{"id":160767,"date":"2026-08-17T14:25:59","date_gmt":"2026-08-17T22:25:59","guid":{"rendered":"https:\/\/xira.com\/p\/2026\/08\/17\/dharshi-harindra-is-right-ai-doesnt-replace-judgment-it-exposes-it\/"},"modified":"2026-08-17T14:25:59","modified_gmt":"2026-08-17T22:25:59","slug":"dharshi-harindra-is-right-ai-doesnt-replace-judgment-it-exposes-it","status":"publish","type":"post","link":"https:\/\/xira.com\/p\/2026\/08\/17\/dharshi-harindra-is-right-ai-doesnt-replace-judgment-it-exposes-it\/","title":{"rendered":"Dharshi Harindra Is Right: AI Doesn\u2019t Replace Judgment. It Exposes It."},"content":{"rendered":"
\"\"
<\/figcaption><\/figure>\n

If you\u2019ve spent any time on LinkedIn over the past year, you\u2019ve almost certainly seen the slogan: \u201cAI won\u2019t replace lawyers. Lawyers who use AI will.\u201d It\u2019s become the legal profession\u2019s favorite reassurance because it sounds optimistic, practical, and actionable.<\/p>\n

The message also gives lawyers something tangible to do. Learn prompting, experiment with new tools, become AI fluent, and you\u2019ll be ready for the future. It is comforting advice, but after my recent conversation with Dharshi Harindra, I don\u2019t think it captures the real transformation happening inside the legal profession.<\/p>\n

\n
\n<\/div>\n<\/figure>\n

AI Isn\u2019t The Story. Judgment Is.<\/strong><\/p>\n

Dharshi Harindra, assistant general counsel for APAC at FTI Consulting, challenged that familiar narrative in a way that has stayed with me ever since. She wasn\u2019t arguing that lawyers shouldn\u2019t embrace AI. She was arguing that we\u2019ve been asking the wrong question.<\/p>\n

\u201cThe phrase oversimplifies what\u2019s really happening,\u201d she told me. \u201cIt frames the future of law as primarily a tooling problem.\u201d That observation immediately shifted the conversation away from technology and toward something much more interesting: human judgment.<\/p>\n

One comment, in particular, stopped me in my tracks. \u201cAI is a mirror,\u201d Dharshi said. \u201cIt reflects back at us the existing judgment that we have.\u201d<\/p>\n

That sentence deserves a second look. We often assume AI makes lawyers smarter, but what it really does is make us faster. Speed without judgment is not a competitive advantage. It\u2019s simply a faster way to make bigger mistakes.<\/p>\n

Information Is Becoming Abundant. Judgment Is Becoming Scarce.<\/strong><\/p>\n

I\u2019ve spent the past several years building legal AI products, and one lesson keeps repeating itself. The difficult part has never been finding the law \u2014 machines are becoming remarkably good at that. The difficult part has always been deciding what to do with the law once you\u2019ve found it.<\/p>\n

That is judgment.<\/p>\n

Dharshi described judgment in a wonderfully practical way. She said it means deciding \u201cwhat matters, what doesn\u2019t matter, what happens next, and who owns the consequences.\u201d Notice what isn\u2019t part of that definition. It isn\u2019t legal research, drafting, or remembering every case. It\u2019s making decisions under uncertainty.<\/p>\n

That sounds much closer to the work in-house lawyers actually perform every day. Most legal departments aren\u2019t paid because they know where to find information. They\u2019re paid because businesses constantly face situations where several reasonable answers exist, none of them are perfect, and someone has to recommend a path the company can actually execute and live with.<\/p>\n

Maybe AI Didn\u2019t Create The Problem<\/strong><\/p>\n

One idea kept surfacing during our conversation. I found myself wondering whether AI has created far less disruption than many people think and has instead exposed something that has always been there.<\/p>\n

For decades, legal education rewarded finding the right answer. Law firms rewarded technical excellence, responsiveness, and precision. Judgment developed through experience, observation, mentorship, and thousands of decisions that were never written down in a playbook.<\/p>\n

We called that experience. We admired it. We rarely tried to explain how it actually worked.<\/p>\n

Then AI changed the economics of information almost overnight. Drafting, summarizing, researching, and comparing documents became dramatically faster. What remained stubbornly difficult was the one thing we had never really documented in the first place: good judgment.<\/p>\n

As Dharshi observed, perhaps the profession has rewarded being technically right more than being strategically wise. I think she\u2019s right. Technical expertise still matters, but technical expertise is becoming easier to augment. Judgment is becoming easier to see.<\/p>\n

