What Does a Human-Centric Approach to AI Mean for the Ministry of Justice
Over the past year, colleagues across the Ministry of Justice have made real progress in building confidence using AI tools. From summarising case notes to improving information for people leaving prison, AI is already becoming part of everyday work across MoJ.
But as we move from experimentation to scale, a bigger question emerges: what does a human-centric approach to AI actually mean for us?
We often hear that AI works best when there’s a ‘human in the loop’ as it ‘frees up time for higher-value work’. These ideas assume technology is human centric because it automates routine tasks so people can focus on more meaningful activity.
AI is changing.
It is increasingly able to take on aspects of professional judgement and decision-making, the very parts of work many people value most.
This matters for MoJ because our work is not just technical, it is deeply human. We are not simply solving engineering problems, we are dealing with justice, risk, rehabilitation and trust.
A human-centric approach means recognising that technology is not destiny. The way AI is used is a choice. It is not about putting ‘humans in the loop’ of automated systems, but about designing systems where AI supports human judgement, not replaces it.
For MoJ, this starts with valuing the full nature of our work. Effective public service depends on a combination of knowing, doing and connecting: professional expertise, practical intervention, and the ability to build trust with people in difficult circumstances. AI should strengthen this combination, not fragment it.
It also means rethinking productivity. Success cannot be measured simply by time saved or tasks completed faster.
In our context, productivity is about better outcomes: safer communities, reduced reoffending, improved wellbeing, and more sustainable workloads for staff. Doing things better, not just quicker, must be the goal.
Trust cannot be automated.
Much of what probation and wider justice services deliver is relational.
The risk is that poorly designed AI could increase standardisation and central control, reducing the discretion and professional judgement that frontline staff rely on. A human-centric approach does the opposite, it empowers people across the organisation, not just at the top.
This requires a clear commitment to ethical practice. AI systems must be transparent, fair and accountable. Crucially, responsibility for decisions and their consequences, remains with us and not machines.
Within the Justice AI Unit (JAIU), our commitment to delivering better services and productive value, is based on empowering the frontline.
And although early days, we’ve shown how it’s possible to accelerate innovation within clearly defined guiderails, enabling the safe and ethical adoption of AI across the MoJ and its agencies, to deliver faster, better and more tailored services.
The progress we have made so far shows what is possible. The next step is to ensure that, as AI becomes more embedded in our work, it helps us deliver not just more efficient services, but more human ones.
Professor Phil Brown
Last updated on