Interview with Sabra Lane, ABC AM

Sabra Lane
Artificial Intelligence; ChatGTP; Deepfakes

SABRA LANE: The Federal Government's taking its first steps towards regulating artificial intelligence, saying new safeguards are needed to protect us. AI is developing rapidly with advice to governments saying it'll affect everything from banking to education and creative industries, while others are warning it could lead to human extinction. The Federal Government's releasing a discussion paper today and the Minister for Industry and Science, Ed Husic joined me earlier. Ed Husic, welcome to AM.


SABRA LANE: Artificial intelligence makes it so hard to know what's real. There's so much misinformation. How can the government beef up regulation?

ED HUSIC: I think everyone recognises AI has become a part of their lives and they do obviously appreciate some of its benefits. But I think there's sort of a feeling in the community that they want to have the assurance, the confidence that the technology isn't getting ahead of itself and that's not being used in a way that creates disadvantage or risk for people. And that's why as a government, the Albanese government wants to basically set up on the next reforms that can give people confidence that we're curbing the risks, maximising the benefits and giving people that, as I said, that assurance that the technology is working for us and not the other way around.

SABRA LANE: What do you have in mind? The discussion paper that's been released flags things like facial recognition and social scoring. China uses those things. Would they be banned here?

ED HUSIC: We're opening it up for broad discussion. We want, as we shape up the next reforms into the way in which AI gets regulated in this country, we want to invite not just people that are involved in technology, but we want the broader public and others to have their say as well. 

There's probably, Sabra, close to about a dozen different laws at the moment that in one way or other have some give some regard to the impact of technology, particularly like AI. But whether or not we're able to say with full confidence that those laws are keeping up with technology and the developments that we've seen over the last six months, for example, well, that's why we're doing the consultation. 
We want to make sure that Australia has got a fit-for-purpose regulatory regime when it comes to AI, particularly what we're seeing around generative AI. And again, the aim is to give people confidence that we are curbing the risk and maximising the benefit of new technology.

SABRA LANE: As you say, the technology is racing ahead. Have public schools been too quick to ban ChatGPT?

ED HUSIC: Well, my colleague and friend Jason Clare the other day has basically referred to the House of Representatives Education Committee an inquiry into the use of AI in education. And I do know that states and territories, they're already starting to think about this as well, Sabra. And again it's about recognising, there are in some cases, people use ChatGPT to help enhance the way they do their research. 

But we've also got to ensure that the way that that is done is proper, that people are confident and that people are comfortable with the way that that's being used. And in an education sense you want kids to go through the full education learning process and to be able to develop those skills within them without necessarily relying heavily on the technologies.

SABRA LANE: Do you think companies and people should be allowed to create and muck around with deep fakes given fake images and disinformation are pretty rife now, I think.

ED HUSIC: That's a big concern that people have. They want to have the assurance that what they see on their screens is legitimate, particularly from a news perspective, completely understand from an entertainment - if people are fully aware, they're fully conscious of the fact that something might be a fake, but it's used in a particular context that might be about entertainment, then that's a totally different thing. 

But the way in which misinformation might influence the way in which we as a community make our decisions about things, and particularly from a parliamentary perspective that might be guided in that way, you ought to think carefully about it. And so online safety and within the realm of the Minister for Communications, Michelle Rowland, the eSafety Commissioner, is thinking a lot around that and working with companies on that. But that is clearly something that again is another example of the technology being used in a way that has outcomes that people won't be comfortable with whatsoever. And we do need to have some sort of regulatory response to deal with it.

SABRA LANE: Hundreds of tech chiefs, including the bosses of some of the biggest AI companies, yesterday released a statement warning that this could lead to the extinction of humanity. Should be - people be frightened and on alert over this?

ED HUSIC: Not be careful about - look, my take Sabra, I mean, you may be aware I followed technology for years, well before -

SABRA LANE: A long time.

ED HUSIC: - entering into politics. I always take this view, I'm not an evangelist about it and I don't catastrophise. Technology will be used in a range of different ways. The main thing is we want to use it for benefit rather than ill. They want the assurance, the confidence that those type of risks are being, one, recognised and something's being done about it, but not in a way that throttles the use of the technology because this can have benefits. 

I mean AI, for example, if you've got a choice, you're worried about your health, you're concerned you might potentially have cancer, you've had your screening done, and you can use AI to be able to draw off hundreds of thousands of images that have been able to detect trends in the way that cancer has evolved and that you can get a definite diagnosis within a short space of time, you'd want AI to do that. 

But at the same time, would you want that diagnosis just to be spat out on a receipt out of a machine? Or would you want someone to walk you through that? And this is the classic, I guess, case here, Sabra, that we want the technology to be able to deliver for us, but we want people, the people to people interaction that helps gives us that comfort and assurance about how the technology is being used. And it's, in a sense, a double check to make sure that everything is okay. So, we got to get that balance right. That's the big thing here.

SABRA LANE: Ed Husic. Thanks for talking to AM.

ED HUSIC: Thank you. Sabra.

SABRA LANE: That's Ed Husic, the Minister for Industry and Science.