The Hardest Part Of Legal AI Isn\u2019t AI<\/strong><\/p>\n

One of my favorite parts of the discussion had nothing to do with prompting or large language models. It was about what happens when you actually try to build systems that capture legal judgment.<\/p>\n

I shared a story from my own experience. I can create workflows that seem perfectly reasonable to me, only to hand them to a lawyer with 30 years of experience in a particular specialty who immediately points out that, while everything sounds plausible, nobody in that field would ever make decisions that way.<\/p>\n

That experience has repeated itself enough times that I no longer find it surprising. It reminds me that subject-matter expertise isn\u2019t simply knowing more law. It\u2019s seeing patterns, context, exceptions, and practical realities that never appear in statutes or casebooks.<\/p>\n

The best experts often struggle to explain how they think because so much of their reasoning has become instinctive. It\u2019s like asking someone how they recognize a friend\u2019s face or asking a bird how it flies. The answer skips over thousands of tiny decisions happening beneath the surface.<\/p>\n

Building useful legal AI forces us to surface those invisible decisions. That may be one of the most important things AI contributes to the legal profession because it pushes us to understand our own judgment more deeply than we ever have before.<\/p>\n

The Future Doesn\u2019t Require Every Lawyer To Become A Programmer<\/strong><\/p>\n

Dharshi also challenged another assumption I hear constantly. She argued that lawyers should spend less time worrying about becoming programmers and more time documenting how they think.<\/p>\n

I agree completely.<\/p>\n

The future does not require every lawyer to write code. It does require experienced lawyers to explain how they evaluate tradeoffs, recognize patterns, identify missing context, and decide which risks are acceptable. Those thought processes are valuable organizational assets, yet most of them still live inside individual lawyers\u2019 heads.<\/p>\n

That\u2019s a much bigger opportunity than learning the latest AI tool. If we can preserve and scale judgment, we create something that benefits every future lawyer, every business client, and every AI system we build.<\/p>\n

Judgment Is The Competitive Advantage<\/strong><\/p>\n

Another assumption deserves reconsideration. Many lawyers worry that AI levels the playing field because everyone has access to the same technology.<\/p>\n

I\u2019m not convinced.<\/p>\n

Everyone can purchase the same software. Not everyone asks the same questions. Not everyone notices the same risks, understands the same business context, or recognizes which recommendation a company can realistically execute. Those differences become even more valuable when everyone has access to similar technology.<\/p>\n

AI democratizes information. It does not democratize judgment.<\/p>\n

The Lawyer Worth Becoming<\/strong><\/p>\n

Near the end of our conversation, I asked Dharshi how she would define an exceptional lawyer in the AI era. Her answer had nothing to do with prompting techniques or coding skills.<\/p>\n

She described someone who can \u201cnavigate risk and technology and complexity without losing sight of the humans that are affected by their decisions.\u201d That feels like a much better definition of legal excellence than simply being the fastest person to produce an answer.<\/p>\n

As lawyers, we sometimes underestimate how unusual our training really is. We spend years learning to hold competing truths at the same time, navigate ambiguity, and advise people when certainty simply doesn\u2019t exist. Those capabilities are not becoming obsolete. They are becoming easier to recognize and more valuable than ever.<\/p>\n

Dharshi\u2019s insight changed how I think about the next chapter of our profession. AI doesn\u2019t replace judgment. It exposes it. Once you see that, it\u2019s hard to look at the future of law the same way again.<\/p>\n

\n

Olga V. Mack is the CEO of TermScout, where she builds legal systems that make contracts faster to understand, easier to operate, and more trustworthy in real business conditions. Her work focuses on how legal rules allocate power, manage risk, and shape decisions under uncertainty.<\/em><\/strong> A serial CEO and former General Counsel, Olga previously led a legal technology company through acquisition by LexisNexis. She teaches at Berkeley Law and is a Fellow at CodeX, the Stanford Center for Legal Informatics.<\/em><\/strong> She has authored several books on legal innovation and technology, delivered six TEDx talks, and her insights regularly appear in Forbes, Bloomberg Law, VentureBeat, TechCrunch, and Above the Law. Her work treats law as essential infrastructure, designed for how organizations actually operate.<\/em><\/strong><\/p>\n

\n

The post Dharshi Harindra Is Right: AI Doesn\u2019t Replace Judgment. It Exposes It.<\/a> appeared first on Above the Law<\/a>.<\/p>\n

