{"id":145702,"date":"2026-03-10T07:12:00","date_gmt":"2026-03-10T15:12:00","guid":{"rendered":"https:\/\/xira.com\/p\/2026\/03\/10\/genai-a-slippery-slope-of-too-much-kool-aid\/"},"modified":"2026-03-10T07:12:00","modified_gmt":"2026-03-10T15:12:00","slug":"genai-a-slippery-slope-of-too-much-kool-aid","status":"publish","type":"post","link":"https:\/\/xira.com\/p\/2026\/03\/10\/genai-a-slippery-slope-of-too-much-kool-aid\/","title":{"rendered":"GenAI:\u00a0A\u00a0Slippery Slope Of Too Much Kool-Aid?"},"content":{"rendered":"

Legalweek\u00a02026<\/a>, the massive trade show more or less\u00a0focused on Biglaw, kicked off on Monday, March 9, in New York.\u00a0According\u00a0to the organizers, this year\u2019s show will attract some\u00a06,000 registered attendees and over 400 speakers. That\u2019s big by any standard.\u00a0<\/p>\n

The show kicked off with a series of workshops. One\u00a0workshop track was, not surprisingly,\u00a0entitled\u00a0the\u00a0\u201cAI workshop.\u201d The topics included how AI is shaping the profession, how to look at\u00a0AI\u00a0ROI, and\u00a0how\u00a0to lead and thrive in \u201cthe AI-transformed\u00a0Legal Workplace.\u201d The sessions were led primarily by\u00a0Jeff\u00a0Reihl<\/a>\u00a0of LexisNexis.<\/p>\n

It was the last of these\u00a0workshops, which dealt with human advancement\u00a0with\u00a0AI that offered some interesting insights into the thinking of\u00a0lawyers\u00a0and legal\u00a0professionals.\u00a0Particularly of those in-house\u00a0and consultants.<\/p>\n

T<\/strong>he\u00a0<\/strong>Legal GenAI\u00a0<\/strong>Cliche<\/strong><\/p>\n

The\u00a0central\u00a0idea of\u00a0the panel\u00a0played\u00a0off\u00a0the now\u00a0clich\u00e9\u00a0concept that AI\u00a0won\u2019t\u00a0replace\u00a0lawyers,\u00a0but it may replace lawyers that don\u2019t use\u00a0AI or use it well.\u00a0The idea seemed to revolve around the notion that\u00a0those who master AI and use\u00a0it\u00a0appropriately\u00a0will have more time for strategic and critical thinking. (Assuming they know how to do that.) As workforces tighten (since AI can do more), it will be these\u00a0folks\u00a0who will be left standing.<\/p>\n

Following\u00a0this logic, the notion also seemed to be that\u00a0the profession will need completely different\u00a0evaluation\u00a0processes and advancement\u00a0criteria\u00a0based in large part on how well\u00a0AI\u00a0is being\u00a0used.\u00a0As\u00a0one panelist put it,\u00a0in the future,\u00a0evaluation criteria will center around \u201cwho are the best people\u00a0using these skills.\u201d\u00a0<\/p>\n

In addition,\u00a0this resulting increased use of AI will mean a decline in\u00a0the\u00a0specialist in legal and more emphasis on the generalist who presumably can supplement their lack of training and experience by using AI. All this means of course an emphasis on AI\u00a0all the way\u00a0around.<\/p>\n

Too Much Kool-Aid?<\/strong><\/p>\n

I think there is a little too much AI\u00a0Kool-Aid\u00a0being consumed right now.\u00a0Too much GenAI writing.\u00a0You know what I mean:<\/p>\n

The one sentence paragraph.<\/p>\n

The room suddenly got quiet statement.<\/p>\n

The real question is not this but that.<\/p>\n

The dramatic two sentence conclusion.<\/p>\n

Etc.<\/p>\n

The problem is that in the process\u00a0of embracing GenAI as the be all and end all\u00a0we may be\u00a0ignoring\u00a0a few dangers.\u00a0For example,\u00a0I have noticed\u00a0of late\u00a0the\u00a0amount\u00a0of AI-written slop being produced\u00a0is increasing\u00a0every day.\u00a0<\/p>\n

The problem\u00a0is not only that\u00a0it\u2019s\u00a0clear that a human didn\u2019t take the time to write this slop or even edit\u00a0it\u00a0but it\u2019s also just not very good writing. \u00a0And\u00a0if we keep\u00a0emphasizing\u00a0AI skills over all else, this kind of writing and, for that matter, thinking, will become the norm.\u00a0So much so that what we now consider good writing,\u00a0and critical thinking, will no longer be the standard by which we\u00a0evaluate. And what is now good will no longer be considered good.<\/p>\n

