Trellis Data's Michael Gately on sovereign AI and selling to govt

So talk to me about trust, I guess.

I mean, right now, actually, we might as well go here now.

We're recording this on the morning after Prime Minister Albanese has just talked about OpenAI having hacked into the Medicare system.

And obviously he was somewhat stroppy about that and got on the phone to Sam Altman.

So it's a huge story this morning, and I'm assuming it's going to be a pretty big story over the next several days.

We'll see.

But it goes to the heart of trust in these systems.

So just talk me through how Trellis approaches that.

Yeah.

So I think there's a lot of noise out there in the AI market and it's difficult for some people to sort of separate what's real from perception.

And what I mean by that, trust is a classic case in point.

Sovereignty, we hear about sovereign AI, we hear about all these things and we are a sovereign AI company for Australia, of course.

And And there's a lot with trust that makes AI difficult to really adopt at a business layer.

And you're talking intellectual property, customer contracts, the very things that make a business grow and people get paid from.

And when adopting AI, there has been this, well, we need frontier AI or we need some sort of heavy integrated AI, whatever that looks like.

without necessarily the knowledge of where that data is going or where the AI processing is occurring.

I think that is creating an anxiety in businesses for adoption.

When we see news articles like the one overnight with the Prime Minister and OpenAI, it does lead to more distrust, I would imagine.

And so when we're saying to dentists, you can trust us, It's difficult for that message to cut through, I think, because it's another AI platform.

That's how the perception is to some degree.

And so it's really incumbent on Trellis data to really get the message out about what does it mean to trust AI and not AI? What does it mean to trust Agentes? And so that idea that you can be completely safe and that your data will never be looked at, never.

I think that is a very persuasive argument that we're hoping to cut through into the marketplace.

Because if you can trust it, if you actually can trust it completely, which you can with Agentes, then you can start experimenting.

You can start using the AI to do things that maybe you wouldn't have tried before because you're worried about the actual finances from zero or the actual contracts of your customers or your intellectual property and research and product roadmaps or all those things can actually now go into Agentes.

And nothing's changed from when they weren't in there.