It’s past dinner time and you’re still polishing the same trust paragraph you started after lunch. Meanwhile, your feed is full of posts about AI drafting entire contracts in seconds. You can’t help but wonder, should you be using tools like ChatGPT in your practice?
Let’s unpack that question honestly. Because the real answer isn’t yes or no — it’s how and where AI should fit into a small firm’s day-to-day reality.
The State Of AI In Legal Practice
According to Clio’s 2024 Legal Trends for Solo and Small Firms, only 19% of law firms were using AI in their practice in 2023. In just one year, that number exploded to 79% of legal professionals using some sort of AI tool. Why? Because the biggest benefits they see aren’t about replacing lawyers — they’re about saving time and reducing tedium.
Deterministic vs. Non-Deterministic AI: Why ChatGPT Feels “Smart” (But Sometimes Isn’t)
Not all AI is created equal. AI is a broad term and is used with varying levels of accuracy. OpenAI’s ChatGPT brought AI technology to the mainstream so many people use the term AI synonymously with generative AI tools like ChatGPT. You don’t need a deep background on the history of AI as an academic discipline. Probably the single most important distinction we as attorneys need to keep in mind is deterministic vs non-deterministic.
- Deterministic systems — like your document-automation platform or intake forms — follow strict logic. The same input always produces the same output.
- Non-deterministic systems — like ChatGPT — use probability to generate text. The same prompt can yield different answers each time.
Think of deterministic AI as your paralegal following a checklist. Generative AI is more like a sharp new intern: quick, creative, but occasionally confident about things that aren’t true. That distinction matters. Deterministic tools (like Estate Engine) are ideal for repeatable, rule-based drafting. Generative tools are great for brainstorming, summarizing, and polishing — but they still need human review.
Assuming you want to be confident in the quality of the documents you provide to your clients, drafts created by generative AI tools would need to be reviewed line by line. That kills the time-savings you turned to them for in the first place.
Where Generative AI Actually Fits in Estate Planning Work
High-Value vs. High-Time Tasks
Every week you juggle two kinds of work:
| High-Value | High-Time |
|---|---|
| Advising clients | Draft formatting |
| Tailoring trusts | Proofreading |
| Navigating family dynamics | Client follow-ups |
| Explaining complex choices | Intake reminders |
Generative AI shines in that right-hand column — the repetitive, mental-energy-draining tasks that don’t require a J.D. According to the 2024 Clio Legal Trends Report, 74% of billable tasks in law firms are potentially automatable, especially documenting, gathering, and analyzing information.
Automation doesn’t threaten good lawyers — it frees them to do more of what clients actually value: judgment, empathy, and strategy.
The Real Risks Of AI: Accuracy, Ethics, And Economics
I’m a big proponent of using AI tools in my own practice. But there are concerns to be aware of. I look at these as fitting into one of three categories.
1. Accuracy And Hallucination Risks
ChatGPT can sound persuasive while being completely wrong — citing non-existent cases or misquoting statutes. Always verify every AI-generated line of legal text.
2. Confidentiality And Data Privacy
Typing client details into a public AI tool may breach confidentiality. The ABA Formal Opinion 512 reminds lawyers that they must supervise technology and protect client information. A safer path is to use enterprise AI tools with data-control settings or purpose-built legal platforms.
3. The Billing Model Problem
Under traditional hourly models, automation reduces billable hours — potentially $27K per lawyer per year (according to the Clio report introduced earlier). Firms that shift toward flat-fee pricing can turn that same efficiency into higher profit and faster case turnaround.
A Framework For Using AI Responsibly
None of this should scare estate planning attorneys away from utilizing AI in their practices. Here’s how you might approach it:
Step 1: Start With Tasks, Not Tools
List your time drains first: intake follow-ups, template version control, proofreading, routine emails. Then choose tools that handle those specific jobs.
Step 2: Pair Generative And Deterministic AI
Use deterministic automation for drafting and data consistency, and generative AI for summaries, cover letters, or marketing content.
Step 3: Build Human Oversight In
Every AI output should be reviewed by someone with a law degree. Document your review process and update client-data policies accordingly.
How To Evaluate A New AI Tool
Before you sign up, ask:
- Does it clearly save measurable time?
- Can I verify every output?
- Does it protect or isolate client data?
- Is it compatible with my billing model (flat fee vs. hourly)?
- Will it free up my attention for higher-value work?
What Smart Small Firms Are Doing Now
Firms embracing client-facing tech — online schedulers, intake forms, and e-signatures — see 51% more leads and 52% higher revenue on average. The takeaway? The best use of AI in estate planning isn’t drafting wills — it’s eliminating the friction that keeps you from serving more clients.
AI won’t replace estate planning attorneys. But attorneys who use AI to eliminate busywork will have more time for what no algorithm can do — understand human relationships, family nuances, and legacy goals.
If you treat ChatGPT as a thought partner — not a paralegal — you’ll get the best of both worlds: more time, less tedium, and a practice that feels a little more modern every week.
