AI Agents x Law: Hands-On Workshop and Live Participatory Demo
A hands-on Stanford workshop giving participants practical experience
using AI agents for legal and professional work—through guided demonstrations,
exercises, discussion, and a live participatory multi-agent collaboration demo.
Location: Room 270, Stanford Law SchoolCapacity: ~60 participantsAdmission: In-person, invitation onlyRecording: Edited version released free after event
Request an Invitation
Complete the Stanford Law registration form to request an invitation. This is required for all attendees.
Accepted participants will have access to pre-reading, setup walkthroughs, and live teach-in sessions to help get a supported AI agent running before the workshop.
Report-outs and group reflections from the collaborative AI agent session
4:00 PM
Damien Riehl
The Agentic Spectrum: Shipping Outcomes, Not Architectures
4:15 PM
Bryan Wilson
Computational Law Report announcement
4:20 PM
Matt Pollins
The Rise of Vibe Coding and AI-Native Law Firms (Prerecorded)
4:25 PM
Dazza Greenwood
Closing reflections, next steps, and community resources
4:30 PM
END
Workshop concludes
What to Expect
The workshop has two parts, moving from learning to doing.
Attendees will leave with hands-on experience using real agent tools
and a practical understanding of agentic workflows for professional work.
The session will also explore how a new generation of AI Native lawyers
and legal organizations is emerging across in-house teams, law firms,
and other legal-service contexts.
Part 1: Agent Teach-In
Practical show-and-tell: learn to use AI agents by doing.
Guided demonstrations with OpenAI Codex, Anthropic Claude Code, and Google Gemini CLI
Individual exercises and group activities
Workflow patterns that generalize across agents and platforms, including Agent Skills
Guest speakers from AI Native law firms and in-house legal teams sharing how they already run professional work through AI agents
Part 2: Live Interactive Agents
Capstone participatory demo: see agents working together in action.
Live multi-agent collaboration on Interlateral.com
Participants bring their own agents into a shared working environment
Real-time cross-participant agent interaction
Experience how agentic workflows support research, drafting, analysis, and coordination
AI Native Lawyers
A new wave of lawyers and legal organizations is being built around AI agents
from the start—not adding agents to existing workflows, but building the
workflow around the agent. These AI Native practices span in-house corporate
legal teams, new-model law firms, and emerging legal organizations.
Several of our guest speakers represent this cohort; additional speakers
to be announced.
In-House Legal Teams
In-house legal teams at the forefront are deploying AI agents directly
into the business workflows they serve—research, contract review,
compliance, and cross-functional coordination—and measuring agents
against the organization's own standards, not generic benchmarks.
AI Native Law Firms
A new wave of law firms is building client service around agentic
workflows from the ground up—using AI agents for research,
drafting, and matter management as the foundation of practice,
not an add-on to it.
Workshop Leaders
The workshop is led by Dazza Greenwood, with guest speakers representing
the emerging wave of AI Native lawyers and legal organizations—spanning
in-house practice, law firm work, and newly emerging legal organizations
built around agentic workflows.
Instructor & Facilitator
Dazza Greenwood
law.MIT.edu · Civics.com
Workshop design, facilitation, and framing: why legal teams need practical agent skills.
Can answer questions about OpenAI Codex desktop setup and workflows.
Short Talk
The Rise of Vibe Coding for Lawyers & AI-Native Law Firms
Matt Pollins delivers a fast-paced talk on the rise of vibe coding for lawyers and AI-native law firms. He covers the shift from traditional software development to AI-assisted “vibe coding,” where lawyers can build tools using natural language prompts. Matt demos live builds, walks through real projects on vibecode.law, and introduces AI Native Law Firm Index — a growing directory of AI-native law firms. The talk highlights how these tools are reshaping legal practice and lowering the barrier for lawyers to create their own technology.
Important: This is a hands-on session. In-person participants
must arrive with a supported or approved AI agent already installed and functioning
before the workshop begins.
Supported Agents
If you have any of the following agents installed on a working laptop (or available via remote access),
you can participate directly in the workshop exercises and activities.
If you have a different agent that is comparably capable to the above (such as OpenClaw),
you may request permission to use it. We will do a quick run-through to ensure it works
well with the workshop exercises and activities.
Pre-Event Teach-Ins
Live sessions via Zoom for anyone who wants help getting a supported AI agent installed and running,
or for anyone who already has one and wants a walkthrough or quick test before the April 13 workshop.
Open to all registered participants.
Zoom link will be sent to registered participants by email. If you can, please have at least one supported agent installed before the session.
Asynchronous tutorials will also be available. IDE plugins (e.g. Claude Code or Codex for VS Code) and other comparable agents like OpenClaw are also welcome — just let us know what you’re using so we can confirm compatibility.
Tutorials
Getting Started with Claude for Legal Professionals
Matt Pollins & Dazza Greenwood · Workshop Prep Session
In this practical hands-on tutorial, Matt Pollins walks you through Claude AI and shows exactly
how legal professionals can start using it today. You’ll discover that Claude isn’t
just one chatbot — it’s actually three powerful tools (Claude.ai, the desktop agent
Claude Cowork, and the advanced agentic Claude Code). Matt explains why the agentic versions
are a game-changer for law, how top engineers are already letting AI handle most of their work,
and the ridiculously simple way to download the Claude desktop app and begin exploring it in
under a minute.
