Workshop otherwise.

And the second one, are there any other good ideas of value which hasn't been tested? Now, normally that, again, it depends a bit on how many issues you're dealing with, but it's just one big critical issue.

There's normally not any further new ideas particularly.

Although I think it's a question that was one of the sessions there, you know, how useful is AI going to be as a watchdog? And I've just been downloading a whole lot of different AI models to try them out to see how well they work as standalone because the idea you're going to sort of put secure systems real-time online through a cloud AI – I don't think you're going to do that because that means that information knowledge becomes part of the AI cloud.

So the ones you're going to be looking for are the ones that are local, that are secure.

And I've got to say, I've been a very...

I mean, obviously we run Macs with the unified memory and we seem to have a bigger...

So it just makes it very tricky, doesn't it?

Yeah.

When you look at the current world and you're sort of looking at all the oil shocks and the hoothies and the different things happening in the Middle East and the things that are going on, you're sort of looking at a future uncertainty like that.

That is a completely different kind of uncertainty.

It's not just a mathematical one or the other.

And this is sort of then sort of explained in one way with this hallucination things they talk about or confabulations, as Geoffrey Hinton put it, from the AIs coming, because you get a lot of people sort of using AI more or less as truth than gospel, and sometimes it's not exactly precisely what's going on.

And the best way to explain it, I could think of it, was that if you think about it as a series of, I don't know if you remember about Markov changes, all the best, but basically it's a whole series of probabilities.

In a sense, you could think of it as fault trees integrating below.

It will give you your most likely outcome, a bit like Monte Carlo simulations will give you your most likely outcome.

But

I think the point, you know, if you ever wind up going to court, I mean, from our point of view, it's...

Too late.

Too late.

I mean, you've still got to go through the process.

But I was reading David Howarth, the professor of law and public policy.

We invited him to Melbourne before COVID, a long time ago now.

So by focusing on that – you can then focus on the controls that you're going to put in place, which is the key thing, and all your stakeholders are then on the same page.

But I thought what we'd do is we'd go through what people traditionally think of the hierarchy controls from the Code of Practice, how it also relates to safety and design and then the hierarchy controls in the WHS legislation, but then go to the elimination option.

Yeah, that's right.

And the reason why I talk about the hierarchy in the elimination option, from a criticality viewpoint, that's where you always want to start.

So what I thought I'd do is just summarise what the code of practice for OHS hierarchy control roughly is with a simple example.

So, I mean, most people remember we've got the elimination option, then you've got sort of substitution, isolation, engineering controls, then you've got admin and then PPE.

That's the normal way of thinking about it.

And that's been done for what you might call common hazards.

Yeah, that makes sense.

What steps did you take to make that a reality?

I mean, just looking at the numbers, you know, just inspired me and encouraged me.

Like there's less than 4% minorities doing what I do and less than 1% African-American men who do what I do.

So like, let's bridge that gap, right? Let's get these young African-American men and women and minorities in this profession and let's get them in the right spaces and places.

And that opportunity is not always given to people who look like me.

You know, so how can I get those people in the room, get them in front of the right people, you know, in the right professions? Because growing up, I just thought everybody in the hospital was a nurse or a doctor.

I didn't know they had respiratory therapists, physical therapists, dental hygienists, things like that.

That's is there something that she said or did or was what was going on in your mind that you were like, hey, this is this is what I want.

You know, this maybe not be my spot." But instead of saying, "This isn't my spot," you said, "Hey, this is my spot, and I'm going to do more with that and provide opportunities for other people to find their belonging in these spaces." What steps did you take to make that a reality?

reality?Um, I mean, just looking at the numbers, you know, just in- inspire me and encourage me.

Like, there's less than 4% minorities doing what I do, and less than 1% African American men who do what I do.

So, like, let's bridge that gap, right? Let's get these young African American men and women and minorities in this profession, and let's get them in the right spaces and places.

And that opportunity is not always given to people who look like me, you know? So how can I get those people in the room, get them in front of the right people, you know, in the right professions? Because growing up, I just thought everybody in the hospital was a nurse or a doctor.

I didn't know they had respiratory therapists, physical therapists, dental hygienists, things like that.

So it was eye-opening to me as I became an adult.

What was going on in your mind that you were like, "Hey, this is, this is what I want"?

And do you remember what your GPA was when you applied?

When I applied, oh, wow.

Okay, so real quick.

A lot of people don't know that I have severe dyslexia and severe test anxiety.

So I didn't learn how to read and comprehend until I was 36, Dana.

Wow.

Yeah.

Tell us about it.

How is it swinging right now? And then how does a product like th- the ones Robin.AI are putting out lead society, lead all of this in a better direction?

I think a lot of this comes down to validation.

Sam Altman said something that I thought was really insightful.

He said that the algorithms that power most of our social media platforms, X, Facebook, Instagram, they're the first example of what AI practitioners call misaligned AI at scale.

And what we mean by that is systems where the AI models are not actually helping achieve goals that are good for humanity.

So this, the algorithms in these systems, this is before ChatGPT, but they are using machine learning to work out what kind of content to surface for people and it...

so it turns out people are entertained by really outrageous, (laughs) really extreme content.

Are we perhaps creating a new class of problems or overlooking some areas even as these brilliant systems are coming online?