- Mark, who leads a team of product lawyers for Anthropic's enterprise offerings, and Anna, global co-head of AI at Freshfields, discuss how lawyers actually work with Claude.
- They argue the real power comes not from chat-based Q&A but from leveraging Claude's ability to use a local computer to produce the documents knowledge workers actually use — Word, Excel, PowerPoint.
- Despite AI's strength at synthesizing large bodies of information and spotting themes, human judgment, junior associates' instinct for process, and a firm's role in navigating risk remain central.
Working Like a Lawyer with Claude
- The trickiest area for legal teams is the "context gap" — understanding what the product spec looks like today versus how it may differ tomorrow as engineering teams move at the speed of light.
- Most lawyers start with chat-based question-and-answer experiences, but the more powerful unlock is leveraging code — Claude using your local computer to create the files knowledge workers rely on.
- Claude excels at making sense of large bodies of information and spotting themes, so uploading memos, docs, product roadmaps, and new legislation into one place lets you converse with Claude to create a new artifact synthesized across all of them.
- Freshfields, one of the top global law firms, has roughly 5,700 staff and lawyers worldwide using AI, framing the core question as how to add value and be most fluent in clients' needs.
- The value question is about delivering advice in the most seamless, contextual way possible, and bringing everything the firm separately does with AI to the center of the conversation.
- Junior associates often have an inherent understanding of both the benefits and limitations of AI and the right process for checking and staying in the loop.
- A law firm's role is described as helping clients sail a ship through rocky straits — judgment is the thing that keeps the rudder straight amid risk.
- Anthropic's value of "holding light and shade" captures the duality: significant risk in the technology that lawyers help clients navigate, alongside a lot of promise.
Product lawyer — A lawyer who reviews and advises on new features and products as they are developed.
Context gap — The challenge of legal advice keeping pace with product specs that change rapidly as engineering moves fast.
AI product counseling — Advising on the legal aspects of building and deploying AI products.
Artifact — A new document or output Claude creates by synthesizing across multiple source materials.
Knowledge worker — A professional whose work centers on producing and reasoning over documents and information.
Human in the loop — Keeping people involved to check, verify, and apply judgment to AI output.
Holding light and shade — An Anthropic value capturing both the risk and the promise of the technology.
I'm excited to chat with you today, Anna. I know, me too, Mark. I help lead a team of product lawyers who focus on Anthropic enterprise product offerings. That includes the platform, things like Claude Code and other enterprise verticals. And so we're often reviewing new features and products that are coming out from Anthropic, and sometimes I need advice and counsel from trusted experts. And so I call Anna.
I am the global co-head of AI at Freshfields, which means I help head up our AI practice globally. And my own personal practice, I do a lot of what you might call AI product counseling. We have, I think, 5,700 staff and lawyers worldwide who are using AI. And usually one of the stickiest and trickiest areas for us to navigate is the context gap — understanding what does the spec look like today, and the fact that it might look different tomorrow, and your engineering teams are moving at the speed of light.
A lot of lawyers, when they first start using Claude, are just using it for these sort of chat-based question-and-answer experiences. But the more powerful stuff to be unlocked is by leveraging code, which is the ability for Claude to use your local computer to create the types of files that you would typically be using as a knowledge worker. So that could be a Microsoft Word document or an Excel spreadsheet or even a PowerPoint.
If there's one thing that I feel like AI is very good at, it's making sense of large bodies of information and spotting themes across them. And so being able to upload a lot of existing memos and docs and product roadmaps and new legislation that's coming out and just putting that all in one place and then being able to have a conversation with Claude to create that new artifact that's based on all the things it's seeing across those different surfaces.
Everyone's asking this question: how does AI make us different in the future? What does the future bring to knowledge work or otherwise? And from my perspective, one of the core questions in that is how do we add value if we are one of the top global law firms? What does it mean to be the most fluent in our clients' enterprises and their kinds of hopes and dreams? How do we actually deliver on advice in the most seamless, contextual way possible? We as an institution are doing all these great things. But what if we took everything we were separately doing with AI and then brought it to the center of the conversation?
I really appreciate sometimes our junior associates. Sometimes they have such an inherent understanding of the benefits, but also the limitations and the right process for checking and being in the loop. The role that we play, ideally for our clients, is to help them sail a ship. Sometimes they're sailing through a really rocky strait and there might be rocks to the left and rocks to the right. And they're being slammed by hail and all of these things. But at the end of the day, our judgment is the thing that keeps the rudder straight.
At Anthropic, we have this saying of holding light and shade. And that's one of our values. And for me, that can represent that. I feel like there's so much risk in this technology, and that as lawyers, we often help our clients navigate that risk. But there's also a lot of promise.
