7. Future Visions of Digitalised Criminal Justice

Courtroom obsolescence in favour of fully virtual courts arose during interviews and SCJ5 speculated:

As the technologies evolve, this quote recognises that the debate around remote hearings will become redundant (Wilkman et al., 2024). Certain participants accepted that ‘we’re in this world now where we accept that the court can be a digital one’ (DL23), and the court system should ‘go completely digital’ (P11). Fewer courthouses will be needed because, increasingly, much work can be done online (M4), and everyone is now on the same ‘cyber street’.¹ ‘Gone are the days’ when court has to be in-person (P7), and AVL is ‘here to stay’ (SCJ5), especially for traumatised complainants.

By contrast, DL19 speculated on ‘an extremely repellent dystopian nightmare’ of a ‘completely virtual justice system’ in which a defendant’s ‘entire life has been decided on a television screen’ in a small prison room. M1 thought that courtrooms would remain ever-present ‘important symbols to the community’ and felt ‘uncomfortable’ about a futuristic use of, for example, holograms in courtrooms and the ‘the lack of human interaction’. SCJ6 said that physical courts are ‘where justice is dispensed’ and that role cannot be substituted by ‘an avatar’ (compare Rossner & Tait, 2024). Despite many people having a ‘belief in the infinite possibilities of technology’, SCJ1 suggested that we will continue to need the special space of criminal adjudication, on the basis that courtrooms deal with ‘real human beings’. The ‘bricks and mortar’ (M4) of courthouses will remain essential due to the ‘social contract’ with citizens (P10), and their symbolism as legitimate ‘pillar(s) of democratic societies’ and open justice (Flower, 2025, p. 10; Mulcahy & Rowden, 2019). The majority of participants expected there will be a continuum of ‘a physical courtroom with a physical jury and a physical judge … But … there should be, and will be, more uptake of online methods so that witnesses from all over the place, can give their evidence online’ (P14). Jury trials can’t become completely virtual because it would be impracticable to bring together 12 random community members; they need ‘human interaction’ (P13).

However, my study gathered evidence on how the pandemic saw ‘a shift to videolinks almost the entire way [for] someone’s progression through the criminal justice system’ from arrest, bail, remand, through to finalisation: they might never attend court in-person (DL11). So, while physical courtrooms might continue to exist, might it be that, increasingly, defendants will never step inside one? There are clear indications that defendants will be increasingly excluded from courtrooms despite the fact that they have, traditionally, been the central figure in criminal procedure (Leader, 2010). For instance, during the pandemic, there was a period where defendants would never meet face-to-face with anyone involved ‘in the system that they were attempting to participate in’ (DL11). End-to-end, automated dispute management, without human intervention, is already used in civil disputes and for low-level traffic penalties in various jurisdictions (Mulcahy, 2024), yet the human journey through the process needs to be recognised (Mulcahy & Tsalapatanis, 2024).

Concerns were expressed regarding whether AVLs will replace regional justice including bush or circuit courts. M1 argued that while virtual courts can be functional, they don’t satisfy remote communities who are entitled to the presence of the courts as important social structures, and to see judicial officers ‘get a bit of red dust on their boots’ (DL25). This extends to physical bush or circuit courts, although DL1 suggested that some judges don’t like bush courts, preferring to use AVL ‘so they can just send people to gaol from 2,000 miles away’. The establishment of NSW Centralised Bail Courts, facilitated by AVL, points to more virtual centralised courts (Local Court, 2024), and a survey respondent wrote about watching: ‘a Saturday bail court at Parramatta all done via AVL. Extremely efficient, but totally isolating for the defendants in often extremely vulnerable circumstances’.

