Many judges and lawyers still view artificial intelligence with skepticism, but rapidly evolving technology shows great promise. Are the courts ready for the coming push towards the use of more electronic materials in litigation?
But inside the courtroom trials are run largely as they were in the typewriter age.
People may have laptops or iPads in front of them but it’s hit or miss whether there’s actually an Internet connection. And while there is likely some electronic evidence, bankers’ boxes of paper documents remain a fixture in the majority of corporate cases.
“We still litigate the way we did a couple of centuries ago,” says David Bilinsky, a lawyer for the Law Society of British Columbia and a Fellow of the National Center for Technology and Dispute Resolution at the University of Massachusetts.
“Sure, we’ve got [real time] transcripts now and electronic equipment and we bring in monitors and that sort of thing, but by and large most people still bring paper data.”
In fact, Ontario Superior Court Justice David M. Brown took the entire profession out to the woodshed last year over its continuing attachment to paper trials, saying the courts are treated like “some kind of fossilized Jurassic Park.”
The judge made the remarks after asking lawyers at a pre-trial case conference whether an October hearing should be conducted as an e-trial. Some counsel indicated they preferred to work in paper.
Wrong answer. He not only mandated an e-trial, he took the opportunity to blast lawyers and judges over the stodgy pace of change. If the profession doesn’t wake up to the fact that it has to evolve, he warned, the courts are going to discover they have become “irrelevant museum pieces.”
The music industry moved from vinyl to eight-track, cassettes, CDs and digital to keep pace with the needs of its users, he said, asking “why should courts and lawyers be any different?
“Our court must choose: are we a court of the past or a court of the future? I vote for a court of the future. I know there are judges available who are chomping at the bit to conduct more e-trials.
“Paper must vanish from this court and, frankly, the judiciary cannot let the legal profession or our court service provider hold us back.”
As a clarion call to a generation of Canadian lawyers and judges, it was worthy of Bob Dylan.
Justice Brown may be frustrated with the pace of change but the revolution has begun. Software developers have come up with litigation apps that will allow users to do everything from conflict searches to opening new files and designing courtroom presentations right on their smartphones or tablets.
The new technology is promising to transform the way litigation is practised — starting with the use of artificial intelligence pre-trial.
“Artificial intelligence is not pie-in-the-sky anymore. It’s here and it’s real,” says Martin Felsky, National e-Discovery Counsel at Borden Ladner Gervais LLP in Toronto. “It can really do things.”
People who work in the field say it’s not much of a stretch to imagine that computer programs will be used in lieu of judges to decide low-value litigation in the not-too-distant future.
The truth is, they’re capable of it now.
When most people think about litigation technology they think of e-discovery, which formed the first frontier in the 1990s in response to the birth of email. Litigators faced with reviewing thousands of electronic messages, maybe a warehouse full of boxes, turned to outsourcers and computers to help.
But electronic information today is coming at a blistering pace that was unimaginable even 20 years ago.
Material that needs reviewing can be found on desktops, laptops, phones, iPads and external hard drives as well as a company’s servers.
While email is still key, admissible evidence can also be found in records of search engine requests, visits to various websites, security cameras, email attachments, spreadsheets, GPS geo-locators and voice-mail messages. It can even be embedded in a document’s metadata, such as the date and time it was written, who sent and received it, and whether it was edited or forwarded.
To get a sense of how overwhelming the amount of pre-discovery information can be today, consider that large corporations can have more than 10,000 tapes of information. Each one, if printed, would create a stack of paper more than 300 kilometres high.
The generation of e-discovery software that relied on key-word search programs may no longer be enough to weed information down to manageable levels. Programmers racing to keep pace have turned to artificial intelligence.
Felsky says technology-assisted review, also known as computer-assisted review or predictive coding – software that uses machine learning to pull relevant documents – is exactly the kind of thing the new generation of litigation technology excels at.
You teach the computer what you think is relevant based on a review of a sample set and the computer then applies the newly learned algorithms – or logic – to all other relevant documents.
“It’s a step beyond what you would do with a keyword search,” he says, adding he’s convinced artificial intelligence is better than humans at some aspects of document review because where people can get tired, bored or just make mistakes, a software program cannot.
Yet artificial intelligence is still viewed with skepticism by many in the profession, he says, because it’s so new there are competing platforms and no accepted standard way of doing machine learning.
“The technology needs to get more established and develop some standards around it before lawyers feel really comfortable recommending using it to their clients, so it’s still not in widespread use.
“I’d say we’re in a transition between two worlds. One is the world of the old-fashioned keyword search combined with a manual review of one document at a time. The other is the world of let’s push a button and here are the relevant documents.”