\"\"
<\/figcaption><\/figure>\n

If you\u2019ve spent any time on LinkedIn over the past year, you\u2019ve almost certainly seen the slogan: \u201cAI won\u2019t replace lawyers. Lawyers who use AI will.\u201d It\u2019s become the legal profession\u2019s favorite reassurance because it sounds optimistic, practical, and actionable.<\/p>\n

The message also gives lawyers something tangible to do. Learn prompting, experiment with new tools, become AI fluent, and you\u2019ll be ready for the future. It is comforting advice, but after my recent conversation with Dharshi Harindra, I don\u2019t think it captures the real transformation happening inside the legal profession.<\/p>\n

\n
\n<\/div>\n<\/figure>\n

AI Isn\u2019t The Story. Judgment Is.<\/strong><\/p>\n

Dharshi Harindra, assistant general counsel for APAC at FTI Consulting, challenged that familiar narrative in a way that has stayed with me ever since. She wasn\u2019t arguing that lawyers shouldn\u2019t embrace AI. She was arguing that we\u2019ve been asking the wrong question.<\/p>\n

\u201cThe phrase oversimplifies what\u2019s really happening,\u201d she told me. \u201cIt frames the future of law as primarily a tooling problem.\u201d That observation immediately shifted the conversation away from technology and toward something much more interesting: human judgment.<\/p>\n

One comment, in particular, stopped me in my tracks. \u201cAI is a mirror,\u201d Dharshi said. \u201cIt reflects back at us the existing judgment that we have.\u201d<\/p>\n

That sentence deserves a second look. We often assume AI makes lawyers smarter, but what it really does is make us faster. Speed without judgment is not a competitive advantage. It\u2019s simply a faster way to make bigger mistakes.<\/p>\n

Information Is Becoming Abundant. Judgment Is Becoming Scarce.<\/strong><\/p>\n

I\u2019ve spent the past several years building legal AI products, and one lesson keeps repeating itself. The difficult part has never been finding the law \u2014 machines are becoming remarkably good at that. The difficult part has always been deciding what to do with the law once you\u2019ve found it.<\/p>\n

That is judgment.<\/p>\n

Dharshi described judgment in a wonderfully practical way. She said it means deciding \u201cwhat matters, what doesn\u2019t matter, what happens next, and who owns the consequences.\u201d Notice what isn\u2019t part of that definition. It isn\u2019t legal research, drafting, or remembering every case. It\u2019s making decisions under uncertainty.<\/p>\n

That sounds much closer to the work in-house lawyers actually perform every day. Most legal departments aren\u2019t paid because they know where to find information. They\u2019re paid because businesses constantly face situations where several reasonable answers exist, none of them are perfect, and someone has to recommend a path the company can actually execute and live with.<\/p>\n

Maybe AI Didn\u2019t Create The Problem<\/strong><\/p>\n

One idea kept surfacing during our conversation. I found myself wondering whether AI has created far less disruption than many people think and has instead exposed something that has always been there.<\/p>\n

For decades, legal education rewarded finding the right answer. Law firms rewarded technical excellence, responsiveness, and precision. Judgment developed through experience, observation, mentorship, and thousands of decisions that were never written down in a playbook.<\/p>\n

We called that experience. We admired it. We rarely tried to explain how it actually worked.<\/p>\n

Then AI changed the economics of information almost overnight. Drafting, summarizing, researching, and comparing documents became dramatically faster. What remained stubbornly difficult was the one thing we had never really documented in the first place: good judgment.<\/p>\n

As Dharshi observed, perhaps the profession has rewarded being technically right more than being strategically wise. I think she\u2019s right. Technical expertise still matters, but technical expertise is becoming easier to augment. Judgment is becoming easier to see.<\/p>\n

The Hardest Part Of Legal AI Isn\u2019t AI<\/strong><\/p>\n

One of my favorite parts of the discussion had nothing to do with prompting or large language models. It was about what happens when you actually try to build systems that capture legal judgment.<\/p>\n

I shared a story from my own experience. I can create workflows that seem perfectly reasonable to me, only to hand them to a lawyer with 30 years of experience in a particular specialty who immediately points out that, while everything sounds plausible, nobody in that field would ever make decisions that way.<\/p>\n

That experience has repeated itself enough times that I no longer find it surprising. It reminds me that subject-matter expertise isn\u2019t simply knowing more law. It\u2019s seeing patterns, context, exceptions, and practical realities that never appear in statutes or casebooks.<\/p>\n

The best experts often struggle to explain how they think because so much of their reasoning has become instinctive. It\u2019s like asking someone how they recognize a friend\u2019s face or asking a bird how it flies. The answer skips over thousands of tiny decisions happening beneath the surface.<\/p>\n