While, on one hand, there was a push for centralisation, on another, there were many suggestions to bolster localisation and regional justice digital infrastructure. Given Australia’s vast distances, increasing regional capacity and infrastructure would improve access to justice for everyone (WAS2), and address the need for high-quality technologies in a localised setting. One initiative suggested nationally involved ‘justice hubs’ to connect people with courts and services (WAS3). Similarly, DL30 recommended a small ‘booth’, ‘a little supported place’ with good screens and audio, so that clients could access the court system with ease and without taking a day off work or arranging childcare. To some extent, such measures are already in place and M2 observed that South Australia has community hubs in libraries from where people can appear if they don’t have digital access at home. In Western Australia, Legal Aid has a network of ‘virtual offices’ (DL37) in libraries and other community service agencies for clients to hold legal appointments with a lawyer. Participants advocated for more hubs in neutral ‘multidisciplinary centres’ (WAS2), not police stations (P16). Justice hubs would address issues concerning a lack of facilities in older courthouses and how the circulation routes for vulnerable witnesses and accused people often intersect.

Top of many participants’ wish lists was investment in better and more technologies. AVL is here to stay and will expand: ‘We’ve just got to get it right’ (DL26). There were calls for ‘a fit-for-purpose system that will carry us into the future’ including ‘an abundance of remote witness facilities’ (M5), and going beyond just ‘retrofitting courtrooms’ (M3). M5 explained that ‘the government’s put their big toe into the water’, but hasn’t funded the system adequately to reap optimal benefits. According to VJ1, it isn’t sufficient that technologies are merely functional, they need to be optimal, to ‘be as real as it can be’, retaining ‘the atmosphere’ of court. SCJ5 said ‘as the tech gets better, the difference … between in court and AVL’ will diminish.

While there is a lack of evidence to inform decisions regarding which technologies and configurations are optimal (Martschuk et al., 2024), DL18 identified the need for better facilities, such as headphones, for people with hearing difficulties because so often, remote defendants ‘literally can’t hear what the judge is saying’. DCJ1 recommended video cameras that are trained ‘on whoever is talking at the time’ so that the remote person is hearing voices connected with relevant faces on screen. Improved AVL could also provide more suitable facilities for interpreting, including Australian sign language, Auslan (WAS1). One of the ‘biggest bugbears about access to justice’ involved the complexity of linking remote interpreters with remote people (M1). The issue for interpreters is they need greater abilities to either interpret in-person or via enhanced access to optimal AVL equipment (Skinner et al., 2018).

Aligned with optimal technologies is the issue of the uniformity of systems across jurisdictions and agencies. Videoconferencing providers, VJ1, recommended: ‘Let’s all do this together and … get a completely interoperable system’. WAS1 said that ‘if we could have some nationally and accepted IT platforms, that would make everybody’s life a lot easier’. Police, lawyers, legal aid and courts should all be on the same integrated system (M4). However, interoperability cannot be done ‘on a piecemeal, silo basis, because then you’re going to have one arm not operating with the other’ (M5). The key is collaboration amongst the many justice agencies.

Many participants addressed improving the capacities of the summary jurisdiction which ‘does over 90% of all of the work’ (M4). P7 spoke of the ‘high volume’ Local Court, an ‘exhausting’ and ‘devastating’ workplace, which could be a leader in improving the system. However, improvements would require ‘solid funding’ for ‘future-proofing the Local Court’ (M5). Several interviewees suggested that low-level offending, particularly traffic offences, could easily go ‘straight in the online Magistrates’ Court’ (DL30) and ‘be done even by registrars in virtual courts … [the] creation of a fourth tier is something that we’re working on to push down that low-end work to a virtual courtroom’ (M5). M4 also promoted ‘the fourth tier’ explaining ‘the first tier is the Supreme Court. The next tier is the District Court, and the final tier is the Local Court’. Repetitive work, like parking or transport matters, can be done by registrars or an e-court, freeing up magistrates to do the ‘more complicated, contentious work’ (M4).