Felsky likens the situation to self-driving cars, saying while they are now legal in California, “I’m sure if you got in one you’d be a little nervous. You’d feel a little vulnerable. So it’s a bit like that. It’s a new technology and lawyers are used to looking at documents, reading documents and understanding documents themselves — so to push a button and let the software determine what’s relevant is scary.”
He feels certain that while using artificial intelligence may cut down on the number of lawyers needed on a case, it will never completely do away with the requirement for human oversight.
“When we go on planes we rely on autopilot to get us from one airport to another, but we still have a pilot in the cockpit to apply judgment. So what proponents of technology-assisted reviews – and I’m one of them – will tell you is we’re not just handing over the keys to a computer, we’re using a very powerful tool to help us apply our own judgment. That’s why I think it will grow in importance, and eventually become the norm.”
People who do a double take at the thought of artificial intelligence might keel over backwards at the notion of the Watson Debater, a computer being developed by IBM that can synthesize information – the facts and the law – then develop arguments for different sides of an issue.
Robert Weber, Senior Vice President and General Counsel of IBM, has said Watson could have “implications in the law.” IBM did not respond to requests for an interview, but Weber has said, “I think Watson could pass a multistate Bar exam without a second thought.”
Using computers to develop arguments or even preside over disputes is not as far out there as it sounds. Online dispute resolution that uses technology to facilitate the resolution of disputes between parties, and technologically assisted mediation are both becoming much more common.
Bilinksy advises lawyers and law firms on practice management and he points to eBay’s online dispute resolution system as a perfect example. The e-commerce giant handles an estimated 60-million disputes a year between buyers and sellers – and roughly 90 per cent are solved with the two parties interacting with the software.
Bilinsky, who stresses he is expressing his personal views and not the Law Society of British Columbia’s, says another example is BC’s new Civil Resolution Dispute Tribunal, which offers people an online alternative to going to small claims court.
Available to parties involved in disputes of up to $25,000, the tribunal’s online negotiation tool will guide the two sides through a structured negotiation phase, with the aim of helping them reach a settlement without the need to show up at a hearing. They’ll be able to access the platform via the Internet — and carry out party-to-party negotiation at their own times and their own convenience.
So how far does that take us?
Fred Galves, a professor at the Pacific McGeorge School of Law in Sacramento, California, says as algorithms become increasingly more sophisticated, automated decision-making systems could eventually be used to determine all sorts of disputes.
“There are certain types of litigation cases that might lend itself to this — a divorce, for example, where you have child care that needs to be calculated and property to be divided. If you can reduce it to math and you can reduce it to steps in logic then the computer, based on gaming theory, might actually be able to litigate some of this stuff for you.
“Who knows what the brave new world holds?”
Galves, a leading US scholar on the use of technology in the courts, says the issue right now is not so much that computers will someday replace lawyers as it is lawyers who use computers will replace lawyers who do not.
It’s already happening in the area of litigation-presentation technology, he says, as it becomes increasingly apparent that the savvy use of technology in the courtroom can affect a trial quite dramatically. Studies show people retain about 10 per cent of what they hear and 20‒25 per cent of what they see — and that goes way up when you combine the two.
Visual aids in the courtroom aren’t a new concept, he stresses, it’s just that the flipcharts are giving way to inexpensive apps like TrialPad that allow litigators to prepare sophisticated graphics and video to help make dense subject matter more digestible for judges and juries alike.
But Galves, who teaches civil procedure and evidence, says the real potential of trial technology goes far beyond that, especially in bet-the-farm litigation that is not as cost-sensitive as might normally be the case.
Where the client is willing to pay the cost of having all the examinations videotaped, he says, the video can later be synchronized with the transcript of the exact words the person used.
That means at trial, if the witness were to contradict himself, the litigator would be able to pull up the transcript and page number with the original statement within seconds — and play the contradictory testimony for the court on the spot.
The traditional way of handling contradictory evidence, Galves says, is to go back to the office when court adjourns and comb through the transcript to find the discrepancy. “You go to court the next morning prepared to show how the witness was lying by reading the testimony. But it’s the next day, you’ve sort of lost the moment, you have to remind everyone what was going on.
“Contrast that with what happens on TV news when the media catches a politician lying. They show a copy of the transcript, or play the clip of the speech where he said something different, back-to-back and it’s very powerful. If you can do that in court it’s very effective.”
Live streaming is another potentially powerful tool, with real-time court reporting allowing lawyers and their experts back at the office to watch testimony and text the litigation team with inconsistencies, potential objections or possible arguments through a closed chat room.
New software will even help litigators to decide whether to take a matter to the Supreme Court.
Addison Cameron-Huff, a Toronto technology lawyer, analyzed 4,261 Supreme Court of Canada judgments and wrote software code that offers an odds calculator based on the court’s past decisions.
