Why The GAiGE 2.0 starts with AI governance

We launched The GAiGE in May to answer a question we kept hearing from the organisations we work with: are our AI tools worth what they cost? It's still a good question, and The GAiGE still answers it. But as more organisations tried it, and as we talked it through with their leadership teams, something else kept coming up first. Before they could think about the return on their AI tools, they wanted to get their AI governance right. So that is where version 2 begins.

What we kept hearing

Many of the organisations we spoke to had started with AI the way most did: licences for ChatGPT or Copilot, a sensible hope that people would find good uses for them, and perhaps a policy written in a hurry when the tools first appeared. That was a reasonable first step. It was never meant to be the destination, and the tools have moved on a long way since.

What their leaders wanted now was a clear picture of which AI tools were in use across the organisation, what had been agreed about each of them, and something they could show a board, a client or a regulator if asked. Most of them also felt a little stuck. They knew this needed sorting out, but the only way they could see to do it was months of expensive outside help.

What's at stake for directors

The concern is well founded, and it is a board matter. In Australia, the corporate regulator, ASIC, has set out that directors' and officers' duties of care and diligence apply to AI: "These duties extend to the adoption, deployment and use of AI." ASIC adds that directors and officers should be aware of how AI is used within their companies (Source: ASIC, Report 798: Beware the gap: Governance arrangements in the face of AI innovation, 2024). The Australian Institute of Company Directors said the same in its 2026 guide for boards: overseeing AI and its risks is already part of directors' duties under the Corporations Act (Source: Australian Institute of Company Directors and Human Technology Institute, A Director's Guide to AI Governance, Version 2, 2026).

Meanwhile, a good share of the AI used at work was never chosen by the organisation at all. A third of Australian workers (34%) use AI tools at work without their employer knowing (Source: Employment Hero, AI Paradox at Work research, 2026, a Focaldata survey of 1,634 employed Australians). Two years earlier, Microsoft and LinkedIn found that 78% of knowledge workers using AI at work were bringing their own AI tools, in Australia and worldwide alike (Source: Microsoft and LinkedIn, 2024 Work Trend Index). We've written before about what that looks like from the inside.

Put the two side by side and the exposure is plain. Directors are expected to be aware of how AI is used in their company, while a third of workers use AI tools their employer doesn't know about. The gap closes in steps, and the first ones are straightforward: find out which AI tools are in use, agree what's allowed, and tell people clearly.

Why governance comes first

It's hard to judge whether a tool is worth what it costs when nobody has agreed that it should be in use, who owns it or what it's for. Governance gives you those decisions: which tools are approved, for what, within what limits, and who is answerable for each one. Once they're in place, looking at the return becomes a fair test of a choice the organisation made, rather than a guess about tools nobody chose.

There's a second reason, and it's the one closest to how we work. Good AI governance is a set of decisions you come back to as the tools and the organisation change, which makes it the same agile loop we use for adoption: find out where you are, decide, act, review and adjust. A policy written once and filed away goes out of date as quickly as the tools it describes. A decision with a reason, an owner and a date to look at it again can be changed when it needs to be.

Taking the legwork out of it

We're a consultancy, so we're aware of how the next part sounds coming from us. A lot of the work of AI governance is legwork: surveying everyone about the tools they use, keeping a register up to date, drafting policies, explaining them to each group, checking who has read them and putting together the agenda for each review. It matters, and it takes time, but very little of it needs a consultant in the room. What does need people is the judgement: how much risk the organisation will accept, how fast it wants to move, and what each tool should be used for. That's where we'd rather spend our time with clients.

So over the last several months we've built version 2 of The GAiGE to do most of that legwork and leave the judgement with your leaders. With Standard, you get:

• Finding out. The AI State of Play is a survey of everyone, which takes each person under 10 minutes, so you know which AI tools are really in use and what for. The Industry AI briefing adds outside context: how organisations like yours use AI and where it could help, researched from your website, with its sources.
• Deciding. With the Leadership Position tool, your leaders agree where the organisation stands on seven decisions about AI, with a dated record of what they agreed. The AI tools register keeps every tool in one list, with its decision, its owner, its costs and when it renews. Every decision and every policy starts from our recommendation and the reason for it, and you decide.
• For your people. Each person sees the policies that apply to them and why they're there, and you can see who has confirmed they've read them and remind the rest in one go. The AI Adoption Portal gives everyone their own page, with a chatbot that answers from your published policies, and the questions it can't answer show you where a policy has a gap. The browser extension, for Chrome and Edge, shows your decision on a tool's own website the first time someone opens it, and reads nothing on the page.
• Looking again. Three short pulses, The AI Pulse, The AI Review and The AI Capability Check, run on a schedule and ask whether each tool is helping, whether it's worth keeping and what skills people need. Reviews, monthly or quarterly, bring the findings, decisions and renewals that are due into one place, with the agenda put together for you.
• Showing others. The How We Use AI Report is a dated PDF of your tool decisions and policies, ready for clients, suppliers or the board. Automated decisions keeps a record of the decisions software makes about people, with wording drafted for your privacy policy, ready for the Australian Privacy Act requirement that starts on 10 December 2026.

The full list is in what's new in The GAiGE 2.0.

Once your governance is in order, the question we started with comes back: is each tool worth what it costs? With Pro, you also get ROI by tool, which values the time people say each tool saves at your hourly rate and sets it beside what the tool costs. The AI Impact Report puts the hours saved, their value and the licences going unused into a PDF for whoever approves the renewals. You also get a plan of your AI initiatives that everyone can see, AI Champions, team results for Group Leaders, and the pulse library and custom pulses.

Where we fit

The GAiGE stands on its own. Governance set-up takes an Owner or Admin through each step in order, and an organisation can work through it by itself. If you'd like a hand, that's where we come in. Our free AI Maturity Assessment shows where your organisation stands across strategy, technology and people, and our AI Adoption Roadmap sets out what to do first, and why. From there we work alongside you, hands-on at first and then increasingly hands-off, through the same loop The GAiGE is built around. You can start with either, and add the other when it helps.

You can give The GAiGE a go at thegaige.com. The trial runs for 30 days, for up to 200 people, with no card needed, and it starts on Pro, so you can see everything before you choose a plan. If you'd like a hand working out where to begin, find your starting point with our free AI Maturity Assessment.