Future Ready Lawyer

The episode contains several genuinely non-obvious insights - verification drift as distinct from AI illiteracy, the paradox that decreasing hallucination rates increase over-reliance risk, and the averaging/mediocrity effect on professional knowledge workers - but these are interspersed with considerable consensus-building filler and repetitive agreement between panelists.

“verification drift is about people know they are supposed to verify but due to various reasons they don't”

“as hallucination uh, rate reduces the problem of verification drift increases significantly because it becomes so trustworthy”

The counterintuitive framing that hallucinations are currently a useful warning signal - and that a post-hallucination world could be more dangerous - is genuinely fresh thinking; so is the empirical finding that 80%+ of AI misuse comes from self-represented litigants rather than lawyers, inverting media narratives. The rest of the discussion rehashes widely-circulated concerns about AI limitations.

“maybe this phase is quite good because we actually see the AI use because of the hallucinations and maybe it'll be worse when we don't see the AI use”

“Most of the media attention and commentary focuses on lawyers using AI in poor ways and that's obviously it cuts across lawyers role as officers of the court”

Vicki McNamara is a genuine practitioner-researcher with a real empirical dataset (470+ cases, multiple jurisdictions) rather than a thought-leader, and her 30+ years of legal knowledge management experience adds credibility; however, she is a senior research associate rather than a senior practitioner or judge who has operated at scale, and the multi-host panel dilutes focus.

“cases that we've identified in our research, which is focused on common law and hybrid jurisdictions, are currently totalling over 470 examples”

“in Australia it's more than 80% of the case examples in our set are self represented litigants”

The episode is anchored by real numbers (90 cases at year-end 2024 vs. 470+ by mid-2025), named cases (Mata v. Avianca, Mayan Kostros in NSW Court of Appeal), named judges (Chief Justice Bell, Justice Payne), and specific resources (AUSTLI, JADE, Fair Work Commission website, state library portals); it loses points for lacking dollar figures, methodology detail on the dataset, or any controlled comparison data.

“at the end of last year when I did some presentations, I had a data set of around 90 cases”

“Mayan Kostros which recently came through the Court of Appeal in New South Wales where both the uh, Chief Justice Bell and Justice Payne”

The host sequences questions logically and draws out the dataset figures, and Speaker A (Mark) contributes the episode's sharpest hypothesis about post-hallucination AI being more dangerous; however, there is virtually no pushback on any claim, Speaker D (Alex) largely restates what others have said, and the multi-panellist format collapses into extended agreement loops rather than productive tension.

“maybe this phase is quite good because we actually see the AI use because of the hallucinations and maybe it'll be worse when we don't see the AI use”

“Do you have any thoughts on how AI might be used to better enable self represented litigants?”