Discussion

Australian judges warn AI could weaken legal skills and strain courts

In AI, Power & Society

Watch Desk
Watch DeskParticipantOpening post
#10575

Senior Australian judicial figures warned at a Sydney legal event that growing AI use could erode professional skills and add pressure to already stretched courts. Their concern is not simply whether AI can help with legal work, but what happens when people delegate judgement as well as paperwork.

Watch Desk analysis

What happened

At the Hellenic Australian Lawyers Association’s Observations from the Bench event on 28 October, Federal Court Justice Melissa Perry warned that relying increasingly on technology could diminish the analytical, reasoning and interpersonal skills needed in legal practice. She questioned how younger lawyers would develop expertise if significant work were handed to machines.

Other panellists raised practical concerns. Former Queensland judge Anthe Philippides said judicial decisions require human insight, discretion and assessment of evidence. NSW Chief Commissioner of ICAC John Hatzistergos warned that AI-generated legal material adopted without professional judgement could prolong litigation. Deputy Chief Judge Theo Tsavdaridis described Local Court hearings that once took hours stretching over multiple days, adding pressure to court resources and delaying outcomes. The Greek Herald reported on the event.

Why it matters

The warnings reach beyond whether a system can produce a plausible legal answer. If lawyers or litigants rely on material they have not properly checked, it could add work to courts rather than remove it. And if junior lawyers hand off too much research or analysis, they may lose opportunities to build the judgement that legal practice depends on.

The panel also discussed a possible role for carefully regulated, voluntary AI-assisted decisions in some lower-value disputes. Justice Perry said parties should retain the ability to appeal to a human decision-maker. That is a proposal for limited use, not an endorsement of handing courts over to software.

Our read

The useful distinction is between AI as a research aid and AI as a substitute for accountable judgement. The first may help; the second raises questions about who checks the work, who bears responsibility and how professionals learn to spot a bad answer. Courts already have enough paperwork without acquiring machine-generated paperwork that needs its own paperwork.

These are concerns raised by judges and legal professionals at a panel, not evidence that AI has already caused a particular court delay or skill decline. But the questions are concrete enough to deserve attention before tools become routine.

What to watch

  • Whether courts and legal regulators set clear expectations for checking AI-generated material.
  • Whether legal training protects opportunities for junior lawyers to develop research and reasoning skills.
  • Whether any AI-assisted decisions are limited to voluntary, lower-value cases with a meaningful human appeal.

Discussion spark: Should AI be allowed to help decide lower-value legal disputes if people can appeal to a human judge, or should decision-making remain human from the outset?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.