SABRA LANE: The Federal Industry and Science Minister Ed Husic has signed Australia onto an international agreement on artificial intelligence while attending a London summit on AI safety. It's known as the Bletchley Declaration. Bletchley Park is where the conference is being held. It's also where during World War II, code breakers cracked German coded messages. A conference is happening in the same week that the US President signed an executive order with rules for safe and responsible use of AI technology. Mr Husic joined me earlier.

Minister, this agreement says there's potential for serious, even catastrophic harm, either deliberate or unintentional stemming from AI models, and that understanding these potential risks and actions are urgent. But there aren't any policy goals with this document. Is that a bit of a problem?

ED HUSIC, MINISTER FOR INDUSTRY AND SCIENCE: I think what's happening here is a pretty seismic shift in thinking around technology. One, it's really rare that you'd have events like this bringing governments together to think about regulating technology. Usually, it's the case technology spans the world and people – government’s reluctant to do anything. You've had people brought together in this way to consider that and underneath it is this major movement by the US bringing an executive order to boost AI safety and security. I've previously said that the days of self-regulation for technology are over, and this summit confirms it. Expecting that firms, when they design, develop, deploy and use AI, that they'll do it in a much more safe and trustworthy manner. So, it's a big deal.

SABRA LANE: Well, how soon will there be action on this because already deep fake images are circulating online? People have been denied insurance based on faulty algorithms. Governments are already too late, aren't they?

ED HUSIC: I think there's a determination to start taking action in a number of different ways. One, you've seen coming out of this – decisions by the US and UK Governments to set up safety institutes, to get researchers involved in looking at the models that are used by some of these companies and to try and improve them. There is also going to be work on what's called the State of the Science Report. Australia will have, from the CSIRO, its chief scientist Dr Bronwyn Fox joining other researchers to examine what's being done to improve the safety and security of AI. But there's definitely one of the things that's been expressed, Sabra, is we do need to take steps to counteract some of these concerns that you've raised.

SABRA LANE: As you've pointed out, the US President has acted on this, Europe's taken action. The Federal Government is still consulting on AI regulation. You've had hundreds of submissions and you've actually met too with Sam Altman, the brains behind that ChatGPT. He's pleaded with governments to act, saying it poses a threat to humanity. When will you act, why is Australia taking so long?

ED HUSIC: A number of things. One, while the US and UK are setting up these safety institutes, we've also, through the CSIRO, established a responsible AI network and that is bringing industry, academia, governments together to look at ways in which to improve the safety and trustworthiness of AI. We started the consultations around safe and responsible AI regulations. We've had broad agreement that voluntary commitments won't be enough. Sure, there are different views on the best form of regulation, we're working through that, but there's some that's going to add momentum to the work that we're doing. And importantly, it ensures that we'll have governments working in similar ways to try and improve the safety and responsible use of AI instead of seeing some go – probably not do anything, others go too far. 

I think there's a broad agreement by governments that we do need to do a lot more safety testing and auditing of AI and that we can't just leave the companies to themselves on this, or as it's been described, that they mark their own homework. And that's a really important sign too, saying to companies that they are going to need to step up and work out how to satisfy the concerns of communities as expressed through governments.

SABRA LANE: All right, what are we going to see regulation or laws passed in Australia to deal with this? Will it be in the next six months?

ED HUSIC: So – well, we're working through that and obviously the summit itself gives us the ability to take on board some of the pathways that are being described and the way in which we can work together on that. So, we're hoping to get that done in the very near future. But clearly taking on board what at the conclusion of the summit today, we've got some things that we can factor in, particularly around safety testing of AI locally. There'll be some work that we'll be examining. 

We do need to get the balance right on this. At its best, AI can help improve the quality of life and improve the way our economy functions. At its worst, it can distort and divide. I don't think really that the worry is that the robots will take over, but that potentially AI-generated misinformation will. And so, we recognise the risks that are there. If we get this right, we build trust in the use of the technology. That'll be really good but stamping out some of these things will be really important as well.

SABRA LANE: If I could ask you about Gaza. Several countries, including Australia, have warned Israel that civilian deaths are causing enormous concern. The US President again has asked for a pause to allow civilians to leave, for hostages, possibly to get out and for aid deliveries. From this distance, it doesn't appear to be having much impact. You've already voiced concerns. What are the consequences for Israel if it doesn't heed warnings from its allies?

ED HUSIC: I was worried very early on about where things would head and the impact that would be borne by innocent Palestinian families. And I think it has been a very sizeable impact in terms of lives that have been lost, and particularly around kids. And I think a lot of people in Australia are genuinely worried by that. And look, I think the Foreign Minister is managing the bulk of the Government's response on this, but this is a real issue in terms of potential overreach, and I think the international community is asking – pleading for moderation and hoping that that happens.

SABRA LANE: Minister, thanks for talking to AM this morning.

ED HUSIC: Thank you.

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