Any investment in courtroom infrastructure must be matched in police stations and corrections. There are insufficient facilities available in prisons (M1; DL3) and more facilities are required for confidential legal conferencing with clients in custody (Turner, 2021). DL13 identified the need for soundproofed AVL booths in corrections, recommending dedicated booths that resemble courthouse conferencing rooms (see Kashyap et al., 2018) ensuring access to clients before, during and after court matters. Acoustic issues between prisons and courts, ‘the banging … crashing … wailing … shouting’ (DCJ2), continue and need to be addressed in future spaces (McKay, 2018, 2020).

As discussed in Chapter ‘Remote Vulnerable Defendants’, communicating with clients in police stations by telephone is difficult when they are in the charge room alongside police officers (DL38). Several defence lawyers suggested that police are often hesitant to put people on AVL, ‘or pretend it’s broken’ and tell the lawyer to just do instructions on the telephone, leading to a loss of confidentiality for instructions’ (DL27). While ‘most police do the right thing … the potential for things to go wrong is too high when you don’t have both audio and visual’ (DL26). There is also the need for the ongoing independent scrutiny by lawyers of carceral spaces. When lawyers attend correctional centres and police stations in-person, they see the conditions in which people are being held (DL25). To the contrary, DL36 suggested that ‘gone are the days of face-to-face appointments’.

DL31 said that prisons have ‘a lot of work to do around how they deal with AVL. It’s not acceptable the practitioner can’t talk to a client on the same day when they need to’. Such issues highlight how courts are increasingly ‘an adjunct or an annex to the prison’ (DCJ2), connecting with my earlier research regarding the technological conflation of suspect/inmate management with criminal procedure (McKay, 2022; McKay & Macintosh, 2023). Going forward, greater attention should be paid to the control of the conceptual extension of courtrooms into spaces of detention.

My study highlights the need to promptly screen, identify and respond to the ‘markers of vulnerability’ (Travers et al., 2020, p. 11) on a case-by-case but methodical and universal basis. This aligns with the recognised ‘vulnerability identification deficit’ (O’Loughlin et al., 2024, pp. 67, 81), caused by tight legal definitions and unstructured approaches. Vulnerabilities must be identified and made known to all justice actors from the first point of contact (Hughes et al., 2022), so that communication modes may be tailored. For such purposes, systematic communication assessments, vulnerability screening tools and risk factors have been developed in various jurisdictions for frontline professionals and to facilitate diversion (e.g. Law Foundation, 2025; O’Loughlin et al., 2024; UNHCR, 2016). While police, correctional staff and lawyers should all be well versed in detecting the typical markers of vulnerability, they can still be missed due to ‘knowledge deficits’ requiring judicial officers to sometimes step into the role of assessor (survey respondent). Moreover, vulnerabilities are not always visible or immediately apparent (Menichelli, 2021). Nevertheless, interviewed judicial officers emphasised lawyers’ duties in knowing their clients and witnesses well. In this regard, my dataset points to the continuing benefits of physically meeting with clients or witnesses face-to-face, at least once, as an effective screening tool and means to identify appropriate supports. SCJ3 recommended developing protocols to ensure that lawyers meet their witnesses and clients face-to-face, to better understand people’s communication needs and any reasonable adjustments for effective communication, participation and procedural fairness.

Further training in trauma-informed and responsive practice was suggested by WAS3 who posited, in relation to vulnerable witnesses:

How great would it be to have the trauma-informed criminal justice system where we can integrate the principles of trauma-informed practice … safety, empowerment, choice, they are things that will enable our people to give the best evidence and … maintain a fair trial.

