{"id":5134,"date":"2017-03-23T13:39:59","date_gmt":"2017-03-23T17:39:59","guid":{"rendered":"http:\/\/ncjolt.org\/?p=5134"},"modified":"2020-06-04T20:52:55","modified_gmt":"2020-06-04T20:52:55","slug":"can-monkey-do-this-job","status":"publish","type":"post","link":"https:\/\/journals.law.unc.edu\/ncjolt\/blogs\/can-monkey-do-this-job\/","title":{"rendered":"Can a Monkey Do This Job? Maybe Not, but Artificial Intelligence Can"},"content":{"rendered":"

Discovery methods continue to evolve. Twenty to thirty years ago, lawyers, usually young and expensive associates, manually sifted through hundreds of pages of documents. Today, with advances in technology, the discovery process for a similar sort of case might require review of thousands or tens of thousands of pages of documents instead of only hundreds.
\nArtificial intelligence (AI) may now help that process along. AI does more than just a word or Boolean search of electronic files. AI is a learning technology; it can refine the search as the search develops. It trains itself to \u201chot\u201d documents and can discover stories and concepts<\/a>, regardless of the verbiage employed, within a stack of documents.
\nThe way it works. Lawyers flag certain documents deemed crucial or provide examples of relevant documents to the AI system, which then relates those documents to documents in the search pool. The \u201clearning\u201d continues<\/a> to occur after the system has returned a number of documents to the lawyer and the lawyer has identified returned documents as helpful or not.
\nLawyers use AI for more than just discovery. AI can help with due-diligence review, legal research, contract analysis and docket analytics<\/a> for predicting litigation outcomes.
\nWhat is good about AI in the legal profession:<\/strong>
\nIn short, AI saves lawyers time and clients money. Especially in the discovery phase of litigation, so much time is needed to sift through documents. This process is one that AI developers have an eye on. Since nearly 80% of a company\u2019s data is unstructured<\/a>, AI can make sense of the unstructured data before an attorney lays eyes on it, thereby reducing the number of hours required to review discovery documents.
\nProponents of AI in the legal profession also claim its use is far cheaper for the client and far more efficient for the lawyers. One study published in the Richmond Journal of Law and Technology found that technology-assisted review (TAR) produced a 50-fold savings<\/a> in cost over manual review.<\/p>\n

This is because, according to a study by McKinsey & Co., 23% of lawyer work output can be automated<\/a> by AI.<\/p><\/blockquote>\n

TAR can also help a lawyer by ranking documents prior to review or by pushing the lawyer documents the computer feels would be helpful, thereby reducing the overall number of documents<\/a> to review. TAR can also increase lawyer efficiency by developing a cast of characters<\/a> who are relevant to a litigation, identifying additional key dates, and suggesting alternative search terms that might speed up the search or fill in the story.
\nWhat is bad about AI and the legal profession:<\/strong>
\nSome lawyers worry that the more legal work AI does, the less money lawyers will make. Implementation of TAR would certainly create shifts in the value chain, therefore changing the legal business model in terms of legal services procurement, billing\u2014and margins<\/a>. While paradigm shifts can be helpful in the long term, they are often messy at the start.
\nOf course, as many law students fear, implementation of AI systems will decrease the need for young associates, thereby compounding the difficulty of an already arduous job search.
\nMost importantly, AI cannot yet provide the insight, wisdom, and judgment<\/a> that a good lawyer provides. In the event AI becomes mainstream, clients may constantly force lawyers to explain why certain tasks where not cheaply completed by AI instead of expensively completed by an associate or partner.
\nWhat are some AI tools available to lawyers today?:<\/strong>
\nDiscovery Cracker<\/a> helps lawyers manage electronic documents for litigation. NexLP<\/a> uses AI to analyze data and identify trends. Story Engine<\/a> can read through unstructured data and summarize conversations. Ross Intelligence<\/a> uses the IBM\u2019s Watson cognitive computing system to enhance legal research. Ravn Systems<\/a> is a U.K.-based company that can, among other things, identify tables, price lists, clauses and charts from unstructured data. CaseAssist<\/a> takes information from lawyers, such as the nature of the case, goes through all the documents, and pushes to lawyers documents it deems helpful.
\nIn conclusion, very few believe AI can ever fully replace lawyers. Proponents of legal AI claim AI can add value to a lawyer\u2019s review and research capabilities, increase efficiency, but never supplant human judgment. After all, even AI needs attorneys to provide initial \u201ctraining\u201d or \u201ceducation\u201d on a case before it can understand what the client needs. Just as research moved from time consuming book searches to quicker online searches, AI may simply free up attorney time to complete work that is less tedious.<\/p>\n","protected":false},"excerpt":{"rendered":"

Discovery methods continue to evolve. Twenty to thirty years ago, lawyers, usually young and expensive associates, manually sifted through hundreds of pages of documents. Today, with advances in technology, the discovery process for a similar sort of case might require review of thousands or tens of thousands of pages of documents instead of only hundreds. …<\/a><\/p>\n","protected":false},"author":1,"featured_media":5135,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[51],"tags":[],"_links":{"self":[{"href":"https:\/\/journals.law.unc.edu\/ncjolt\/wp-json\/wp\/v2\/posts\/5134"}],"collection":[{"href":"https:\/\/journals.law.unc.edu\/ncjolt\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/journals.law.unc.edu\/ncjolt\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/journals.law.unc.edu\/ncjolt\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/journals.law.unc.edu\/ncjolt\/wp-json\/wp\/v2\/comments?post=5134"}],"version-history":[{"count":1,"href":"https:\/\/journals.law.unc.edu\/ncjolt\/wp-json\/wp\/v2\/posts\/5134\/revisions"}],"predecessor-version":[{"id":7083,"href":"https:\/\/journals.law.unc.edu\/ncjolt\/wp-json\/wp\/v2\/posts\/5134\/revisions\/7083"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/journals.law.unc.edu\/ncjolt\/wp-json\/wp\/v2\/media\/5135"}],"wp:attachment":[{"href":"https:\/\/journals.law.unc.edu\/ncjolt\/wp-json\/wp\/v2\/media?parent=5134"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/journals.law.unc.edu\/ncjolt\/wp-json\/wp\/v2\/categories?post=5134"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/journals.law.unc.edu\/ncjolt\/wp-json\/wp\/v2\/tags?post=5134"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}