A personal legal AI toolkit: a prompt library organized by matter type, a document review workflow, and a client communication system — all designed to run sustainably without compromising professional responsibility.
Designing Your Practice AI Workflow
Start by mapping where time goes in your practice. The highest-value AI use cases are different for different practice areas.
Transactional (contracts, M&A, real estate):
→ Highest value: issue spotting, clause extraction, summaries
→ Medium value: first drafts of standard agreements
Litigation:
→ Highest value: research planning, memo drafting assistance
→ Medium value: discovery document organization prompts
Regulatory/Compliance:
→ Highest value: regulatory analysis, policy drafting
→ Medium value: compliance calendar and tracking
Client-facing work (any area):
→ Highest value: client communication, plain English summaries
→ Medium value: status updates, client education materialsPick the two or three tasks where you spend 2+ hours per week. Those are your first AI automation targets.
Professional Responsibility Guardrails
Using AI in legal practice raises professional responsibility questions you need to address proactively — not react to when something goes wrong.
Competence. Many bar associations now require basic AI literacy as part of technological competence. This course counts toward that understanding.
Confidentiality. Know what goes into your AI tool and what doesn't. Use enterprise tools with appropriate agreements. Never paste client-identifying information into consumer AI.
Supervision. You're responsible for the AI's output. Review everything before it goes to a client or court. Build review into your workflow as a non-skippable step.
Disclosure. Some courts and clients are starting to require disclosure of AI use. Know your jurisdiction's requirements and build disclosure language into your engagement letters if needed.
Document your AI use. Keep a log of what AI tools you use and how. This protects you professionally and builds institutional knowledge for your firm.
Build Your Legal Prompt Library
Create a document organized by task category. Here are the 8 core prompts for most legal practices:
- Contract Issue Spotting — systematic review from client's perspective
- Plain Language Summary — any legal document for non-lawyer audience
- Clause Extraction — pull specific provisions from multiple contracts
- Research Planning — issue identification and search strategy
- Research Memo Drafting — structure from verified research notes
- Client Status Update — clear, professional, client-friendly update letter
- Policy Drafting — compliance policy from regulatory requirements
- Regulatory Impact Analysis — what a new regulation means for your client
For each template, add notes on: what data is safe to include, what data must never be included, and what to check before using the output. Make it a training document for any associate who joins your practice.
Continuous improvement. After each AI-assisted task, spend 2 minutes noting what worked and what didn't. Over 3 months, these notes become a refined prompt library that's dramatically more effective than what you started with.
What You Learned Today
- How to map your practice area to identify the highest-value AI use cases
- The four professional responsibility guardrails: competence, confidentiality, supervision, disclosure
- The 8 core legal prompts that cover most practice area needs
- Why documenting AI use protects you professionally and builds institutional knowledge
Go Further on Your Own
- Review your state bar's current guidance on AI use in legal practice. Does your current workflow comply? What needs to change?
- Share your prompt library with one colleague in a different practice area. Collaborate to build a combined library that covers both areas.
- Design an AI disclosure paragraph for your engagement letters — one that's transparent about your AI use while reassuring clients about your oversight and professional responsibility
Course Complete!
You finished all 5 days. Ready to go deeper?
Reserve Your Bootcamp SeatWant live instruction and hands-on projects? Join the AI bootcamp — 3 days, 5 cities.