We are already facing\u00a0increased\u00a0usage of AI because it\u2019s so easy. Why take the time to think\u00a0through\u00a0a problem when you can just ask ChatGPT to do the work for you? It\u2019s too tempting. I fear the more we emphasize the need for \u201cGenAI skills\u201d the more we will\u00a0also\u00a0encourage lazy thinking.<\/p>\n

It\u2019s\u00a0like the\u00a0addition of the word \u201cat\u201d at the end of a\u00a0sentence. Good\u00a0grammar\u00a0used to\u00a0demand\u00a0that you\u00a0never\u00a0end a\u00a0sentencewith a preposition.\u00a0It\u2019s\u00a0like\u00a0\u201cwhere\u2019s the coffee at\u201d\u00a0replaced\u00a0\u201cwhere\u2019s the coffee.\u201d<\/p>\n

So much so that good\u00a0grammar\u00a0now sounds, well, weird.\u00a0And\u00a0similarly\u00a0thinking that GenAI slop can\u00a0replace expertise is dangerous for another reason as well.<\/p>\n

The Rise of a\u00a0<\/strong>Mediocre<\/strong>\u00a0<\/strong>Generalist<\/strong>\u00a0at the Expense of Expertise<\/strong><\/p>\n

Moreover, the whole notion that we will no longer need the subject matter expert because AI can replace that expertise so that\u00a0it\u2019s\u00a0not needed ignores what makes\u00a0you an expert. I wasn\u2019t a good mass tort\u00a0defense\u00a0lawyer because I read about it on some GenAI output. I was good because I lived\u00a0through\u00a0several cases. I knew from\u00a0experience\u00a0what would happen. I saw patterns and\u00a0similarities\u00a0in how people would react. Do we really think that can be replaced with a prompt\u00a0that gets\u00a0a\u00a0GenAI answer\u00a0with some platitude?\u00a0The end result: mediocrity at the expense of real expertise.<\/p>\n

Something else: those who are advocating for an increased emphasis on greater AI skills and training and a\u00a0corresponding emphasis on\u00a0advancement based\u00a0on those\u00a0skills\u00a0are already good lawyers. They didn\u2019t become what they are by\u00a0relying on AI.\u00a0They have\u00a0critical\u00a0thinking and writing skills that were developed pre-AI.\u00a0So,\u00a0I fear they have forgotten how those skills were honed.<\/p>\n

In fact, we may be ignoring the rule of unintended consequences. Overemphasizing\u00a0GenAI skills and use risks dragging everyone down to the mediocre. And the\u00a0mediocre becomes more accepted\u00a0than the good. And the good is lost and replaced in the process. That\u2019s the risk.<\/p>\n

So What Can Be Done?<\/strong><\/p>\n

Make no mistake: I\u2019m not an anti-AI curmudgeon. But I am a realist and what I\u2019m seeing is a proliferation and acceptance of GenAI-generated stuff. That worries me since right now it\u2019s humorous but tomorrow it may be accepted.\u00a0<\/p>\n

Practicing law well is exacting and, frankly, hard. It means reading the cases, it means good, thoughtful writing and editing. It means using judgment honed over time. It takes time and energy. \u00a0Practicing mediocre law, on the other hand, is easy and sloppy.<\/p>\n

Instead of focusing on how to\u00a0train younger\u00a0lawyers\u00a0on how to use GenAI, maybe we\u00a0do something\u00a0different:\u00a0let\u2019s\u00a0first\u00a0define and then\u00a0emphasize\u00a0what and\u00a0how to be\u00a0a\u00a0good lawyer.\u00a0Skills\u00a0like problem solving. Understanding clients. Listening.\u00a0Resilience. Knowing the law. Thinking\u00a0critically.\u00a0We can\u2019t\u00a0expect good lawyering by\u00a0saying\u00a0go use AI and\u00a0then just rewarding\u00a0that use if we don\u2019t first define and develop those skills.\u00a0<\/p>\n

If we start there and then figure out how GenAI can enhance those skills, we may end up with lawyers\u00a0with the skill of\u00a0those who are now talking about what needs to change.<\/p>\n

To\u00a0quote ChatGPT:\u00a0\u201cPracticing law well is hard. Practicingmediocre law is easy. The real question for lawyers is which path they are willing to choose.\u201d<\/p>\n

Classic GenAI line.\u00a0Sigh.<\/p>\n

\n

Stephen Embry is a lawyer, speaker, blogger, and writer. He publishes\u00a0TechLaw Crossroads<\/a>, a blog devoted to the examination of the tension between technology, the law, and the practice of law<\/strong><\/em>.<\/p>\n

The post GenAI:\u00a0A\u00a0Slippery Slope Of Too Much Kool-Aid?<\/a> appeared first on Above the Law<\/a>.<\/p>\n

