Why is demand important in justice?

Justice systems are predominantly analysed in terms of the quality of justice they provide and their level of accessibility. However, justice systems, like all public institutions, operate under resource constraints and must allocate those resources efficiently to achieve the best possible outcomes. In practice, a perpetual challenge for justice system administration is managing fluctuations in demand, i.e., changes in the volume of cases lodged with courts or tribunals by litigants.

Increases in the volume of cases being brought can be a signal that systems are operating more efficiently in relation to promoting access to justice. However, significant or rapid increases in case volumes can also create backlogs and delays, and ultimately risk compromising the quality of justice and trust in the system.

What causes demand fluctuations within justice systems is a complex and under-researched question. Often, jurisdiction-specific, local factors are likely highly relevant. For instance, new legislation reforming rights, entitlements, or procedures might affect the number of claims lodged in related courts and tribunals.

What are senior judges saying about AI-driven demand?

More recently, though, some senior members of the judiciary have suggested that the growing availability and use of AI chatbots are driving increased demand, by raising the number of cases being lodged. This, of course, has the potential to have a much more global effect on the demand that justice systems experience, cutting across jurisdictions within the justice systems as well as across national boundaries.

Last month, the Master of the Rolls—the senior judge who serves as Head of Civil Justice—Sir Geoffrey Vos said in a lecture at the Old Bailey:

AI is now being used by almost every individual litigant in person and small business. The first port of call used to be a lawyer if one was available and affordable. Now the first port of call is ChatGPT or CoPilot. Whatever answer generative AI gives, the would-be litigant in person can easily use it to transform a mass of documents and personal information into an arguable legal claim. This means that, in future, we will see many more civil claims because AI can create them for free, whereas previously claim numbers were limited by the availability and cost of lawyers. Judges and courts need to be ready to deal with an AI revolution that may vastly increase the number of civil, family and tribunals claims with which the justice system has to cope.

10,000 miles away, at a presentation to the Victorian Bar Association, Justice Adam Hatcher went further. Justice Hatcher is the President of the Fair Work Commission, a tribunal which handles industrial relations cases. In discussing the ‘recent impact of AI on the work of the Fair Work Commission,’ he explained:

If I had to describe how the FWC has experienced that impact in a few words, I couldn’t do better than refer to the famous Ernest Hemingway quote about how you go bankrupt: ‘Gradually, and then suddenly’. It became clear over the course of 2025, after some signs that this was beginning to occur in previous years, that the FWC’s operations were being significantly disrupted by the availability and use of AI tools… The ‘normal’ up until about 2023 was a bit above 30,000 matters a year. However, there was a sudden increase in 2023–24 to around 40,000 matters in total. That was a record year. Then in 2024–25, there was a further jump in 2024–25 to about 44,000 matters. That was another record year. For 2025–26, we are looking at a number between 50,000–55,000, which will be another record. There is no sign of this growth trend plateauing out, and we have no idea what the ‘new normal’ will be… That this is principally being caused by the increasing use of AI tools by potential litigants is, in my view, the only reasonable inference which can be drawn.

He went on to describe how this increased volume of cases has meant that, despite efforts to increase capacity, the Commission’s ‘performance has now started to slip.’ He pointed, by way of example, to cases taking longer to resolve due to the scale of the overall caseload.

Is AI really driving demand?

While it is still a conversation overshadowed by debate about the possible use of AI within institutions, the capacity of AI to generate increased caseloads and traffic for public institutions is increasingly recognised, and there are signs it could prove to be a major challenge for justice systems.

However, it is important that explanations that increased caseloads are being caused by AI are evaluated rigorously and critically. Part of any such evaluation is an obvious question about evidence: whether access to AI chatbots is having this effect, and, if so, exactly how it is operating. It is not clear at present if, where, or to what extent AI use is responsible.

Importantly, if AI is responsible in some meaningful way, attention must be paid to the specific mechanisms through which AI could be driving increases in caseloads. Is it because AI is reducing learning burdens by better signposting the availability of justice to litigants? Is AI increasing the awareness of litigants that their problem is one that could be justiciable? Is it because the burdens of complying with justice procedures are lessened when a litigant can quickly generate text? Or are there other mechanisms at work here?

The suggestion that AI-driven caseload increases could have a growing impact on the justice system in the coming years ought to also make us think more about what has been accepted about the status quo. In particular, the continuing problems of access and the continuing problems litigants—especially unrepresented litigants—face. While AI-driven demand could prove a serious challenge for the justice system in practical terms, it is also important to reflect on what this might reveal about the justice system itself.

Could we see reforms to manage AI-driven demand?

If what the Master of the Rolls and Justice Hatcher observe transpires to be accurate in the long run, we will no doubt also see measures introduced to respond to this challenge. Justice Hatcher, in his speech, explained how the Fair Work Commission is already reforming its procedures to ‘modify…forms to include an AI disclosure requirement.’

Reforms such as these, if justified, would ideally be shaped not only by the best possible evidence on demand mechanisms, but also an understanding of the likely effects of interventions—including both institutional effects and perceptions of trust in the justice system. They should also not be detached from reflection upon the persistent accessibility challenges within the justice system, of which this may be symptomatic.

We are grateful to Professor Matthew Groves for bringing to our attention the issue arising in the Fair Work Commission. We are also grateful to helpful comments from the UKCLA blog editors.

Joe Tomlinson (Professor of Administrative Law, The Dickson Poon School of Law, King’s College London)

Jed Meers (Senior Lecturer in Law, York Law School, University of York)

(Suggested citation: J. Tomlinson and J. Meers, ‘Will AI Create a Caseload Problem for Justice?’, U.K. Const. L. Blog (16th March 2026) (available at https://ukconstitutionallaw.org/))