There’s a distinction that tends to get lost in the rush to deploy AI across law firms.
Access is not the same as capability.
A firm can roll out an AI platform to every lawyer and staff member in every office — and still find, months later, that usage is inconsistent, outputs are uneven, and people are either overconfident or quietly avoiding the tools altogether.
This is not a technology problem. It’s a development problem. And the data is proving it.
The Numbers Tell a Complicated Story
Individual AI use among legal professionals has more than doubled in a single year, with nearly 70% now reporting they use generative AI tools for work — up from 31% the prior year, according to the 8am 2026 Legal Industry Report. But here’s what that headline obscures: more than half of respondents report their firm has provided no training on the responsible use of AI and has no plans to do so.
Nearly seven in ten lawyers are using AI. More than half of their firms haven’t trained them on how. That gap is not limited to smaller practices — the Wolters Kluwer 2026 Future Ready Lawyer Survey, which focused on law firms and corporate legal departments globally, found that 39% of legal professionals identify inadequate training as a persistent barrier to AI success.
Thomson Reuters frames the strategic consequence clearly: organizations with a visible AI strategy are twice as likely to experience revenue growth from AI adoption, and 3.5 times more likely to experience its critical benefits — yet only 22% report having such a strategy.
The gap between adoption and capability is the defining challenge of legal AI right now. It doesn’t close on its own.
The Gap Nobody Talks About Enough
Think about what happens when a firm upgrades its matter management platform. The technology goes live. A handful of people dive in and find genuine efficiency gains. A larger group follows along at varying speeds. And a meaningful subset sticks to the old way — because the new system is unfamiliar and no one has given them a real path to change.
AI is experiencing the same dynamic, at much higher stakes.
As one legal industry observer has noted, firms will discover that two lawyers using the same AI tools can get radically different results — based entirely on how they frame the task. The lawyers most comfortable with technology are finding genuine productivity gains. Others are prompting in ways that produce mediocre results and concluding AI doesn’t work for their practice. And some aren’t meaningfully engaging at all. None of these outcomes serve the firm or the client.
What’s Actually Driving the Gap
A few patterns surface consistently:
Prompting is a skill, and most people haven’t been taught it. The difference between a lawyer who gets genuinely useful AI output and one who dismisses the tools as unreliable often comes down entirely to how they structure a request. That’s teachable — it just hasn’t been taught systematically.
Risk awareness is uneven. Some lawyers are cavalier about what they put into AI tools. Others are so cautious they’ve essentially opted out. What’s needed is calibrated judgment — and that requires training, not just a policy memo.
The supervisory layer is underprepared. Partners and senior associates are increasingly overseeing AI-assisted work product — their own and their teams’. How do you review an AI-generated draft? How do you supervise a junior associate’s AI-assisted research? How do you explain that process to a client? These are daily questions at most firms now, and most haven’t built consistent muscle around them. Thomson Reuters found that 46% of professionals report skills gaps within their organizations, with 31% specifically citing technology and data skills.
There’s Also a Professional Responsibility Dimension
ABA Formal Opinion 512, issued July 29, 2024, provided the ABA’s first formal ethics guidance focused on lawyers’ use of generative AI — making clear that competent representation now requires understanding the capabilities and limitations of any AI tool a lawyer chooses to use. Critically, that duty is ongoing: a one-time training does not satisfy it if the tool or the firm’s use of it has materially changed. Multiple states — including California, Florida, New Jersey, New York, and Pennsylvania — have also issued their own ethics opinions on AI use. Firms that haven’t built structured AI training aren’t just leaving capability on the table. They’re accumulating professional responsibility risk.
What This Means for Development Programs
The firms making the most progress have treated AI capability as a development priority, not a one-time event — and they’ve integrated it into their broader learning architecture rather than siloing it as a standalone initiative.
That means recognizing that a first-year associate using AI in their research has meaningfully different needs than a sixth-year drafting a brief, a paralegal managing documents, or a partner overseeing an AI-enabled team. It also means recognizing that AI fluency doesn’t exist in isolation from the rest of professional development. The associate who struggles to manage up will find it harder to communicate AI-assisted work to a supervising partner. The one who can’t read the room in a client meeting won’t know when AI-generated content needs a more human touch. Increasingly, the AI skills conversation and the professional skills conversation are the same conversation.
The question is no longer whether AI will become embedded in daily legal work — it already is. The question is whether development programs will evolve quickly enough to support that reality. Thomson Reuters reports that while only 15% of law firm respondents say AI is currently central to their workflow, 78% believe it will become central within the next five years. Firms that build AI capability deliberately and systematically — and connect it to the full arc of professional development — will be better positioned to serve clients, support their people, and compete as expectations on both fronts continue to rise.
The AI Fundamentals for Law Firms curriculum from SkillBurst by BARBRI — now spanning 33 modules across five integrated learning domains — was built specifically to address this shift: moving firms from fragmented AI usage to consistent, practical capability across roles, experience levels, and real-world workflows. Learn more at skillburst.com/professional-skills.
Interested in reading more? Fill out the form to read the full article.