Trauma-informed approaches are seen as providing more humane and less stressful environments for vulnerable individuals. The aim is to support individuals’ coping mechanisms and to minimise re-traumatisation, thereby supporting their effective and coherent communication, full participation and inclusion, and securing their accurate, complete and reliable evidence—all attributes of procedural fairness. This is achieved by judicial officers and lawyers having greater awareness of ‘trauma-informed questioning’ (SCJ3), and how trauma might manifest through difficult or agitated behaviours (Judicial Commission, 2022). Measures include modifying the court environment and procedures to provide a greater sense of safety and, as my study finds, decisions whether remote modes will mitigate or intensify stress. Trauma-informed approaches involve training in understanding trauma, recognising its signs and impacts, and applying precautionary approaches, especially for Aboriginal and Torres Strait Islander peoples (Judicial Commission, 2022; Kendall, 2024). However, the benefits of trauma-informed practice and the recognition of communication needs ought not to be restricted to prosecution witnesses. A fairer future would ensure that vulnerable defendants’ communication needs are also systematically accommodated. For instance, DL37 spoke about communication difficulties with a remote client held in a police station, noting she was a ‘particularly vulnerable person who comes from an extremely traumatic background … known to be a bit mute at times’. Other literature has recognised the trauma-inducing aspects of criminal procedure on defendants as well as witnesses, complainants and victims (Kendall, 2024). This connects with the acknowledged overlap between offending and victimisation discussed in preceding chapters.

Greater training for court and corrections personnel in effectively operating and scheduling the various technologies was identified by many participants. Training for police regarding pre-recorded interviews was recommended to avoid admissibility ‘problems with recordings and visibility issues that arise from cameras being in the wrong spot’ (P18) and ‘leading questions’ (DCJ2). CCJ1 suggested the need for more ‘defined rules about the format and the technology’ used during police interviews to ensure the vulnerable person’s face and subtle gestures can be seen.

The dataset includes many responses concerning whether there are individuals for whom AVL is not appropriate, for instance, witnesses or defendants with cognitive impairments, people with strong accents or language needs, and witnesses who lack a private location to discuss sensitive issues. There are also circumstances when AVL is not appropriate according to many respondents: trials, defended hearings, crucial or contested testimony, credibility assessments, or for viewing complex evidence/exhibits or CCTV footage.

It seems unlikely that we can draw a bright line regarding when AVL should and should not be used. It would be impractical, and unhelpful, to suggest that all vulnerable individuals be either always included or always excluded from AVL, and greater attention needs to be directed to screening tools or assessment mechanisms to assist in the triage process. For the purposes of clarity regarding options, some respondents suggested changes to the legislation or AVL court practice notes, although these shouldn’t be too prescriptive or inflexible as technologies evolve. P10 stated:

technology is here to stay, but I think it’s a creative tension as to how to use it properly … there should be no settled rules … ultimately, it should be the courts that make the decision as to what’s going to happen in a particular case, guided by legislation, of course … it shouldn’t be too static.

DL31 suggested that with electronic justice becoming the norm, more standardised court practice notes are required to clarify when the three levels of courts are ‘happy to see AVL appearance and where they require in-person appearance’. Finally, more guidance is required for the use of AVL from police stations for bail applications because police are insufficiently aware of their obligations and safeguards (DL27).

Providing a more specific AVL opt-in/opt-out system seems vexed. Several respondents suggested that it effectively already exists, noting some limited choice, but also observed that making it about individual choice might add to a vulnerable person’s stress.

Regarding vulnerable witnesses, many prosecutors and witness assistance officers suggested that they all should, by default, give evidence remotely unless they explicitly opt for court. However, it’s important to keep choices available for victims (WAS2), and provide sufficient ‘information about the pros and cons’ (VS1). P15 asks vulnerable witnesses to consider: ‘At the end of the day, what do you want to achieve?’, and P14 tells witnesses that remote modes are designed to minimise stress but they can still attend court if they prefer. One survey respondent said that vulnerable witnesses should only appear remotely if adequately supported; they shouldn’t be expected ‘to give evidence about harrowing sexual assaults or violence at home alone and unsupported’. This links with ensuring that vulnerable witnesses are using facilities that are secure and meet technical standards and court protocols (James, 2025).

