AI-assisted legal drafting, correspondence, template-based drafting, and iterative refinement

AI definitely has a place in legal services. I use it to review voluminous files, prepare preliminary drafts, and double check certain work for any additional suggestions. Given my legal training and three decades of experience, I know how to prompt it, I know how to review it, and I know when it’s right and when it’s wrong. It is still very often wrong.

I continue to see very bad AI created documents coming from clients and potential clients. Please understand that while these documents might have the right label, they lack the content necessary to solve problems and avoid litigation. I am deeply concerned that when problems arise, business owners and executives will lack the tools in their contracts and governance documents necessary to resolve them.

Please do not fuel the litigation pipeline. Consult with counsel about your legal needs and find counsel who can use AI appropriately as a tool to reduce legal fees, expedite services, and provide the best outcome.

You can draft an agreement using AI without going to a lawyer.

You can take medicines after asking AI about your symptoms without going to a doctor.

You can write a LinkedIn post without hiring a content strategist.

But none of that makes you a lawyer, a doctor or a content strategist.

Every time I open LinkedIn, someone is posting about how AI now does their drafting, their writing, their thinking. And you can see the result. Everything reads the same. Everyone sounds the same.

I use AI. Most days. It saves me time on the parts of the work that were never the valuable parts.

But it does not know my client.

Someone will say I can just prompt it with everything I know about the client.

You can, but it still does not do the part that matters, because that part is not information, it is judgment.

AI will never tell a client that the post they want is a bad idea. It will write it, and it will write it well. It will not say that this is the wrong month to be visible on this issue, because the firm is on the other side of a matter that has not been reported yet. It will not say that a post which performs very well can still cost a referral relationship that took eight years to build.

It optimises for the post. I am being paid to think about the practice.

A lawyer works the same way. AI will list the compliances under labour law. A lawyer tells you which one the inspector actually checks, which one will surface in due diligence three years from now, and which one is not worth the cost of fixing today.

That gap is the entire job.

Having access to a tool is not the same as knowing what to do with it. And the people who confuse the two are usually easy to identify, because their work is indistinguishable from everyone else's.

AI Governance and Compliance: How I Use AI as In-House Counsel?

If you had asked me in 2022 how I used AI in my legal work, I probably would not have had a substantive answer. Since early 2023, AI has become part of my regular workflow—for summarising emails, organising information, preparing first drafts, extracting contract obligations, creating checklists, and supporting legal research.
However, every AI tool cannot be used for every legal task. A tool approved for email summaries may not be approved for contract redlining, depending on its licensing, data-processing terms, security controls, and organisational guardrails.

That is why AI governance matters.
The key questions are:
What data can be entered?
Is the tool approved for the specific use case?
Will the data be retained or used for training?
Is human review required?
Can the output be verified and audited?

The legal AI market is projected to grow from USD 3.11 billion in 2025 to USD 10.82 billion by 2030. As adoption grows, legal teams must focus not only on using AI efficiently but also on using it responsibly. The future of legal AI will depend on innovation with accountability.
Read more: MarketsandMarkets Legal AI Software Market Report
#AIGovernance #LegalAI #LegalTech #InHouseCounsel #Compliance #ResponsibleAI #LegalInnovation #FutureOfLegalWork

We have all heard that AI is going to reshape the legal profession. But after spending significant time integrating it into my day-to-day work, I have realized a fundamental truth: AI isn’t a search engine. It’s a junior associate. And just like working with a junior associate, the quality of the output is entirely a reflection of the quality of your instructions.

If you want partner-ready drafts, you need a structured framework. In this short video, I break down the four pillars I use every day to get precise, usable results:

⚖️Role: Anchor the perspective (e.g., “Act as senior commercial counsel”).
📝 Task: Use precise legal verbs (“Draft,” “Extract,” “Redline”).
📖 Context: Define the jurisdiction, the client's side, and your risk appetite.
🧠 Judgment: Demand structural constraints and verification safeguards.

The golden rule I always follow? Trust it to accelerate the draft, but distrust it enough to verify every word.

I would love to hear how other practitioners are prompting their workflows. What’s working for you, and what are you still refining? Let's discuss in the comments.

