An estimated 52% of law firms use ChatGPT.

But if your team is on Free or Plus, your conversations train OpenAI's models by default.

OpenAI's own terms (updated October 2024) also prohibit consumer plans from being used for "the provision of tailored advice that requires a license such as legal or medical advice."

The Business plan fixes both: no data training, SOC 2 Type II certified, compliant for professional services.

For a 5 to 50 person firm, it's the sweet spot.

I walk through each of these on screen in the companion video.

Here's what they do, why they matter, and a real workflow for each.

A reusable AI assistant with your firm's standards and playbooks baked in permanently.

Example: Your litigation team reviews NDAs every week - each associate applying the playbook from memory.

With a Custom GPT, you encode those standards once, share it via a single link, and every associate gets the same checklist, the same flags, the same output.

The benefit: Consistency across every review, without retraining or supervision.

A dedicated workspace per matter where ChatGPT already knows the context before you ask a question.

Example: Your M&A team sets up a project for a $32 million acquisition - buyer's counsel, California law, deal documents uploaded.

Anyone on that matter can ask about risk factors in the disclosure schedules and ChatGPT already knows the deal.

Memory learns your preferences across sessions - tables over paragraphs, risks flagged red, yellow, green.

The benefit: Stop re-explaining context every session.

Watch out: Claude's Projects handles more files and has a larger context window - stronger for heavy due diligence.

The Legal AI Comparison Matrix shows you exactly which tool wins for each legal workflow - contract review, regulatory research, drafting, data analysis, and more. Filter by your priorities and get a clear answer.

Takes a complex question, breaks it into sub-queries, searches dozens of real-time web sources, and produces a structured report with citations.

Example: Your compliance team needs data breach notification requirements across four jurisdictions - normally most of a day for a junior associate.

Deep Research delivers a structured comparison in about 14 minutes.

The benefit: Regulatory research condensed from a full day into minutes.

Watch out: 25 queries per month, no Westlaw or LexisNexis access.

Perplexity is faster for the same type of research.

Verify every citation regardless of tool.

A dedicated editing workspace where you and the AI work side by side - not a chat window where you copy-paste text out.

Example: Your corporate team drafts an engagement letter, then refines the fee structure - Canvas updates only those clauses while the rest stays intact.

Partner says the language is too dense for the client - a reading level slider simplifies the entire letter in one action.

The benefit: Targeted edits rather than full rewrites, without leaving the workspace.

Watch out: Claude arguably produces higher quality reasoning, but for the drafting experience - inline editing, tone adjustments, typing directly into the document - Canvas is more polished.

Upload a spreadsheet and ChatGPT runs calculations, finds patterns, and produces charts - all from plain English questions.

Example: Your finance partner needs over-budget matters, realization rate by partner, and write-off trends by practice group before Monday's committee meeting.

That's usually half a day in Excel.

Three charts back in about two minutes.

The benefit: Partner-ready insights in minutes, not hours.

Watch out: No other chat-based LLM generates charts this cleanly.

For defensible citations, NotebookLM answers only from your uploaded documents - zero fabrication.

For massive document sets, Gemini and Claude handle over a million tokens; ChatGPT tops out at roughly 272K.

For pre-built legal workflows, Claude's Legal Plugin has slash commands out of the box; ChatGPT you build yourself.

The answer is never one tool.

It's knowing which tool to reach for at which moment.

YOUR NEXT STEP

Everything here, I walk through live on screen in the companion video.

If any of these features sound useful, the video shows you exactly what to expect.

Which ChatGPT feature would save your team the most time this week?

A) Custom GPTs - standardized contract review

B) Projects + Memory - persistent matter context

C) Deep Research - multi-source investigation

D) Canvas - collaborative document drafting

[Reply with your letter choice] - I'll share the results in the next edition.

We've now covered Claude, Gemini, Perplexity, and ChatGPT.

The pattern is the same every time.

No single tool wins everything.

Claude is more conservative with citations and handles larger documents.

Gemini has NotebookLM and the deepest web search.

Perplexity is the fastest researcher with the best citation accuracy.

And ChatGPT has Canvas and Custom GPTs that genuinely change how your team drafts and reviews.

The firms getting the most out of AI right now aren't the ones who picked one tool and went all in.

They're the ones who know which tool to reach for at 9 AM on a Tuesday.

Hit reply and tell me - which feature are you trying first?

— Liam Barnes

We help legal teams implement AI workflows that actually work.

Grab some time to chat

(if you don’t see a suitable time, just shoot me an email [email protected])

How Did We Do?

Your feedback shapes what comes next.
Let us know if this edition hit the mark or missed.

Too vague? Too detailed? Too long? Too Short? Too pink?

👍 Loved it | 👎 Not for me | 💡 I Have a Suggestion

OR: [email protected]

Was this week’s newsletter forwarded to you?

Sign up, it’s free.