\"\"<\/figure>\n

Legalweek\u00a02026<\/a>, the massive trade show more or less\u00a0focused on Biglaw, kicked off on Monday, March 9, in New York.\u00a0According\u00a0to the organizers, this year\u2019s show will attract some\u00a06,000 registered attendees and over 400 speakers. That\u2019s big by any standard.\u00a0<\/p>\n

The show kicked off with a series of workshops. One\u00a0workshop track was, not surprisingly,\u00a0entitled\u00a0the\u00a0\u201cAI workshop.\u201d The topics included how AI is shaping the profession, how to look at\u00a0AI\u00a0ROI, and\u00a0how\u00a0to lead and thrive in \u201cthe AI-transformed\u00a0Legal Workplace.\u201d The sessions were led primarily by\u00a0Jeff\u00a0Reihl<\/a>\u00a0of LexisNexis.<\/p>\n

It was the last of these\u00a0workshops, which dealt with human advancement\u00a0with\u00a0AI that offered some interesting insights into the thinking of\u00a0lawyers\u00a0and legal\u00a0professionals.\u00a0Particularly of those in-house\u00a0and consultants.<\/p>\n

T<\/strong>he\u00a0<\/strong>Legal GenAI\u00a0<\/strong>Cliche<\/strong><\/p>\n

The\u00a0central\u00a0idea of\u00a0the panel\u00a0played\u00a0off\u00a0the now\u00a0clich\u00e9\u00a0concept that AI\u00a0won\u2019t\u00a0replace\u00a0lawyers,\u00a0but it may replace lawyers that don\u2019t use\u00a0AI or use it well.\u00a0The idea seemed to revolve around the notion that\u00a0those who master AI and use\u00a0it\u00a0appropriately\u00a0will have more time for strategic and critical thinking. (Assuming they know how to do that.) As workforces tighten (since AI can do more), it will be these\u00a0folks\u00a0who will be left standing.<\/p>\n

Following\u00a0this logic, the notion also seemed to be that\u00a0the profession will need completely different\u00a0evaluation\u00a0processes and advancement\u00a0criteria\u00a0based in large part on how well\u00a0AI\u00a0is being\u00a0used.\u00a0As\u00a0one panelist put it,\u00a0in the future,\u00a0evaluation criteria will center around \u201cwho are the best people\u00a0using these skills.\u201d\u00a0<\/p>\n

In addition,\u00a0this resulting increased use of AI will mean a decline in\u00a0the\u00a0specialist in legal and more emphasis on the generalist who presumably can supplement their lack of training and experience by using AI. All this means of course an emphasis on AI\u00a0all the way\u00a0around.<\/p>\n

Too Much Kool-Aid?<\/strong><\/p>\n

I think there is a little too much AI\u00a0Kool-Aid\u00a0being consumed right now.\u00a0Too much GenAI writing.\u00a0You know what I mean:<\/p>\n

The one sentence paragraph.<\/p>\n

The room suddenly got quiet statement.<\/p>\n

The real question is not this but that.<\/p>\n

The dramatic two sentence conclusion.<\/p>\n

Etc.<\/p>\n

The problem is that in the process\u00a0of embracing GenAI as the be all and end all\u00a0we may be\u00a0ignoring\u00a0a few dangers.\u00a0For example,\u00a0I have noticed\u00a0of late\u00a0the\u00a0amount\u00a0of AI-written slop being produced\u00a0is increasing\u00a0every day.\u00a0<\/p>\n

The problem\u00a0is not only that\u00a0it\u2019s\u00a0clear that a human didn\u2019t take the time to write this slop or even edit\u00a0it\u00a0but it\u2019s also just not very good writing. \u00a0And\u00a0if we keep\u00a0emphasizing\u00a0AI skills over all else, this kind of writing and, for that matter, thinking, will become the norm.\u00a0So much so that what we now consider good writing,\u00a0and critical thinking, will no longer be the standard by which we\u00a0evaluate. And what is now good will no longer be considered good.<\/p>\n

We are already facing\u00a0increased\u00a0usage of AI because it\u2019s so easy. Why take the time to think\u00a0through\u00a0a problem when you can just ask ChatGPT to do the work for you? It\u2019s too tempting. I fear the more we emphasize the need for \u201cGenAI skills\u201d the more we will\u00a0also\u00a0encourage lazy thinking.<\/p>\n

It\u2019s\u00a0like the\u00a0addition of the word \u201cat\u201d at the end of a\u00a0sentence. Good\u00a0grammar\u00a0used to\u00a0demand\u00a0that you\u00a0never\u00a0end a\u00a0sentencewith a preposition.\u00a0It\u2019s\u00a0like\u00a0\u201cwhere\u2019s the coffee at\u201d\u00a0replaced\u00a0\u201cwhere\u2019s the coffee.\u201d<\/p>\n