#LegalAI #LegalTech #ArtificialIntelligence #InHouseCounsel #Lawyers #GenerativeAI

Since launching its Clio Work AI product in October 2025, Clio says it has been the fastest-adopted product in the company’s 18-year history. But there was one place where Clio Work was not available — Canada, the country where Clio was founded and is headquartered.

Now Clio has brought home the bacon, so to speak, launching Clio Work in Canada — and doing it roughly six weeks after acquiring the Canadian legal data company that made the launch possible.

Clio Work went live in Canada on July 23, enabled by a corpus of more than 470,000 Canadian cases drawn from Jurisage, the Toronto-based legal AI and data company Clio acquired in June.

Here is the first installment in a short video series on how I use AI to streamline my daily work for Practical Counsel Advisors. It covers using AI to develop template documents and emails - a simple operational task that is low risk and an example of one way to get comfortable using AI. Future video topics will include vendor and tool evaluation, preparing for speaking engagements, and potentially your requests.

I do not use a dedicated legal AI review platform. I use a frontier general-purpose model on commercial terms, with a structured set of instructions built around it that enforce the discipline the platforms sell as a feature. Generally, Claude (on their Pro Max plan) at the time of writing, but ChatGPT’s Work (released this week) is looking decent too.

Building on Libra by Wolters Kluwer’s early leadership, strong market momentum and advanced AI innovation in Italy, its capabilities are now directly available in One via the Libra add-in.

Milan, July 8 2026 – Wolters Kluwer Legal & Regulatory today announced the integration of Libra by Wolters Kluwer AI workflows in One, its leading legal research platform in Italy. This step further reinforces the strong commercial success and rapid market adoption of the Libra legal AI workspace in the country. With the integration generative AI features are directly available in the research workflow of legal professionals allowing a seamless experience without switching between tools.

There are dozens of AI prompt frameworks that lawyers are being told to use. But giving good instructions is already a core legal skill. We don't need another acronym. We just need to draw on those same principles:

1️⃣ Role

Tell the AI who it is and what expertise to draw on. Being specific is good, but like all things legal, there’s a balance with being too specific as the data might be too limited. In my example, I’ve referenced professional athletes, but within the broader category of high net worth individuals.

2️⃣ Purpose

Identify the client and state what they're trying to achieve. Try to provide as much context as possible. It can also help to refer to any notes you may already have.

3️⃣ Task

State the specific deliverable that we need to prepare. If the draft email is going to a client or a supervisor, state that. It’s helpful also to include any examples, checklists, risk matrices etc. that should be followed. Same as if you were giving a junior a precedent to use instead of making them start from scratch.

4️⃣ Key facts

Set out the info that isn’t publicly available or that the AI should focus on. In this case, we are asking them to consider 2 specific options instead of every contract extension variant.

5️⃣ Constraints

Here we build in guardrails. For example, the format, word count, tone, house styles, things to avoid etc. The more key constraints we can include, the more relevant the final output will be.

6️⃣ Iterations

This bit is something I like to do whenever I give instructions to people, and I’ve found it helpful when prompting AI. It’s best to think of prompting as an iterative process instead of something you need to get perfect in one go.

NB: This guide is tool-agnostic. Follow your organisation's rules for AI use.

What’s your prompting approach? Do you just give instructions like you would to a human, or do you have a system that works better?

----------

I'm building an e-library with the 110+ how-to guides I've written over 6 years. If you want early access when it launches, sign up to the free newsletter on my website / profile.

Starting this week, CoCounsel Legal customers will get early access to the next generation of CoCounsel, ahead of the full launch planned for later this summer starting in the U.S, and followed by rollouts in Canada, UK and Australia. The product will keep getting better. But it is already good enough that keeping it from you any longer felt like the wrong call.

The next generation works the way an attorney works with a teammate. Describe your matter in plain language. CoCounsel Legal creates a plan, dives deep into legal authority, reasons through the legal issues, retrieves what it needs from your own precedents, and from Westlaw and Practical Law, and drafts with citations. When new information changes the picture, it adapts. Instead of getting piecemeal answers, you get a professional standard response that you can iterate on.