Remarks on the Role of Technology and Innovation in Advancing the Delivery of Justice

Justice Susan Glazebrook[1]

Question: I understand that New Zealand is currently in the process of introducing “Te Au Reka”, a digital case management system that promises increased transparency and ease of access for court users.[2]  But as I have previously discussed, such interventions face numerous barriers to successful implementation.  In your view, what are the main challenges that the New Zealand courts face in implementing digital technologies, and what can we learn from these challenges?

Answer: The framework you mention, Te Au Reka, was conceived and implemented against the background of past failed technology projects, not just in New Zealand but around the world.  The causes are varied and there is not time to cover them all, but I will discuss the ones I see as most significant for future projects.

Often failures were a result of poor planning, with systems being designed without proper input from users, meaning they were not user friendly and did not meet user needs.  They were technology and management led rather than user focused.[3]  Some of the failed projects were also too ambitious, including those designed from scratch rather than making appropriate use of tried-‍and-tested, off-the-shelf solutions.[4]  And, in many cases, underlying inefficiencies in systems were not addressed as an essential part of the planning process.[5]  There is not much point in spending money on technology if it does not improve systems and enhance access to justice.

There is also a real issue with the pace of technological change.  In past projects, many systems or parts of systems were redundant even before they were implemented.  This means future proofing, to the greatest extent possible, is vital.  The digital divide, both within countries[6] and between countries, can also affect access to justice and the implementation of technologies.  Measures to address this are essential.  More generally, adequate ongoing support and training is needed for any technology project to be successful.

New Zealand’s Digital Strategy for Courts and Tribunals, of which Te Au Reka is a part, was judge led and designed after extensive consultation with users and potential users.  The document sets the framework for all uses of technology in the New Zealand courts.  It is intended to be a dynamic document, and its underlying philosophy is that of promoting meaningful access to justice for all, with projects that are user led and focused.[7]

Question: In your 2019 paper “Women and Technology”, you wrote that “[t]here are many ways in which [technology] has changed our lives for the better.  It does, however, have its dangers.”[8]  Do you think technology is changing justice systems for the better?  More broadly, how is technology impacting justice systems, and how can we ensure that it supports access to justice for the more vulnerable?

Answer: I answer these questions with particular emphasis on new technologies like generative artificial intelligence (AI)[9] and I think the jury is still out on whether these are changing, or will in the future change, justice systems for the better.  It is early days, although the pace of change has been rapid and no doubt there will be continued improvement in the products being offered.[10]

Certainly, in the short-to-medium-term, generative AI products have the potential to save time for lawyers, judges and court staff through their ability to order and summarize the content of large quantities of documents accurately.[11]  Given the pressures on the system and the delays, saving time can only be a good thing.

Using generative AI tools for legal research should be treated with a bit more caution at present,[12] although the tools being released by legal publishers work off established databases and therefore are less likely to have the hallucination issues other tools suffer from, including making up case citations.[13]  Nevertheless, at present generative AI can still get answers very wrong.[14]

Using technology for actually making decisions must be treated with even more caution.[15]  Technology has not taken a judicial oath and it is not required to take moral responsibility for the decisions it makes.  It is not able to pay respectful attention to a fellow human being and it is not a member of the community to which any decisions it makes will apply.  Such tools are also wont to suffer from the same biases as the material they are drawing on.[16]  With proprietary systems the way they work is also effectively not accessible and this creates issues with testing the material generated.[17]

I also mention, in terms of other dangers, that deep fakes generated by AI have the potential to make evaluating evidence much more difficult.[18]  Issues of confidentiality and security would also need to be addressed in relation to all the new technologies.

In terms of the ability of technology to provide better outcomes for the most vulnerable, there is no doubt that generative AI has the potential to make the law more accessible.  Virtual hearings also make it much easier for court users to participate without the need to travel or to take time off work.[19]  Some users may also feel much more comfortable online than in court.  Courts are an alien and stressful environment for most people.

Unless the digital divide is addressed, however, technology has the potential to make matters worse rather than better for the vulnerable, especially where digital filing is concerned.  Systems and support structures must be designed in a way that promotes and supports access, and that ensures technology does not exacerbate existing inequalities.