Are you feeling like there's a lot more you could be doing with AI today than what you're actually doing?

Well, you're not alone.

Let's look at the three drivers of powerful, impactful, fast AI adoption, so you can work out where you're actually at, and what you can do today to close those gaps.

The three drivers

Capability, that's our knowledge. Knowing what AI can do for us today.

Utility, that's really putting it into workflow. Putting it into use.

And safety, of course.

Now, the gap is hardly ever in capability.

Yes, there's always a lot more to know in AI, but that's not what’s missing.

The gap is almost always in utility. In actually putting what we already know into workflow.

I could keep teaching you new capabilities every single week, but if none of it makes it into the way you work, what’s the idea?

So the question isn't what you know. It's what you're doing with what you already know.

So where are you?

Let's look at utility on its own, and put ourselves into three groups. Behind, mainstream and ahead.

Behind. How do you know if you're behind with AI today?

Well, it's mid 2026, and if you're not using AI every day, a few times a day, for work, you're behind.

You are perfect, loved, and valid, but we’ve got to figure out what's blocking you and get you unstuck.

Is it that you don't have access to secure tools? That you're not sure what you're allowed and not allowed to do? That you haven't found the use cases? Whatever it is, that's the thing we need to change, so you can come into the mainstream.

Mainstream. Mainstream has been using AI as chat so far: So I'm doing something in my work, and I jump over to AI to ask a question, and I come back, and I copy and paste across.

And now, the mainstream is moving from small wins to big wins through reusability and through the entry level agentic AI.

What we're doing now is creating reusability

So we're saving our prompts in ways that we can use again, we can expect relatively consistent behaviour and we don’t need to give context every time again.

We're creating reviewer agents that review our GST, our income tax returns and our advice letters. We're creating drafters that draft those same workpapers and letters. We're creating extractors that find data in messy information.

We go pattern by pattern, creating those reusable cases, and that takes us from small wins of "oh, it helped me with this task here", to big wins like “It's cutting 45 minutes every time we're doing a GST return. Or it's just drafted the AML risk assessment for 600 clients.”

And then we roll it out

Last week a director in a firm showed me how he used AI to review an income tax return. It took the review process from an hour down to fifteen minutes, and it even found some things that he’d missed. How awesome.

But now the question is, how many reviewers are there in that firm who could use this?

And really, how many preparers are there in the firm who could use it before they even give the work to the human reviewer?

And how do you actually do that?

Well, it depends on the tool you're using.

They can be called agents, skills, or so many other names.

It doesn't matter.

As long as you capture your good prompts, you give them knowledge to ground them in, checklists and templates, and you roll that out to your team.

That shift, from small wins to big wins, from I'm using it on the side to everyone's using it in their workflow, is what mainstream businesses are doing with AI today.

And safety, of course

What goes on in mainstream today is that we've got our secure tools that we're comfortable using in our process, with client information and confidential information, because we've researched them and we know what's right.

We've got a good AI policy that covers the capabilities of agentic AI and integrations, not just chat.

And our team all know how to use AI safely.

That’s the benchmark: secure tools (that are properly set up) and an up to date policy.

Your homework this week

Find one way you use AI that's a small win, and make it into a big win by making it reusable and sharing it with your team.

Take that good prompt, make it into an agent or a skill (or whatever your AI tool calls it), and share it.

And in the area of safety: make sure everyone knows what their safe AI environment is, and that your AI policy covers agentic AI.

How do you keep on top of all of this?

Well, I'm not going to say ask AI.

Because if there's one thing that AI so far is not amazing at, it's self awareness. It doesn't know what it can do. It lives in the box and it doesn't see the box. And it gets up to date from what's public online.

But what's online is the innovators on social media talking about emerging capabilities that are immature and experimental and not relevant for us today. And on the other hand, things that are a bit behind.

For example, if you ask AI how to write a good prompt, it'll tell you what was right in 2024 and 2025. It'll give you those structures for a prompt that takes a page or two. That's not how we prompt today.

The best way I know to learn about AI, to keep up to date, and to know what others are doing, is to talk to others and see what they're doing. Yeah, humans talking to humans, I know…

Want to do this together?

If you're in accounting, bookkeeping or otherwise in the finance world, we do that together in our AI club. We share prompts, use cases, policy templates and news updates.

If you're in other professions, invest the time in finding a group that does this together.

This is the best way to learn.-

Inbal Rodnay

Guiding Firms in AI Adoption and Automation

Keynote speaker | AI Workshops | Executive briefings | The Tech Savvy Firm

When you are ready, here is how Inbal can help:

Are you behind, mainstream or ahead?

The three drivers

So where are you?

What we're doing now is creating reusability

And then we roll it out

And safety, of course

Your homework this week

How do you keep on top of all of this?

Want to do this together?

© 2026 Inbal Rodnay. All rights reserved.

Are you behind, mainstream or ahead?

The three drivers

So where are you?

What we're doing now is creating reusability

And then we roll it out

And safety, of course

Your homework this week

How do you keep on top of all of this?

Want to do this together?

Are your prompts too long?

© 2026 Inbal Rodnay. All rights reserved.