In this practical hands-on tutorial, Dazza Greenwood hosts Marcela Jabór for a live
demonstration of how non-technical professionals can deploy powerful autonomous AI agents
directly on their own computers. You’ll learn how to turn an AI into a true digital
coworker with access to your local files and environment. We also touch on points in the
process when the user is expected to set permissions and authorizations and some of the
risk-oriented balances and decisions involved in setting these configurations. Marcela walks
you step-by-step through installing Codex Desktop, setting up a secure local workspace, and
mastering “agentic prompting” — giving the AI a high-level goal and letting
it handle the technical details on its own. Watch as the agent autonomously browses the web,
runs shell scripts to download files, cleans messy HTML into readable text, performs
cross-document analysis, and ultimately generates a complete, polished PowerPoint presentation
from scratch — all without writing a single line of code.
Recommended background reading to get the most out of the workshop.
Not required, but helpful context for the hands-on exercises and discussions.
This list is a work in progress and will be updated frequently. Check back for additions.
Recommended Reading
How To Train Your Agent
By Matt Pollins (Agents.law) · Jan 2026
How Skills — folders of instructions and reference materials — transform AI agents from
generalists into specialists for legal tasks. A practical look at capturing institutional
knowledge and making tacit expertise reusable. By one of our workshop speakers.
A look at what it means to build a law firm around Claude as the core operating layer —
not as an add-on tool but as the foundation of how the firm works.
By one of our workshop speakers.
By Daniel Martin Katz, Michael J. Bommarito II & Jillian Bommarito · 273 Ventures
A comprehensive guide to strategy, economics, and execution for firms building on AI
as foundational infrastructure. Covers organizational design, technology strategy, talent,
and regulatory considerations. New chapters released weekly — relevant to AI-native
law firms and other AI-native professional services firms.
California Proposes AI Amendments to Rules of Professional Conduct
State Bar of California · Public Comment Open Through May 4, 2026
Proposed amendments to six existing ethics rules (competence, client communication,
confidentiality, candor to tribunals, and supervision) to clarify that traditional
professional duties apply when lawyers use generative AI and AI agents. Not new rules —
targeted clarifications making explicit what competent, ethical practice already requires.
Directly relevant to anyone using AI agents in legal work.
AI Agent ID: Identity, Delegation, and Accountability for Autonomous Agents
By Dazza Greenwood · Nov 2025
As AI agents move from experiments into production, who are they acting as — and who is accountable?
Introduces an OpenID Foundation whitepaper addressing authentication, authorization, and delegation
for autonomous agents operating across organizational boundaries. Directly relevant to
the workshop’s focus on multi-agent collaboration and the Interlateral demo.
Various Posts on AI Agents: A Grounding Collection
By Dazza Greenwood · May 2025
A curated set of posts covering how AI agents are reshaping transactional systems and legal
frameworks — from agency law and loyalty obligations to Agent-to-Agent protocols and error
handling. Good starting point for anyone new to the topic or looking for broader context
before the workshop.
Discover reusable AI automations built by legal experts. Use them instantly
in your favorite AI assistant. A growing library of ready-made agent skills for legal work.
A practical walkthrough of setting up Claude Cowork with custom skills, memory files,
and activity logs — using complementary AI tools together rather than as competitors.
Useful prep for workshop participants getting their agents ready.
A focused, online-first session hosted from Stanford to kick off a 2026
research and convening agenda on custom AI evaluations for
legal teams and other organizations deploying AI systems and AI agents.
Why this kick-off mattered
Legal and compliance teams are being asked to sign off on AI
systems and AI agents that operate in high-stakes environments.
Leaderboards and generic benchmarks are not enough. What they
need are custom evaluations that reflect real
tasks, risk tolerances, and institutional duties.
Looking toward 2026
This December 3 session served as a short, high-signal kick-off:
aligning on goals, shared language, and priority use cases.
The work continues in 2026 through a series of focused
research sprints, working groups, and convenings on evaluation
practices for legal AI and agentic systems.
The Dec 3 Brain Trust
This initial group helped set the direction for the 2026 agenda,
bringing perspectives from legal practice, agent evaluation platforms,
consumer protection, and computational law.
Host & Convenor
Dazza Greenwood
law.MIT.edu · Stanford CodeX · Stanford Digital Economy Lab
Opening remarks and framing: why legal teams need custom evals, not just benchmarks.
Legal Domain
Tara Waters
Vals AI · TLW Consulting
Reflections on legal AI evaluation work and what matters most to law firms and clients.
Agent Measurement
Roman Engeler
Co-founder & CTO, Atla
Overview of evaluating AI agents and what teams are learning in practice.
Agent Eval Infrastructure
Darius Emrani
Founder, Scorecard
Agent evaluation infrastructure for high-stakes legal systems.
Fiduciary & Loyalty
Dan Leininger
Consumer Reports Innovation Lab
Early thinking on measuring loyalty and duty-of-care for consumer-facing AI agents.
Computational Law
Robert Mahari
Associate Director, Stanford CodeX
Computational law perspective on how evaluation connects to legal systems, precedent, and governance.
Pre-reads
Quick reads on why custom evaluations matter in law and what practitioners are learning from real-world assessments.
Beyond AI Benchmarks
By Dazza Greenwood · Sep 2025
Argues that leaders must own “evaluation-as-policy,” turning domain standards into golden datasets and
LLM-judge rubrics so AI systems are measured against what the organization truly values.
Explored contract terms for agentic transactions, authenticated delegation,
and legal frameworks for AI agents. That work provides important context
for the continuing focus on AI agents in law.
