of time. However, since the legal profession is quite competitive, there is concern that clients will simply go elsewhere if a lawyer or firm does not meet their increasing expectations. See also Canadian Bar Association, Contributing Perspective: The Clients’ Perspective, CBA Legal Futures Initiative (Ottawa: Canadian Bar Association, 2013), online: <http://www.cba.org/CBAMediaLibrary/cba_na/PDFs/CBA%20Legal%20Futures%20PDFS/The-Clients-Perspective-Linked-eng.pdf>, archived: <https://perma.cc/SEH5-Q8PK>.
14 The increase in self-representation is primarily due to high fees for lawyers and the sluggish pace of the court system in general. See Grant, supra note 6; Charlie Gillis, “Why People Representing Themselves in Court are Clogging the Justice System”, Maclean’s (4 February 2013), online: <http://www.macleans.ca/news/canada/courting-a-crisis>, archived: <https://perma.cc/Q7C4-4LEE>. Today, the Saskatchewan Court of Appeal (“What if I Don’t Have a Lawyer?”, Courts of Saskatchewan, online: <https://sasklawcourts.ca/online-help-guide/introduction>, archived: <https://perma.cc/F28Z-N7ED>) and the Supreme Court of Canada (“Resources for Self-Represented Litigants” (9 August 2017), Supreme Court of Canada, online: <https://scc-csc.ca/res/unrep-nonrep/index-eng.aspx>, archived: <https://perma.cc/YBT8-PAKU>) publish resources online to help self-represented litigants engage effectively with the courts.