Judge Legora’s UK training by supervised legal work

Legora has expanded its legal AI training programme to UK law schools and signed its first university partners, Non-Billable reports on September 29. The announcement places Legora’s training effort inside legal education rather than limiting it to software deployment in firms. It does not identify the university partners, describe the curriculum, state how students are assessed or report any result from the programme. The event is therefore a launch of a training programme, not a demonstrated change in how quickly lawyers become capable.

That distinction matters because the durable advantage belongs to firms that use AI to develop capable lawyers faster, not merely to firms that hire fewer beginners. Legora’s move points in the right direction: it treats legal AI as something lawyers must learn to use, rather than only as a mechanism for removing work from junior roles. But the announcement does not establish that the programme develops judgment, improves legal work or shortens the path to independent practice. LawNext podcast reports that Latitude Legal has seen a 52 percent jump in new law firm and corporate clients in the first half of the year, while describing growing demand for lawyers with real judgment to direct AI, check it and own the work. That supports the importance of the capability Legora would need to develop, not the claim that Legora has developed it.

The strongest objection is that AI will eliminate associate training rather than improve it. Jordan Furlong gives the answer that law firms will not train associates after AI and will instead train partners. That is a clean challenge to the premise that training can preserve a development pipeline: if junior lawyers lose the work through which they acquire judgment, a course in AI use cannot by itself replace the missing practice. Legora’s programme may answer that objection only if it connects instruction to supervised legal work and produces lawyers who can exercise judgment, not simply users who know where to put a prompt. Nothing reported about the UK expansion shows that connection.

The announcement therefore proves one narrower proposition: Legora is expanding legal AI training to UK law schools through university partnerships. It does not prove the stronger proposition that the programme develops capable lawyers faster. A company’s training launch is an input. The relevant output is better legal performance produced in less time, with the supervision and responsibility required for real work. Thomson Reuters Institute describes the need for law firms to redesign new attorney onboarding for the AI era, but that headline supplies no result from any redesigned system. The same standard applies here. A training programme can be useful, widely adopted or well received and still fail to accelerate lawyer development.

The next confirming fact would be a published evaluation from Legora or one of its university partners that compares trained students or junior lawyers with an appropriate baseline on legal judgment, work quality, correction rates and time to complete defined tasks. The programme would need to show that participants can direct, check and own AI-assisted work, rather than merely complete training. A result of that kind would turn the UK expansion into an example of firms and schools building lawyers faster. If the programme reports only enrolment, course completion or tool usage, it will remain evidence of adoption. If it shows that associates are no longer being trained and partners absorb the work instead, Jordan Furlong’s objection will stand. We score the launch as promising direction, not proof of the advantage.