So much so that good\u00a0grammar\u00a0now sounds, well, weird.\u00a0And\u00a0similarly\u00a0thinking that GenAI slop can\u00a0replace expertise is dangerous for another reason as well.<\/p>\n

The Rise of a\u00a0<\/strong>Mediocre<\/strong>Generalist<\/strong>\u00a0at the Expense of Expertise<\/strong><\/p>\n

Moreover, the whole notion that we will no longer need the subject matter expert because AI can replace that expertise so that\u00a0it\u2019s\u00a0not needed ignores what makes\u00a0you an expert. I wasn\u2019t a good mass tort\u00a0defense\u00a0lawyer because I read about it on some GenAI output. I was good because I lived\u00a0through\u00a0several cases. I knew from\u00a0experience\u00a0what would happen. I saw patterns and\u00a0similarities\u00a0in how people would react. Do we really think that can be replaced with a prompt\u00a0that gets\u00a0a\u00a0GenAI answer\u00a0with some platitude?\u00a0The end result: mediocrity at the expense of real expertise.<\/p>\n

Something else: those who are advocating for an increased emphasis on greater AI skills and training and a\u00a0corresponding emphasis on\u00a0advancement based\u00a0on those\u00a0skills\u00a0are already good lawyers. They didn\u2019t become what they are by\u00a0relying on AI.\u00a0They have\u00a0critical\u00a0thinking and writing skills that were developed pre-AI.\u00a0So,\u00a0I fear they have forgotten how those skills were honed.<\/p>\n

In fact, we may be ignoring the rule of unintended consequences. Overemphasizing\u00a0GenAI skills and use risks dragging everyone down to the mediocre. And the\u00a0mediocre becomes more accepted\u00a0than the good. And the good is lost and replaced in the process. That\u2019s the risk.<\/p>\n

So What Can Be Done?<\/strong><\/p>\n

Make no mistake: I\u2019m not an anti-AI curmudgeon. But I am a realist and what I\u2019m seeing is a proliferation and acceptance of GenAI-generated stuff. That worries me since right now it\u2019s humorous but tomorrow it may be accepted.\u00a0<\/p>\n

Practicing law well is exacting and, frankly, hard. It means reading the cases, it means good, thoughtful writing and editing. It means using judgment honed over time. It takes time and energy. \u00a0Practicing mediocre law, on the other hand, is easy and sloppy.<\/p>\n

Instead of focusing on how to\u00a0train younger\u00a0lawyers\u00a0on how to use GenAI, maybe we\u00a0do something\u00a0different:\u00a0let\u2019s\u00a0first\u00a0define and then\u00a0emphasize\u00a0what and\u00a0how to be\u00a0a\u00a0good lawyer.\u00a0Skills\u00a0like problem solving. Understanding clients. Listening.\u00a0Resilience. Knowing the law. Thinking\u00a0critically.\u00a0We can\u2019t\u00a0expect good lawyering by\u00a0saying\u00a0go use AI and\u00a0then just rewarding\u00a0that use if we don\u2019t first define and develop those skills.\u00a0<\/p>\n

If we start there and then figure out how GenAI can enhance those skills, we may end up with lawyers\u00a0with the skill of\u00a0those who are now talking about what needs to change.<\/p>\n

To\u00a0quote ChatGPT:\u00a0\u201cPracticing law well is hard. Practicingmediocre law is easy. The real question for lawyers is which path they are willing to choose.\u201d<\/p>\n

Classic GenAI line.\u00a0Sigh.<\/p>\n

\n

Stephen Embry is a lawyer, speaker, blogger, and writer. He publishes\u00a0TechLaw Crossroads<\/a>, a blog devoted to the examination of the tension between technology, the law, and the practice of law<\/strong><\/em>.<\/p>\n","protected":false},"excerpt":{"rendered":"

Legalweek\u00a02026, the massive trade show more or less\u00a0focused on Biglaw, kicked off on Monday, March 9, in New York.\u00a0According\u00a0to the organizers, this year\u2019s show will attract some\u00a06,000 registered attendees and over 400 speakers. That\u2019s big by any standard.\u00a0 The show kicked off with a series of workshops. One\u00a0workshop track was, not surprisingly,\u00a0entitled\u00a0the\u00a0\u201cAI workshop.\u201d The topics […]<\/p>\n","protected":false},"author":3,"featured_media":145703,"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-145702","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\/03\/Filevine_Thumb_AI_guide_thumb1-ukerwP.webp?fit=950%2C1400&ssl=1","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/posts\/145702","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=145702"}],"version-history":[{"count":0,"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/posts\/145702\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/media\/145703"}],"wp:attachment":[{"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/media?parent=145702"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/categories?post=145702"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/xira.com\/p\/wp-json\/wp\/v2\/tags?post=145702"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}