Victims’ needs were raised by survey respondents as a reason for everyone to be in-person in court: ‘often on sentence, for cultural or personal reasons, victims want to face an offender in-person, but if that offender is in custody already, they often appear remotely’. On this basis, offenders should not be allowed to attend sentencing by AVL. Caselaw evidences the reticence of some offenders to apologise to victims in-person, instead offering to engage via AVL.⁴

Regarding defendants, DL27 felt that while AVL is great for administrative issues in criminal courts such as adjournments, it ‘shouldn’t be everything by AVL. I don’t think that would be good for anyone’. DL18 said the choice to be transported from gaol should be left to the defendant, being the person directly impacted by the process and outcome. Transportation certainly deters defendants from opting to appear in court because:

the cells in the courthouse are terrible … border[ing] on inhumane. But being transported and searched … from the gaol to the courthouse, should be the choice of the defendant and not a judicial officer, or a correctional officer, and it shouldn’t be borne out of convenience or cost. What the defendant wants should be the overriding factor. (DL18)

The problem with this approach is that vulnerable defendants might not be aware of what is in their best interests. Other studies have suggested that remote hearings should only occur with the defendant’s informed consent (Bellone, 2024; UNODC, 2017) while the EHRC (2020) found that video hearings are often unsuitable for vulnerable defendants.

Whilst ‘not necessarily a technological solution’ (P19), several suggested that witness intermediary schemes ought to be expanded. WAS1 spoke about including a broader range of vulnerable witnesses who can automatically be entitled to use the remote witness rooms, videolinks or the intermediary scheme. Wi1 went further and argued for more jurisdictions to expand the scheme to vulnerable defendants because:

the line between victim and accused is very blurry, often, and the majority of people who come into contact with the criminal justice system as the accused will have been or will be in the future a victim themselves … It’s the same person.

Therefore, it’s not logical to differentiate between parties who have communication needs purely on whether they’re a defendant or witness. Regardless of vulnerability, P12 suggested that ‘everyone should have a witness intermediary’ due to the unfamiliar and stressful environment. Other reports have argued for a parity of special measures and intermediaries for both ‘genuine’ and ‘deserving’ vulnerable witnesses as well as vulnerable defendants (O’Loughlin et al., 2024, pp. 75–76, 82; EHRC, 2020). One day, perhaps, digital companions or bots could support remote defendants (Kremens & Peristeridou, 2025).

Going forward from this dataset, clearly AVL and associated measures have a part to play in the future criminal justice system (Smith et al., 2021). Overall, participants recognised the ‘need to embrace technology’, especially that which enhances court procedure (P13). Without doubt, there are many universal and non-contentious uses of AVL that ameliorate digital vulnerabilities, so courts and the profession ‘shouldn’t be standing in the way’ of continuing technological progress ‘without good reason’ (SCJ5). Ultimately, most agreed that the challenge is ‘how we can do it better … work smarter’ and keep it meaningful (P7), while maintaining a focus on addressing ‘social inequality and minimising people’s entrance into the justice system: earlier and better managed interventions’ (DL19). Improvements will involve the collaborative efforts of all ‘here in the trenches’ with a view to enhancing the system, ‘so that it’s not so “groundhog day”’ (P7).

With the possibilities of not only virtual courts but also AI and automated decision-making, there is a need for careful oversight to ensure that vulnerable individuals are adequately identified and not processed through the whole system as digital objects without human contact and scrutiny. Perhaps, particularly for vulnerable defendants, as criminal process is increasingly digitalised and the significance of physical presence diminishes, the question arises regarding how a suitable mechanism might operate to ensure defendants have been seen in-person, at least once. At present, the burden seemingly lies on defence lawyers to ensure adequate human contact.

Whether the shift to digitalised criminal justice guarantees procedural fairness remains unclear. The key, according to DCJ1, is to ensure that any remote individual ‘is, in fact, participating fully in the proceeding. That they understand it. That they can hear it. That their needs are being taken into account adequately, and that everyone works to achieve an adequate level of comprehension’. To achieve and operationalise this, technologies must be of high quality and used in ways that support vulnerable individuals’ resilience. Moreover, court procedure and legal principles require ongoing recalibration to align with digitalised modes.