Q+A: The hidden risks as accountants embrace AI

Welcome to Fear and Greed Q&A, where we ask and answer questions about business, investing, economics, politics, and more. I'm Sean Aylmer. Artificial intelligence is rapidly finding its way into accounting firms, helping with everything from research and document reviews to client communications and compliance work. but it's not all smooth sailing. The pressure to adopt AI is growing while concerns around accuracy, confidential client data and staff using unimproved AI tools aren't going away. This episode was recorded at Class Ignite 2026, proud sponsors of our podcast. Jason Entwistle is Director of Strategic Development at Hub24. He joins us. Hello, Jason. Hello. Dr. Evan Morrison is head of the Innovation Lab at Hub24. Evan, welcome. Thank you. And Jason Noy is manager of accounting at New Haven Group. Hello, Jason.

How are you going, Sean?

Jason, I'm going to start with you. Jason Noy, you're actually using AI inside your firm. What are you using it for? Where's it made the biggest difference?

So we're using AI in a couple of different facets. One, you've got email triage. You're going through telling the AI to sort your emails into those which are important or urgent and require client response versus those that might be newsletters that you can look at later and file those away. We also have used AI for a number of sort of project-related matters. So we've migrated our software from one SMSF platform to another, and we've used AI to compare the data from reports. So AI has been very good at analyzing the data and making sure that that is correct and identifying any areas that we need to review and look at.

Another one you mentioned at the Classic Night Conference was the year's worth of queries for your accounting team and using AI to summarize that and work out where you can improve your processes. That was pretty cool.

Yeah, so we've gone through and downloaded all of the queries that were raised for every single one of our SMSF clients for the 2025 financial year and put that data into a model where we've gone and checked to make sure that where have we had any miscomings, which clients in particular or which pieces of information haven't we requested that we should be requesting upfront to make sure that we've got all the data that we need to start the job, to finish the job, from the start rather than continually going back to the client and asking for more.

And Jason has just given us where efficiencies are fitting in here. Are we actually seeing the productivity gains that AI has promised or is there still a gap between what we expect and what we're getting?

I think that people still believe that there is more to come. However, we are starting to see the productivity gains start to emerge. We're seeing it through the processes that Jason's just talked about where they're able to do more with less. And what we're seeing across the board is where practices are starting to adopt AI to assist with parts of their workflow and they're able to really get advantage from that. And so AI is not coming out and replacing jobs. Where it is coming is actually as an augmentator or as a partner for business professionals.

Jason Entwistle, let's get