The Human Advantage: Professional Growth for Lawyers in the Age of Generative AI

I’m not what you’d call business-minded, so I was surprised when YouTube’s algorithm started serving me videos featuring marketing expert Rory Sutherland. Maybe it knew what I needed instead of what I wanted?

In one of the first videos I watched, Sutherland told a fascinating story. Restaurateur Will Guidara wanted to turn his New York restaurant into one of the best in the world. To do that, he took his team to dine at the top-ranked restaurant on the San Pellegrino World’s 50 Best list. The goal wasn’t to copy success. Instead, Guidara told his staff to look not at what the rival did brilliantly, but at what it missed. Two weaknesses stood out: the coffee was mediocre, and beer drinkers were treated like second-class citizens compared to the wine crowd.

Rather than trying to outshine the competition in its strongest areas, Guidara focused on those gaps. He appointed a single-origin coffee fanatic as the restaurant’s new coffee sommelier, tasked with creating a coffee experience equal to the wine service. Another staff member, a craft-beer enthusiast, became beer sommelier, curating a lineup worthy of pairing menus and tasting notes. Within months, guests could end their meal with a bespoke coffee or enjoy a carefully matched beer throughout dinner.

This approach, known as reverse benchmarking, means improving where others underperform instead of imitating what they already do best. And it worked: Guidara’s restaurant climbed from number fifty to number one in the world.

It’s a story that translates surprisingly well to law. In a profession where change often means “just replicate what top performers do,” we might learn more by looking in the opposite direction and finding the gaps others overlook and building our advantage there. Some people claim that  generative AI is the legal profession’s biggest competition. Maybe it is, maybe it isn’t. But it is definitely a competition and at the very least a new coworker or underling, so a shift in mindset might be exactly what young lawyers and those guiding their development need most.

What Generative AI Can Already Do in Law

Generative AI is no longer a novelty or a pilot project. It is rapidly becoming part of everyday legal workflows. We are seeing it in contract drafting, due diligence, discovery, legal research, and even client communication. With the right prompt, an AI assistant can draft a contract, summarize a thousand-page transcript, or locate precedent across jurisdictions.

Here is what it already handles surprisingly well:

• Drafting and editing: Generating solid first drafts of contracts, pleadings, or memos, and revising them for tone or clarity.
• Summarization and abstraction: Condensing lengthy opinions, diligence files, or transcripts into usable takeaways.
• Search and synthesis: Finding relevant clauses or cases through natural-language queries.
• Analysis and comparison: Flagging missing terms or risky language across large document sets.
• Automation: Routing workflows, creating checklists, and tracking task progress.

It is the kind of work lawyers used to cut their teeth on—the foundational “muscle memory” of practice. AI does it faster, more consistently, and without needing lunch breaks. That is both liberating and a little unsettling.

Still, “fast” does not always mean “right.” Even the best systems can misinterpret context, fabricate citations, or miss subtle distinctions that matter in legal reasoning. Every AI-generated output still requires careful review by a human lawyer, and that raises a new challenge. If AI assistance becomes the default starting point, how will junior lawyers develop the skills to review and correct that work confidently?

Traditionally, young lawyers learned by struggling through the first messy versions of document drafts and reviews themselves. If the machine now writes those first drafts or does the redlining, training programs will need to evolve so that new lawyers still gain the grounding to evaluate what is good, what is weak, and what is legally sound. Without that foundation, “review” risks becoming spell-checking rather than true legal supervision.

For junior lawyers especially, the question becomes: If AI can do the grunt work, where do I learn? The answer lies in what AI still cannot do—the parts of law that depend on judgment, empathy, and human accountability.

What Humans Still Do Better

If machines can draft and summarize, what is left for us? In a word: judgment.

Judgment is not about processing information. It is about context, ethics, and consequence. It is the act of deciding what matters, what is right, and what comes next.

In 1979, IBM issued an internal training manual that included a warning that still feels startlingly relevant today:

“A computer can never be held accountable; therefore, a computer must never make a management decision.”

That single line captures the enduring truth of digital transformation. Responsibility cannot be delegated to an algorithm or AI system. However capable a system may seem, it does not understand what it is doing or why it matters.

In law, accountability is not abstract—it is the profession’s core. A lawyer decides when to settle, what to advise, and how to protect a client’s interests. Those decisions live in the gray areas of trust, risk, and experience. AI can assist, but it cannot own those choices.

And then there is the human side of practice: persuasion, negotiation, advocacy, reassurance. A closing argument, a boardroom pitch, or a tough client call all hinge on empathy, timing, and emotional intelligence. Machines can mimic tone, but they cannot sense anxiety in a client’s voice or hesitation in a witness’s eyes.

That old IBM line doubles as a professional mantra: a lawyer’s value lies in being accountable. As AI absorbs more of the mechanical work, lawyers will define their worth not by the drafts they produce but by the decisions they stand behind.

From Knowledge Accumulation to Skill Differentiation

For generations, young lawyers were trained to master information: memorize doctrine, cite authority, produce flawless work. Progress was measured in output: the memo, the brief, the research note.