TL;DR
- Mark, who leads a team of product lawyers for Anthropic's enterprise offerings, and Anna, global co-head of AI at Freshfields, discuss how lawyers actually work with Claude.
- They argue the real power comes not from chat-based Q&A but from leveraging Claude's ability to use a local computer to produce the documents knowledge workers actually use — Word, Excel, PowerPoint.
- Despite AI's strength at synthesizing large bodies of information and spotting themes, human judgment, junior associates' instinct for process, and a firm's role in navigating risk remain central.
Takeaways
- The trickiest area for legal teams is the "context gap" — understanding what the product spec looks like today versus how it may differ tomorrow as engineering teams move at the speed of light.
- Most lawyers start with chat-based question-and-answer experiences, but the more powerful unlock is leveraging code — Claude using your local computer to create the files knowledge workers rely on.
- Claude excels at making sense of large bodies of information and spotting themes, so uploading memos, docs, product roadmaps, and new legislation into one place lets you converse with Claude to create a new artifact synthesized across all of them.
- Freshfields, one of the top global law firms, has roughly 5,700 staff and lawyers worldwide using AI, framing the core question as how to add value and be most fluent in clients' needs.
- The value question is about delivering advice in the most seamless, contextual way possible, and bringing everything the firm separately does with AI to the center of the conversation.
- Junior associates often have an inherent understanding of both the benefits and limitations of AI and the right process for checking and staying in the loop.
- A law firm's role is described as helping clients sail a ship through rocky straits — judgment is the thing that keeps the rudder straight amid risk.
- Anthropic's value of "holding light and shade" captures the duality: significant risk in the technology that lawyers help clients navigate, alongside a lot of promise.
Vocabulary
Product lawyer — A lawyer who reviews and advises on new features and products as they are developed.
Context gap — The challenge of legal advice keeping pace with product specs that change rapidly as engineering moves fast.
AI product counseling — Advising on the legal aspects of building and deploying AI products.
Artifact — A new document or output Claude creates by synthesizing across multiple source materials.
Knowledge worker — A professional whose work centers on producing and reasoning over documents and information.
Human in the loop — Keeping people involved to check, verify, and apply judgment to AI output.
Holding light and shade — An Anthropic value capturing both the risk and the promise of the technology.
Transcript
I'm excited to chat with you today, Anna. I know, me too, Mark. I help lead a team of product lawyers who focus on Anthropic enterprise product offerings. That includes the platform, things like Claude Code and other enterprise verticals. And so we're often reviewing new features and products that are coming out from Anthropic, and sometimes I need advice and counsel from trusted experts. And so I call Anna.
I am the global co-head of AI at Freshfields, which means I help head up our AI practice globally. And my own personal practice, I do a lot of what you might call AI product counseling. We have, I think, 5,700 staff and lawyers worldwide who are using AI. And usually one of the stickiest and trickiest areas for us to navigate is the context gap — understanding what does the spec look like today, and the fact that it might look different tomorrow, and your engineering teams are moving at the speed of light.
A lot of lawyers, when they first start using Claude, are just using it for these sort of chat-based question-and-answer experiences. But the more powerful stuff to be unlocked is by leveraging code, which is the ability for Claude to use your local computer to create the types of files that you would typically be using as a knowledge worker. So that could be a Microsoft Word document or an Excel spreadsheet or even a PowerPoint.
If there's one thing that I feel like AI is very good at, it's making sense of large bodies of information and spotting themes across them. And so being able to upload a lot of existing memos and docs and product roadmaps and new legislation that's coming out and just putting that all in one place and then being able to have a conversation with Claude to create that new artifact that's based on all the things it's seeing across those different surfaces.
Everyone's asking this question: how does AI make us different in the future? What does the future bring to knowledge work or otherwise? And from my perspective, one of the core questions in that is how do we add value if we are one of the top global law firms? What does it mean to be the most fluent in our clients' enterprises and their kinds of hopes and dreams? How do we actually deliver on advice in the most seamless, contextual way possible? We as an institution are doing all these great things. But what if we took everything we were separately doing with AI and then brought it to the center of the conversation?
I really appreciate sometimes our junior associates. Sometimes they have such an inherent understanding of the benefits, but also the limitations and the right process for checking and being in the loop. The role that we play, ideally for our clients, is to help them sail a ship. Sometimes they're sailing through a really rocky strait and there might be rocks to the left and rocks to the right. And they're being slammed by hail and all of these things. But at the end of the day, our judgment is the thing that keeps the rudder straight.
At Anthropic, we have this saying of holding light and shade. And that's one of our values. And for me, that can represent that. I feel like there's so much risk in this technology, and that as lawyers, we often help our clients navigate that risk. But there's also a lot of promise.