Forget all that stuff about £700million and 1,300 new Probation Officers - the answer as we now know is AI and Justice Transcribe is boosting productivity amazingly!
Justice Transcribe has saved an estimated 266,667 working hours across more than 1.6 million probation meetings since its rollout, according to the National Audit Office. The Ministry of Justice projects the tool will save around 450,000 hours annually for frontline officers.
Ok, there are one or two problems:-
i.e. shitloads of extra work as highlighted by the NAO report: "the pilot... recommended regular quality assurance audits and training on how to undertake a meaningful human review. Training and guidance materials emphasise the importance of reviewing the AI notes and the app included reminders for staff to do this at the point of use. For individual meetings, probation officers are expected to review the outputs from the tool before they add them to the national case management system."
There'll now be new well-paid managerial-level posts created for the chosen chumocrats who will deliver JT training & guidance for the frontline monkeys... and, of course, send a shitload of new emails reminding those monkeys to "review the outputs from the tool before they add them to the national case management system."
It aint gonna end well...
*****
Justice transcribe still requires you to review and edit what it has recorded…. That still takes time…. which it often records names / places wrong. It also doesn’t capture body language or HOW things are said/ the tone. That can be so revealing.
“no intervention completed due to crisis management and stabilisation”… you could play AI buzzword/ buzz phrase bingo!!!
Ok, lets play AI buzzword/ buzz phrase bingo!!!
Over to you.....
https://www.theguardian.com/commentisfree/2026/oct/20/ai-supposed-to-save-time-opposite-government-uk?CMP=Share_AndroidApp_Ot
ReplyDeleteI’ll have ‘meaningful engagement’, ‘explored protective factors’, ‘emotional regulation’, ‘strengths-based approach’, ‘identified areas for further exploration’ and, of course, ‘crisis management and stabilisation’ for a full house, please!
ReplyDeleteMaybe the biggest buzz phrase of all is ‘efficiency savings’.
The Guardian article raises a rather uncomfortable question about AI. Just because something can be produced more quickly doesn’t necessarily mean it creates less work. Sometimes it simply redistributes that work, often to the very people who are already struggling with excessive demands. And that brings us neatly to Justice Transcribe.
We’re presented with huge figures about hours saved, but what's exactly being measured? Does that include the time spent reviewing lengthy narratives, correcting inaccuracies and removing generic language? Or are we simply assuming that meaningful human review takes place because the process says it should?
My concern is that the polished, professional sounding language can create a false sense of confidence in what’s been recorded. When every interaction reads like a carefully structured therapeutic intervention, how critically are we actually examining whether it reflects what happened?
There’s a considerable difference between reviewing a record for accuracy and being impressed by how well it reads.
Because a PP cannot simply assume that a beautifully constructed account is an accurate one. We remain professionally accountable for what goes onto the case record and what others may subsequently rely upon.
There is also something rather troubling about the apparent need to turn every conversation into an elaborate account of meaningful intervention.
Sometimes supervision involves difficult conversations, frustration, uncertainty or very little apparent progress. Sometimes it involves listening, observing and recognising something that isn’t captured by the words spoken. Professional judgement includes knowing the significance of those things without dressing them up in therapeutic terminology.
An impressive narrative isn’t necessarily evidence of impressive practice.
But here’s the bigger question. If Justice Transcribe really does save hundreds of thousands of hours, what happens to those hours?
Do PPs actually regain time to build relationships, undertake meaningful supervision, develop professional judgement and respond to changing circumstances? Or do those theoretical savings simply become another justification for increasing workloads, reducing expectations of staffing capacity and demanding even more from an already overstretched workforce?
Given the current debate about probation being ‘not fit for purpose’, staffing shortages, pay and retention, I think that’s a rather important distinction.
Technology that genuinely gives PPs more time to do their jobs should be welcomed. Technology that merely enables the organisation to claim greater productivity while leaving the underlying pressures untouched is something else entirely.
Perhaps the real bingo winner isn’t ‘crisis management and stabilisation’ after all.
It’s ‘efficiency realised, capacity released, benefits delivered’.
But delivered to who?
House!