AI training for law firms in Chicago

FitLaw firms and legal departments · 3–200 people

• Litigation practiceschronologies, deposition prep, discovery logs
• Transactional and real estate practicescontract comparison, closing checklists, client summaries
• Family, immigration and PI firmshigh-volume intake and client communication
• Firm administrators and COOswho own the policy, the tools and the billing rules
• Firms where associates use personal accountsand partners want it moved onto approved tools
• In-house legal teamscontract review queues and business-facing summaries

The curriculum is the same one behind our AI training for Chicago teams; for a law firm we change the examples, the rules and the labs. A paralegal learns on a fictional produced-documents set, an associate on a fictional research question. Nothing from a real client file is used in the room.

Still choosing a tool? Our comparison of ChatGPT, Copilot, Gemini and Claude for small business covers plans and data controls; firms that live in Outlook and Word should also see Microsoft 365 Copilot training.

What does a law firm prompt look like?

A useful legal prompt names the source, the task, the format and the limits, and it tells the model what to do when the source is silent.

SOURCE   The 40 emails pasted below (Hartwell v. Lakeview Freight,
         a fictional matter prepared for this lab).
TASK     Build a chronology of every communication about the
         March delivery delays.
FORMAT   Table: date, from, to, one-line summary, Bates number.
RULES    Use only the emails provided. Quote, don't paraphrase,
         any sentence about who approved the delay. If a date is
         unclear, write "unclear" and say why.

The rules line does most of the work: it keeps the model inside the documents and makes gaps visible instead of filled. Once the prompt works on the fictional set, it is saved to the firm’s library with an owner and the approved tool it belongs in. Our guide to prompt engineering basics for employees goes further into the structure.

• Client confidences in a personal or consumer accountno contract, unclear training settings, no firm control
• Privileged communications and work productonly in an approved tool, and with client consent where the tool learns from inputs
• Material under a protective orderthe order may limit who can see it; a vendor is a “who”
• Third-party personal data from discoverySocial Security numbers, medical records, financial accounts
• Settlement positions and strategyin any tool that keeps or trains on what you type
• Unchecked citations or quotationsinto anything that will be filed or sent

If a client’s outside counsel guidelines say anything about generative AI, those terms govern that client’s matters. Check them matter by matter; a firm-wide policy cannot override a client’s instruction.

Fabricated citations are the failure everyone has heard of. In Mata v. Avianca (S.D.N.Y. 2023), lawyers filed a brief citing six cases that did not exist and were fined $5,000. A January 2026 article from the Illinois courts cited a Bloomberg Law count of more than 280 court filings with hallucinated citations since 2023. The fix is a habit: a person reads every authority before relying on it. Our AI acceptable use policy template is a starting point for writing these rules down, and the note on shadow IT in small business covers moving staff off personal accounts without a fight.

The ARDC’s Illinois Attorney’s Guide to Implementing AI (October 2025) includes a sample client notice, a staff policy, an informed consent form and a checklist for reviewing a vendor’s terms of service. The ARDC calls these resources nonbinding, and the Supreme Court’s policy controls in any conflict. We follow the guide’s structure when helping a firm draft its policy, and the draft goes to your counsel.

Separately, since January 1, 2026 the Illinois Human Rights Act requires employers, firms included, to give notice when AI is used in hiring and other employment decisions. If your recruiting team uses AI, see the Illinois AI employment notice kit.

This is general information from 312 IT Consulting, not legal advice. We are not a law firm. Confirm how these rules apply to your practice with your ethics counsel or the ARDC Ethics Inquiry Program before relying on it.

1. Free first call. Firm size, practice areas and which AI accounts exist today, personal ones included.
2. Tool and policy check. Your plan, its data and training settings, retention, and whether a written AI policy exists. If not, we start a one-page interim version.
3. Collect real tasks by role. Each group sends two or three weekly tasks; we rebuild them as fictional matters.
4. Foundations. How AI works, tokens and context, prompting, checking the work and privacy, working from your own material, and using AI as a thinking partner on a fictional problem.
5. Applied labs. Three labs on your own processes, for example intake notes to conflict form, produced emails to chronology, and time entries to client guidelines.
6. Saved instructions. The prompts that worked are written up as reusable instructions in a shared library, each with an owner and the approved tool it belongs in. Certificates of completion for every attendee.
7. Follow-up. A few weeks later: which workflows people kept and what the policy missed.

Formats run from a 2–3 hour workshop (one role lab) to a 1-Day Foundations or 2-Day Hands-On program with all three labs. The difference between a standard and a fully custom program, and the confidence results from past sessions, are on the 312 AI training results page. 312 has delivered AI training for companies from small and mid-size businesses up to Fortune 500 firms, and its founder has led AI training for 1,200+ people across four regions. For the minute-by-minute shape of the short format, see the AI workshop in Chicago page.

Tailoring is the main price driver: standard legal examples cost less to prepare than rebuilding your intake form, billing guidelines and a fictional matter for each practice area.

Illinois firms with a physical location may be able to recover part of the cost through the state’s Employer Training Investment Program, which requires pre-approval before training starts. We issue the certificate of completion it asks for. The conditions are in Funding AI training in Illinois, and the wider price picture is in What AI training costs in Chicago.