Stewart McKelvey strategy head presses for better mentoring, law school ties and hands-on training
The automation of legal grunt work is creating a skills crisis that law firm leaders can no longer ignore, Paul Saunders says. Junior lawyers are skipping the tedious tasks that once taught them how the law works, even as firms rush to embed artificial intelligence into their practices.
Saunders, chief strategy and innovation officer and partner at Stewart McKelvey in Halifax, calls it the “AI training conundrum.” As he puts it, “as junior lawyers are increasingly using these technologies … they are not given the same opportunities to develop those foundational skills that they would have before.” He remembers reviewing leases for due diligence on a corporate transaction as a junior lawyer and learning how the contracts function by grinding through them line by line, an experience he now sees disappearing for the next generation.
Now, he says, “you can upload hundreds of leases into one of these AI platforms. It will automatically summarize those terms.” The junior’s role is to validate the output, not to build understanding from scratch, and the risk is clear: “If you’re not doing that lower level work, how do you then graduate to the higher level skills that you will need to be a partner and a senior lawyer?” he asks.
That problem is no longer confined to a handful of early adopters. At Stewart McKelvey, Saunders sees the same pattern across corporate, real estate, litigation, and labour and employment, where research memos that once required lawyers to pore over dozens of cases can now be drafted by AI, with juniors focusing on prompts and quality control rather than deep reading.
For Saunders, the solution starts with a hard reset on what firms are trying to build. He leans on a Wayne Gretzky line – “I skate to where the puck is going to be, not where the puck is” – and argues that firm leaders need to “articulate the vision of the AI-enabled lawyer of the future” and spell out the skills that lawyers will need to thrive, rather than hoping old pathways will somehow survive the shift.
Some of those competencies are timeless: “ethical service, understanding the risk, the needs of the client, deep understanding of the law,” he says. Others are bluntly new. Lawyers must know “how do you incorporate AI into your workflows? How do you validate the output of that content? How do you use the correct prompts to direct that AI tool to get the most correct answers?” They also need to understand pricing, profitability, and project management so they can decide when to automate, when to delegate, and when to push back on unworkable fee expectations, rather than treating AI as a black box that solves margin pressure.
Despite his warnings about AI’s impact, Saunders does not see the technology as the enemy of training; he sees it as a teacher. Tools like ChatGPT, Claude or Gemini “are not the kind of tool you would want to generate a brief for the court or a contract, because they hallucinate, they’re biased,” but he argues they can useful tools to teach and to learn, especially now that lawyers can have a conversation with AI to unpack unfamiliar concepts without waiting for a partner’s time.
He pushes both students and practising lawyers to use these systems constantly rather than treating them as occasional shortcuts. Speaking recently to a Dalhousie University law class, he was struck that only a handful of students knew what “vibe coding” was, and when he asked how many use AI tools every week, his message back to them was blunt: “One hundred percent of you should be using these tools every week. Try to use it every single day,” he says. In his view, “AI will not replace lawyers, but lawyers that use AI will replace those that don’t,” a line he delivers as a warning rather than hype.
That expectation now runs through Stewart McKelvey’s hiring pipeline. When the firm interviews students, Saunders asks, “Do you use AI? Do you use it in your work? Do you use it on your own personal time? How do you use it? What do you like about it? What are its limitations?” he asks them. He also checks whether they understand basic terminology, from large language models to agentic AI, even if they cannot yet code, because a junior who has never engaged with those concepts is already behind the curve.
But individual comfort with technology is only one piece of the puzzle. Saunders insists that compensation and culture must reward coaching if firms want seniors to invest time in junior development. Mentoring, he says, cannot be left to the “natural interactions that folks might have”; firms need structured systems so that lawyers who take juniors under their wing see that work recognized and, in some cases, compensated rather than treated as invisible labour.
Old-school apprenticeship sits at the heart of his model. Firms, he says, need to “double down on traditional mentorship,” reconnecting senior lawyers who “learned the hard way” with juniors who risk missing that foundational experience, and that means bringing junior lawyers to client meetings and boardrooms and having regular conversations so skills and judgment transfer in real time, not just through redlined documents arriving by email.
Saunders wants law schools pulled into this redesign as well, not left to run a parallel, theory-heavy track. He urges faculties to bring in law firm leaders as guest speakers and “have intense curriculum and courses over a couple of days to develop those skills and make law school practical and incorporate those tools into the learning process.” His firm expects lawyers to use AI every day, and he argues education needs to reflect that reality so graduates can “hit the ground running when you get into that environment, not start from scratch,” he says.
Underpinning all of this is classic change management rather than blind faith in technology. Resistance to upheaval is natural, he says, so leaders need open lines of communication and must ensure partners “don’t feel like it’s being forced on them, but they’re owning it, that they have agency, that they’ve been consulted,” with senior figures modelling the behaviour they expect from colleagues instead of delegating the hard parts of change downward.
Looking ahead, Saunders thinks the only safe bet is constant movement. Technology is evolving at a pace “I don’t think we’ve ever seen before in the practice of law,” and he believes this will be “one of the most disruptive periods we’ve seen” in the profession, a period that will expose any firm that fails to rebuild training around AI-enabled lawyers rather than the comfort of legacy workflows.
This article is based on an episode of CL Talk, which can also be found here:
The episode can also be found on our CL Talk podcast homepage, which includes links to follow CL Talk on all the major podcast providers.
