Canadian Lawyer

18 O C T O B E R 2 0 1 8 w w w . c a n a d i a n l a w y e r m a g . c o m T he biggest event in the legal technology calendar is ILTACON, held every August. The International Legal Technology Association is a member-led organization of legal and technology professionals sup- ported by the vendors and consultants in the industry. This year alone, more than 4,000 people descended on Washington, D.C. for a week of educational sessions, networking, training and demos. Both innovation and artifi- cial intelligence were once again scattered throughout the week's agenda, but some of the intense hype I remember from last year seems to have worn off. Back in 1995, Gartner realized that the amount of media hype and the excessive enthusiasm surrounding new and emerging technology affected society's adoption of that innovation. The hype cycle is now pretty much an institution in the high- tech industry and helps explain how a technology evolves over time. Insiders use a roller-coaster-looking image to show the life cycle of a technology and make pre- dictions about when an innovation will mature and achieve mainstream adoption. I think of the hype cycle as a graphical representation of Amara's law: "We tend to overestimate the effect of a technology in the short term and underestimate its effect in the long term." There are five key phases to Gartner's hype cycle for a technology. The first is the "technology trigger" where early and overly excited media interest in brand new tech- nology happens. This enthusiasm and excitement about the latest and greatest builds into the hype that we have all seen for things such as machine learning, blockchain L E G A L I N N O VA T I O N N O W FROM HYPE TO PRODUCTIVITY By Kate Simpson and AI. This hype infects the industry, its end users and investors, which in turn feeds the media even further with more hype, and builds into what is termed the "peak of inflated expectations." That occurred at last year's ILTACON where it was standing room only at the machine learning and AI sessions talking about neural networks and the power of algorithms. Looking around at this year's conference, it seemed to me that we have finally peaked with artificial intel- ligence, blockchain and machine learning. Not that there isn't still some residual enthusiasm about these technologies and their possibilities; it's just that the talks this year struck me as being more focused on the hard and unsexy work that people are undertaking to understand the prac- tical applications. If that's correctly reading where we are, then we're heading into the "trough of disillusionment" over the next couple of years as stories of imple- mentations that fail to deliver reach that same media and interest begins to wane. Excessive enthusiasm will be replaced with excessive disap- pointment in the stories that the media begins to report. But we shouldn't mistake the trough of disillusionment with "it's not going to happen." Rather, it is about resetting those expectations from the inflated hype to the realities of what the technology can do. Early adopters will use this time to explore the technology and understand where the opportunities lie. Take the recent announce- ments by 12 financial regulators around the world, including the OSC, and technologists to create a "global sandbox" (the Global Finan- cial Innovation Network) to test out emerging technologies such as blockchain and AI or the new blockchain consortiums of Enter- prise Ethereum Alliance and the Global Legal Blockchain Consortium. These early adopters are diving into the technology to figure out how to embed it into their products and ser- vices to be ready for the mainstream adoption that will surely follow. Machine learning and AI both rely on good data to train and power up the algorithms. The data is used by humans to train the algorithms to get smarter and faster over time. So, these early firms and technology companies are taking this time to pull the quality data together and to methodically train and adjust the algorithms to do a certain thing well (such as find all change control provisions across a set of agreements). They are figuring out where the most value for lawyer and client adoption will be. Most of this work, however, is not interesting Technologists talk about the five phases of technology hype and legal tech is no exception @k8simpson O P I N I O N

• Cover