Generative AI flips that script. When a well-worded prompt can produce a decent first draft or a clean case summary, knowledge accumulation alone matters less. What matters more is discernment: knowing what to ask, what to trust, and what to change.

That is where reverse benchmarking comes in again. Instead of asking, “How can I do what AI does, only faster?” the better question is, “What can AI not do that I can learn to do better?”

The shift from knowledge accumulation to skill differentiation is the real story of professional growth in the AI era. It means investing in what cannot be automated: judgment, ethics, creativity, and human communication.

Building the Human Advantage: Skills for Lawyers in the AI Era

If you are building a career in law today, these are the areas where human capability still leads and where development should focus.

Critical Judgment and Verification
Generative AI is impressive, but it is not infallible. Lawyers must verify, validate, and challenge what they are given. Verification is the new research, or at least a new step in the research process. The lawyer who can tell when an AI-generated answer is wrong will always be more valuable than one who simply accepts it.

Prompt Engineering as Legal Reasoning
Asking good questions is at the heart of both law and prompting. The skill of crafting clear, contextual prompts mirrors issue-spotting and analysis. The more precise your question, the better your outcome, whether you are interrogating an AI or cross-examining a witness.
Some tools are moving toward making prompting less visible, but understanding what happens “under the hood” still makes you a stronger, more confident user.

Empathy and Communication
AI can draft an email, but it cannot read the emotional undertones of a client in crisis. Lawyers who can translate complexity into calm reassurance will continue to stand out. Emotional intelligence is becoming as valuable as intellectual rigor.

Strategic Framing
AI can generate answers, but it cannot tell you which questions matter. Lawyers who can identify a client’s real problem—the one beneath the surface—will always be indispensable. Framing is strategy, and strategy is human.

Ethical Literacy
Knowing how to use AI responsibly is the new frontier of professional ethics. Confidentiality, bias, intellectual property, and accuracy all intersect here. The next generation of lawyers will be defined not by how fluently they use AI, but by how wisely.

Collaborative Leadership
Legal teams are evolving into human–machine partnerships. Leading those teams will require empathy and accountability, knowing when to rely on AI and when to override it. And, as many of us have learned, thinking carefully about what we can delegate to AI as well as how to explain the issue and desired outcome clearly often makes us better managers of people too.

Continuous Learning
Technology evolves faster than bar exams. The most valuable professional habit is adaptability: curiosity about new tools, humility to learn, and discipline to keep your judgment sharper than your software. As in life, the journey often matters more than the destination.

Rethinking Mentorship and Training

For decades, junior lawyers learned through repetition: drafting, revising, and re-drafting under supervision. It was slow, sometimes painful, but it built intuition. Now that AI can generate a first draft in seconds, what happens to that learning curve?

The answer is not to resist technology, but to rethink mentorship. Senior lawyers need to shift from teaching form to teaching thought. Instead of redlining grammar, explain why a clause matters or why a certain approach fits the client’s goals. Context replaces correction. They also need to explain the relational side of practice: how to interact with clients or tribunals, manage teams, and handle difficult colleagues.

Mentorship does not have to flow in one direction. Reverse mentoring, where younger lawyers share insights about technology while learning judgment from experienced practitioners, creates a two-way exchange of knowledge. Both sides grow.

Technology itself can also play a new role. Generative AI opens the door to AI-assisted training and simulations. A well-designed AI mentor can help new lawyers practice client interactions, explore sample documents, or walk through procedural steps in a low-stakes environment before doing them for real. It can also act as a virtual coach, available around the clock to answer quick, practical questions such as how to phrase an email to a client, what “without prejudice” really means in context, or how to handle feedback on a draft.

Just as importantly, these same tools can help new lawyers build the knowledge and confidence needed to review and supervise AI-generated work. By experimenting with simulated drafting, critique exercises, and “explain your reasoning” prompts, young professionals can develop the analytical muscle memory they will need to assess accuracy, spot bias, and make judgment calls when the machine gets it wrong. Used thoughtfully, AI can make early learning both less intimidating and more rigorous.

Firms that embrace this approach will turn AI from a training obstacle into a learning accelerator, one that helps lawyers at every level develop faster, think deeper, and collaborate more confidently.

Benchmarking Ourselves by What Only We Can Do

Reverse benchmarking is not just a clever management trick; it is a mindset for professional growth. By studying what AI cannot do—care, reason, persuade, empathize—we rediscover what law truly is: a human profession built on trust and accountability.

Generative AI can produce language that looks like law, but it cannot shoulder the responsibility that comes with it. That remains ours.

As IBM warned in 1979, a computer can assist with decisions, but it cannot be held accountable for them. That is not just an ethical boundary; it is the heart of professionalism.

For young lawyers, the goal is not to out-code or out-compute the machine. It is to out-think it, to grow faster in the dimensions technology cannot touch: judgment, empathy, ethics, and leadership.

Reverse benchmarking reminds us that progress does not always mean chasing what is ahead. Sometimes it means turning around, seeing what is missing, and building something only a human could create.