Building useful legal AI forces us to surface those invisible decisions. That may be one of the most important things AI contributes to the legal profession because it pushes us to understand our own judgment more deeply than we ever have before.<\/p>\n

The Future Doesn\u2019t Require Every Lawyer To Become A Programmer<\/strong><\/p>\n

Dharshi also challenged another assumption I hear constantly. She argued that lawyers should spend less time worrying about becoming programmers and more time documenting how they think.<\/p>\n

I agree completely.<\/p>\n

The future does not require every lawyer to write code. It does require experienced lawyers to explain how they evaluate tradeoffs, recognize patterns, identify missing context, and decide which risks are acceptable. Those thought processes are valuable organizational assets, yet most of them still live inside individual lawyers\u2019 heads.<\/p>\n

That\u2019s a much bigger opportunity than learning the latest AI tool. If we can preserve and scale judgment, we create something that benefits every future lawyer, every business client, and every AI system we build.<\/p>\n

Judgment Is The Competitive Advantage<\/strong><\/p>\n

Another assumption deserves reconsideration. Many lawyers worry that AI levels the playing field because everyone has access to the same technology.<\/p>\n

I\u2019m not convinced.<\/p>\n

Everyone can purchase the same software. Not everyone asks the same questions. Not everyone notices the same risks, understands the same business context, or recognizes which recommendation a company can realistically execute. Those differences become even more valuable when everyone has access to similar technology.<\/p>\n

AI democratizes information. It does not democratize judgment.<\/p>\n

The Lawyer Worth Becoming<\/strong><\/p>\n

Near the end of our conversation, I asked Dharshi how she would define an exceptional lawyer in the AI era. Her answer had nothing to do with prompting techniques or coding skills.<\/p>\n

She described someone who can \u201cnavigate risk and technology and complexity without losing sight of the humans that are affected by their decisions.\u201d That feels like a much better definition of legal excellence than simply being the fastest person to produce an answer.<\/p>\n

As lawyers, we sometimes underestimate how unusual our training really is. We spend years learning to hold competing truths at the same time, navigate ambiguity, and advise people when certainty simply doesn\u2019t exist. Those capabilities are not becoming obsolete. They are becoming easier to recognize and more valuable than ever.<\/p>\n

Dharshi\u2019s insight changed how I think about the next chapter of our profession. AI doesn\u2019t replace judgment. It exposes it. Once you see that, it\u2019s hard to look at the future of law the same way again.<\/p>\n

\n

Olga V. Mack is the CEO of TermScout, where she builds legal systems that make contracts faster to understand, easier to operate, and more trustworthy in real business conditions. Her work focuses on how legal rules allocate power, manage risk, and shape decisions under uncertainty.<\/em><\/strong> A serial CEO and former General Counsel, Olga previously led a legal technology company through acquisition by LexisNexis. She teaches at Berkeley Law and is a Fellow at CodeX, the Stanford Center for Legal Informatics.<\/em><\/strong> She has authored several books on legal innovation and technology, delivered six TEDx talks, and her insights regularly appear in Forbes, Bloomberg Law, VentureBeat, TechCrunch, and Above the Law. Her work treats law as essential infrastructure, designed for how organizations actually operate.<\/em><\/strong><\/p>\n

\n

The post Dharshi Harindra Is Right: AI Doesn\u2019t Replace Judgment. It Exposes It.<\/a> appeared first on Above the Law<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"

If you\u2019ve spent any time on LinkedIn over the past year, you\u2019ve almost certainly seen the slogan: \u201cAI won\u2019t replace lawyers. Lawyers who use AI will.\u201d It\u2019s become the legal profession\u2019s favorite reassurance because it sounds optimistic, practical, and actionable. The message also gives lawyers something tangible to do. Learn prompting, experiment with new tools, […]<\/p>\n","protected":false},"author":3,"featured_media":160768,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"_et_pb_use_builder":"","_et_pb_old_content":"","_et_gb_content_width":"","_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[16],"tags":[],"class_list":["post-160767","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-above_the_law"],"jetpack_featured_media_url":"https:\/\/i0.wp.com\/xira.com\/p\/wp-content\/uploads\/2026\/08\/GettyImages-1475813704-scaled-wFP8Uj.jpg?fit=2560%2C1707&ssl=1","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/posts\/160767","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/comments?post=160767"}],"version-history":[{"count":0,"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/posts\/160767\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/media\/160768"}],"wp:attachment":[{"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/media?parent=160767"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/categories?post=160767"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/tags?post=160767"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}