Biglaw

Andrew Yang Says AI Is Replacing Biglaw Associates, Which Is Great News For Malpractice Lawyers

There's a difference between replacing jobs and reducing the number of jobs.

Andrew Yang hasn’t had a lot of success at the polls, but the former Davis Polk associate turned tech entrepreneur still enjoys his Yang Gang following on social media. Over the weekend, Yang offered a bleak assessment of the future of the profession based on a conversation with an unnamed Biglaw partner.

No, it’s not. No, it can’t. No, it’s not. And of the million reasons to second-guess that law school application, this is none of them.

To be a little more fair to the unduly credulous Yang, AI might be doing some work that used to be done by junior associates, but most of that sort of drudgery — the proverbial first-pass review in a Topeka warehouse — had already been outsourced one way or another years ago. The work that departed in favor of contract attorneys in the 2000s may now be in the hands of an algorithm, but that’s not what this Tweet is about.

First of all, AI doesn’t “generate a motion in an hour.” If you’re just looking for AI to do the job, then AI can produce a motion in minutes… as long as you don’t care how much the firm will get sanctioned. AI still screws up. Often. Most of the work involved in a draft motion isn’t generating the words, it’s making sure it’s the best words. If a motion generated by AI “in an hour” is better than what a junior can produce in a week — it won’t be — it probably says more about the partner’s warped recollection of their own genius.

Professional AI solutions absolutely speed up the workflow, perhaps helping that junior associate complete a non-embarrassing first draft in two days instead of a week, but that’s still a draft the junior associate prepares even if the AI sped up the process.

Which puts the lie to the claim “And the work is better.” The work is still a junior associate’s, pushed along by an algorithm spitting out filler text or summarizing the statement of facts or producing better research results teased out of the natural language prompt. If the partner really thinks the work is better, that’s not the AI’s doing. The AI is busy trying to shove Mack v. Armstrong into the brief even though it doesn’t exist.

So students shouldn’t let AI become the reason not to go to law school. Let the Trump administration making it prohibitively expensive do that work for you.

If there’s a lesson to extract from the last few AI hallucination scandals, it’s that the hallucination problem is almost always a partner’s fault. It’s the partners who have convinced themselves that the magic chatbot is replacing their costly associates before bonus season. The associates themselves seem to understand how to actually use AI while the partners keep trying to lawyer via AI about as well as they could explain Italian Brainrot memes. So it’s not surprising that a partner claims — smittenly — that some vague AI can “generate” work that’s “better” than the associates.

The tech bros of the world aren’t helping, perpetually overpromising on what this tool can accomplish in their mad pursuit of VC money. It makes me sound like I’m down on AI, which is not true at all: I think AI is a revolutionary technology that — assuming the bubble doesn’t burst — will accelerate the legal workflow. But it’s not about to replace associates because what we have right now is about as good as it’s going to get. At least for a long while. And then only if the bubble doesn’t burst.

But this is the critical distinction: AI does not in any way replace lawyer jobs.

Whalers were replaced by electric lightbulbs.1 Manual typesetters were replaced by digital printing. AI isn’t replacing associates, it’s a tool allowing them to work faster. In that sense, it’s not unlike the advent of online research: junior associates didn’t disappear because they didn’t spend half their day running back and forth to the library… they just did more research.

That might mean the industry has fewer openings for new lawyers. If junior lawyers do their jobs twice as fast, the firm needs half as many to do the the same work. While the distinction may not be much comfort to the law grad left outside looking in, it’s important because jobs lost to efficiency come back when there’s more work to be done. Indeed, the legal industry probably will get bigger. The world keeps growing in size and complexity and that means more legal work over the long haul and, by extension, more junior associate jobs for partners to systematically devalue by pretending AI is doing everything.

No one is picking Captain Ahab up off the unemployment line. I’d say he should learn to code, but AI actually might replace that.

[1] Or relocating to Carolina in 1997.


HeadshotJoe Patrice is a senior editor at Above the Law and co-host of Thinking Like A Lawyer. Feel free to email any tips, questions, or comments. Follow him on Twitter or Bluesky if you’re interested in law, politics, and a healthy dose of college sports news. Joe also serves as a Managing Director at RPN Executive Search.