Dr. Ruth Buckley

And that's what a lot of people are doing because we want to encourage innovation, Jez.

You don't want to be discouraging people who are coming up with new ideas and new ways.

But at the same time, it has to be done in a sandbox, in a safe space.

And I think that's really important that you have to look at safety first.

We do not want data leakage.

And I think that's the big concern for organisations, whether you're a school, whether you're

you know, a hospital.

There's a really good diagram in the EAU AI Act that has a pyramid of risk and it shows all the different uses for AI, but it also risk profiles them.

So, for instance, you won't be surprised to hear that if it's been used in the health sector, that's seen as high risk.

Whereas generative AI is at the large language models, is at the very end of the pyramid.

So I think risk is really important to consider for an organisation.

But then at the same time, we do need to empower people and enable people.

I can tell you, Jez, what's really important for us is that we've invested in training.

And a lot of government departments, including my own organisation, we don't want anybody experimenting with AI unless you've been trained.

And that's a really important principle.

It's a fundamental principle in the EU AI Act.

If you're going to use this, you must make sure that your staff are trained.

So we've done that.

And we look, we're certainly not there yet.

We're in a process.