Building a Champion AI Team

Sport has a saying: a champion team will beat a team of champions.

People across all industries are experimenting with AI at the moment. In your teams, you may find many of them are teaching themselves around client deadlines and out of hours.

Most firms are currently, albeit not deliberately, building a team of champions. Each person is getting better at AI separately, and 40 people doing that may not add up to a firm that is better at AI.

In Clio's 2026 Legal Insights Report, 98% of the Australian legal professionals surveyed use AI, and 39% have embedded it widely into how their firm works. That difference shows up as time, energy and decisions for your team.

So how does a firm take a room full of AI champions and build a champion AI team?

1Getting comfortable with AI

What we tend to see first is people swapping something they already do for an AI version of it.

AI can give a better answer than the old tool did, hence, why people are using it (even if some of those jobs were already being done well enough). Word has checked spelling and grammar for decades, Excel has done the calculations, and firms have paid for good research tools for years.

The same Clio report found 59% of respondents are relying on generic tools with no understanding of their matters. If replacing an older tool is where the use stops, each person ends up working separately, burning credits on the largest model in the plan to polish an email.

You might have Copilot write up the notes from a recorded meeting and pull out a list of actions. There are no minutes to type up, and the team signs off on one record.

Then what happens after the meeting? What we're seeing is that the actions are often handled the way they were before.

2Where the good work goes

You may have someone in your firm who has worked out something useful. It might be a way of drafting a certain kind of letter that comes back nearly right every time, or a method for pulling the key dates out of a long document. It took weeks of trial and error to get there.

Ask where that method is written down. It could well be in one person's chat history. Firms have kept precedent banks and template folders for years, and a shared place for prompts is the same idea applied to AI.

So another team may end up solving the same problem again, and when the person who worked it out leaves, the method can go with them.

3Many versions of the same job

Client onboarding might be done one way by one person and another way by the next. The same goes for file notes, status updates and first drafts of advice.

People now have tools that can produce a complete, confident draft of almost anything, and each person may direct them a little differently. A different model, or a different account, can give a different answer to the same question.

So the firm's work can go out in several voices. Firms have spent years on brand guidelines for the logo, the letterhead and the firm name. Marketers will remember the Bureau of Meteorology's 2022 rebrand, when it asked media outlets to stop saying BOM and call it the Bureau. According to the ABC, it cost more than $220,000. The same kind of rules can help with how AI writes on the firm's behalf.

One person drafts an email with AI. The person receiving it summarises it with AI and replies with AI, and two people have put their names to a conversation between two machines. For advice, a file note or a set of figures going to a client, where is the human check?

4What is in use

Personal licences, free accounts, a tool one person found and liked, a browser extension, a feature that arrived inside software the firm already pays for. These tend to arrive the way shadow AI does, quietly and with good intentions.

Across 31 countries, Microsoft's 2024 Work Trend Index found 78% of AI users bring their own AI tools to work, rising to 80% at small and medium-sized companies, and 52% are reluctant to admit using it for their most important tasks. Clio's report found 11% of firms have no formal AI policy in place. Where a firm hasn't said what is approved, people tend to pick something and get on with the work.

Ask across the firm:

One instinct is to lock everything down, which is how firms handle shadow IT. With AI, the tool is a website, and the person using it may well be a partner. A firm can only make decisions about the tools it knows about, so the first step is finding out what is there.

5Busier in our own lanes

People are trying hard with AI. They have been told it is enormous and that they should be using it, so they are. That can mean more drafts and more reviewing, without necessarily changing how the firm works as a whole.

MIT's research, reported by Fortune, found about 95% of enterprise generative AI pilots were delivering little or no measurable impact on profit and loss. Clio's report also found 41% of legal professionals invest in technology purely to save time, and 24% invest in it to grow their business.

Chartered Accountants Australia and New Zealand (CA ANZ), reporting Ipsos research via Chartered Accountants Worldwide, found 92% of chartered accountants want AI training and 30% have received any. Microsoft found 39% of people using AI at work have had AI training from their company. Where training hasn't arrived, people tend to teach themselves.

6What a champion team looks like

A champion AI team is the whole firm working to a shared method.

The receptionist, the paralegal, the bookkeeper and the partner all know what AI is for in the firm, how it gets used, who checks it, and what good looks like. When one person finds a better way of doing something, it becomes the way the firm does it.

AI capability is skill that belongs to the firm, so it stays when people move on, and a new starter can be shown it in their first week.

7Five ways to start

All five start with the people and the meetings a firm already has. They need a decision that AI is something the firm does together.

8One question for every use of AI

For any use of AI in the firm, ask:

Spotting that kind of use means looking across the whole firm.

In many firms, the curiosity is already there, and the work that would benefit most is work the firm does every week.

9Where we come in

Our investigation finds out what your people are already doing with AI, where the strongest methods sit, and how long your firm's everyday work takes today, so you have a baseline to build from. The findings belong to your firm whether or not we go any further together.

If you would like to know where your firm sits today, reach out to our team. The first conversation works out whether your firm is the right fit for the AI Uplift Program, and whether we are the right people to help you build your champion team.

Getting started