Technology is having an equally profound effect on settlements and the way large class-action lawsuits are calculated and applied, says Toronto-based Eric Khan, National Director of Deloitte Canada’s class-action services offering. “In the past, you didn’t have the ability to look at an exhaustive set of information. If you’re looking at a class of 200,000 people, in the past you’d try to find a representative sample and based on that you’d extrapolate and make some decisions.
“But today, technology now allows us to review 100 per cent of that population. You can look at 200,000 class members and know with relative certainty how they were impacted by something.”
Everyone agrees the technology is changing fast. But are the courts ready?
Martin Kratz in Calgary, who heads the intellectual property practice group at Bennett Jones LLP and has sat on a number of Alberta commissions looking at moving the courts into the electronic age, says the Bench is generally very open to embracing technology and paperless courts.
The difficulty is, he says, as with most other walks of life, the court system is made up of people with vastly differing abilities in using technology.
“There are very sophisticated tools available, and some judges are more embracing of those tools than others in part because of what they grew up with and what their experiences were. I would say anecdotally – and with the observation there are always exceptions – that the younger judges have used computer technology much more through their legal careers and are much more comfortable with it.
“So I think over time you could reasonably expect to see an increasing push towards the use of more and more electronic materials in litigation.”
For now, uncertainty over the security and integrity of the systems is keeping the brakes on faster adoption of litigation technology, he says, because judges understandably want confidence that what they’re doing is “secure and free from being hacked.”
But Kratz believes a lack of resources is probably the biggest single culprit slowing the move to paperless courts. “All the technology costs money and it changes so quickly that the hardware and software you buy today can be obsolete in three to five years. So there’s a capital cost. But there’s also a training cost because if you adopt systems that change every three to five years, all the participants in court have to be trained to use it again. So human nature and budgetary constraints will tend to slow down the adoption of technology.”
Asked where he sees legal technology headed in the future, like others Kratz sees automated decision-making used in disputes where the law is clear and the facts are simple — a move he says that may help unclog the backlog of cases plaguing so many Canadian courts.
Automated decision-making is already used for the arbitration of domain-name disputes, he points out, and while it has humans involved in the process, “it’s highly automated and all communications are electronic. The cases tend to be simple, typically cybersquatting types of disputes, and the rules of what is bad-faith behaviour are known in advance.
“We’re increasingly using decision-making systems in our cars and appliances to deal with simple fact situations where the variables are known. So in a legal context, if you think of access-to-justice type issues, the cost of the system will drive looking for solutions to widespread basic problems.”
Money is not a factor just for the courts.
As the algorithms and analytics that underpin litigation technology become more intelligent, law firms are going to have to decide whether to try to recoup some of the billable hours lost to computers by doing the technological part of the work in-house.
“The problems law firms face now are do they build something in-house – paying for the software, trying to keep on top of the software, having personnel that are trained to do that on a full-time basis – or do they go to a service provider that can provide on a litigation-by-litigation basis?” asks Corey Fotheringham, National Director of Computer Forensics & eDiscovery Services at Deloitte Canada.
“That’s a difficult decision in this day and age.”
It’s not the only decision technology is forcing on law firms. With so much content and process in both litigation and transactions starting to move into the digital realm, some law firms are looking at whether they can go paperless as well.
Toronto-based Bereskin & Parr LLP is among those that have already made the commitment. “Clients are expecting firms to be more efficient so we are looking to maximize the efficiencies we can gain through technology,” says Victor Krichker, a member of the firm’s executive committee. “We believe this will save money and make us more efficient.”
Walk into the office of any litigator and chances are excellent all the shelf space, and possibly a portion of the floor space, will be littered with piles of paper documents. If you can do away with the bankers’ boxes and piles of paper by putting documents on hard drive, it will cut the need for storage space.
“Once we’re completely paperless, we expect that our office-space needs will decrease,” Krichker says.
It should also save money the firm now spends to move files among its four offices. “We hope not having to move files, look for files, not having to wait for files – having essentially all of our documents available to all of our lawyers regardless of office, and in near-real time – will make us more productive.”
Krichker acknowledges that some people are a bit apprehensive about the change, which the firm is hoping will come into effect by the end of the year. “Any time you’re asking lawyers to change, there’s a certain amount of anxiety and this kind of technology really goes to the core of how professionals at this firm do their day-to-day work. Any significant change in that obviously is going to create some worry, so people are definitely nervous, but I think people also feel this is something that could bring significant benefits.
“We’re hoping one of the main ones will be more flexibility in people being able to work from home. When your files are located online, they can be accessed from anywhere. It will be easier to work outside the office.”
As it becomes easier for lawyers to work outside the office, and easier for courts to adjudicate outside the old-style courtroom, it’s not hard to imagine litigation 25 years from now may look very different.