All summer long, the chorus of voices predicting artificial intelligence will transform the market for legal services’ market has been building to a crescendo: It started in August when Dentons partnered with IBM’s Watson and continued this month when the UK’s Riverview Law purchased a knowledge automation business.
“It is now imperative for legal technology providers to recognize the potential of AI technology within the legal services sector and step up their efforts to develop solutions,” Hugh Logue, an analyst at Outsell wrote last week in a note to the company’s subscribers.
But a new paper in the UCLA Law Review Discourse, entitled “Four Futures of Legal Automation,” attempts to cast doubt on predictions about what the future holds.
“Whenever I see people say there’s only one future, and get on the bus or get run over, that really disturbs me,” said Frank Pasquale, a co-author of the paper and law professor at the University of Maryland. “I really think it’s important for people to be more honest about how open the world is.”
Pasquale, also a visiting fellow at Yale Law School’s Information Society and a member of the NSF-funded Council for Big Data, Ethics, and Society, co-authored the paper with Glyn Cashwell, a computer programmer at the national security firm Vistronix who is studying law.
The thesis of the paper is that focusing only on the technological advances in artificial intelligence is not enough because the level of regulation or deregulation of legal services will play an equal role in shaping the future of legal automation. That is, deregulation of legal services can render extremely complicated legal tasks into simple legal tasks.
In an example of how deregulation influences the practice of law, the authors cited the trend among courts permitting individuals to relinquish their right to join class actions via consent through a terms of service agreement. “A robot could dispose of nearly all cases arising in the wake of such agreements,” the authors write, “if the only legal issue critical for the vast majority of consumers was whether they had ‘agreed.'”
Such deregulation, the paper argues, will make the outcomes of once complicated cases, more predictable, and more easily automated.
But new rules could complicate simple disputes. For example, the authors pointed to European laws that give citizens the right to be forgotten on the internet, meaning individuals can ask Google to omit certain information about themselves such as a prior bankruptcy, which has created complicated issues in expungement law where previously none existed.
Looking at just those two factors, the susceptibility to automation of any given legal task, and the level of regulation, the authors created a framework to imagine four different futures of legal automation. These range from a highly regulated, highly automated future to a society with little regulation and little automation and everything in between.
Ultimately, the authors resist making any predictions: In an interview, Pasquale acknowledged that the future may be some blend of all four scenarios and said he intends to write more on this topic. The paper notes that law makers and policy makers should be aware of the consequences that automation and regulation have on individual’s rights and how these two factors can affect inequality in society.
“We hope the scenarios we have described have demonstrated that the future of law and computation hinges on broader social trends outside of law, and thus is far more open ended than most commentators now suggest,” they conclude.