A sitting General Counsel told me something recently that each of you should be thinking about.

He said that when he looks at outside counsel invoices now, he has started asking himself a new question for each line item: could I have done this myself with an AI tool for next to nothing?

He is not asking what the law firm could have done with AI. He is asking what his team could do before sending work to outside counsel.

And then he went and built the tools to act on it.

Kevin Keller just became the first General Counsel at Neurophos, an AI chip company, after serving as GC at Forward Networks. He spent eleven years as product counsel at Amazon, Facebook, and Instacart before that. Last week he told me why he built an open source legal AI platform that is free, self-hosted, available to any in-house team that wants it. He designed this specifically to help legal departments do more of the substantive work themselves before it ever reaches outside counsel.

Not to replace outside counsel. To arrive better prepared, better scoped, and with a clearer sense of what they actually need a lawyer to do versus what AI already handled.

That is a different kind of client than most law firms are used to dealing with. And more of them are coming.

Kevin also told me about an experience earlier in his career. He scoped a project for outside counsel when we was at Amazon. His estimate was $25,000. The bill came back for $250,000. He nearly lost his job. He eventually learned to be more precise by spelling out exactly what he knew, what he needed, what the expected hours were, and what the fee ceiling should be before he handed anything over.

He got better at managing the gap between what he could see and what he was paying for. AI is now making that gap both more visible and more urgent for everyone.

Early returns from a survey I am about to ask you to take, show that neither firms nor clients can explain when and where AI is being used to produce legal work. And nobody seems to be able to explain how this is impacting fees.

These gaps, between what is on the invoice and what AI contributed to the work underneath it, is exactly what I designed this survey to measure.

A note before you click: the survey is designed for three specific roles: in-house legal leaders, law firm leadership and pricing directors, and law firm practitioners. If you sit outside those categories the survey will thank you and exit you early. That is by design, not a slight. The data only holds if it comes from the right respondents.

TAKE THE SURVEY - FIVE MINUTES

Whether or not the survey is right for you, I would ask one thing: send this to one person in your network who sits in one of those three roles. The second briefing gets better with every qualified response. A single forward from you could be the one that tips a finding from interesting to conclusive.

Here is what the early data is already showing.

Every GC in the survey who asked outside counsel to disclose AI tool usage on a matter got a response they described as vague or incomplete. Not one exception. And when asked what percentage of their outside counsel invoices reflect AI-compressed work without a corresponding fee reduction, the single most common answer was: I have no way to estimate this.

That is Kevin's question, confirmed by data. GCs are starting to ask it. They cannot answer it yet. Neither can their outside counsel.

The second briefing will put a number on this. But only if enough practitioners complete the survey to make the data meaningful.

The survey takes five minutes. It routes you to the right questions based on where you sit: in-house, firm leadership, or practitioner. Everything is anonymized. No organization is identified in the published findings without permission.

Everyone who completes it receives early access to the second briefing before it is published anywhere, and a first invitation to the Brainyacts breakfast series launching this summer in select cities.

The full conversation with Kevin, and with Rob Saccone, who joined us to talk about what all of this looks like from inside law firm infrastructure, is coming soon. Survey respondents hear about it first.

If you have not yet read the briefing that started this conversation, Money Spent, Earned, and Prompted: The New Economics of Legal Work in the AI Era - you can download it free here. It is the foundation for this whole series

Josh.

To read previous editions, click here.

Was this newsletter useful? Help me to improve!

With your feedback, I can improve the letter. Click on a link to vote:

Login or Subscribe to participate

Who is the author, Josh Kubicki?

Josh Kubicki teaches AI and the business of law at Indiana University Maurer School of Law and has trained over 3,000 lawyers on generative AI. He is the author of Brainyacts, read by nearly 10,000 legal professionals worldwide.

AI training, courses, and resources: kubicki.ai

Strategic advisory for firm leadership: joshkubicki.com

DISCLAIMER: None of this is legal advice. This newsletter is strictly educational and is not legal advice or a solicitation to buy or sell any assets or to make any legal decisions. Please /be careful and do your own research.

Keep Reading