The Battle to Control AI: What Australia Does Now

Every government in the world is now trying to answer the same question: how do you control a technology that moves faster than any parliament can? In Australia the debate has arrived as a proposed AI kill switch, a push from MPs across the political spectrum, and a Prime Minister meeting the technology companies in the United States.

Travis Carter and I spent this week's Greater Wealth on what that debate gets right, what it misses, and what it means for your job and your business.

In this episode (39 chapters)

1. 00:00The battle to control AI
2. 00:09Child protection, censorship, and the crossroads Australia is at
3. 00:49Albanese in the United States, and the welcome that was not
4. 01:51Tim Cook, and 20 WA MPs from eight parties write to the PM
5. 02:47What if the technology outruns our ability to control it?
6. 02:59The two levers: regulate too slowly, or too aggressively
7. 04:07The AI kill switch and data centre control bill
8. 04:26Don't build a kill switch. Build a counter-AI
9. 04:50What the Apple meeting says about who protects whom
10. 05:52Right of repair, and what Apple would want in return
11. 06:49From 12th to 52nd: what happened to our internet
12. 07:27Why would developers ask to be legislated?
13. 07:53The Clarity Act, and the lawsuit nobody wants
14. 08:23Live from the chat: two questions from Aaron
15. 08:57When government depends on the technology it regulates
16. 09:15Have you ever used a government website?
17. 09:25The space race is now the AI race
18. 10:21The comparison with the nuclear arms race
19. 10:43The AI the public can see is years behind the military
20. 11:36Rules, and the players who will ignore them
21. 12:21Be proactive: an AI that hunts the scammers
22. 13:13What this means for Australian jobs and business
23. 14:45If you can imagine it, you can probably build it
24. 15:39How a radio station actually uses it
25. 17:25The best thing about AI: no ads
26. 18:43The real dividing line
27. 19:23What people in the city actually say about AI
28. 20:23Can small business keep up with a $10,000-a-day budget?
29. 21:03Making a small business look bigger than it is
30. 21:35Case study: the Clarity Act fails in the Senate
31. 23:09Stablecoins, the banks, and a Bitcoin senator
32. 24:25Get on board, or don't
33. 25:28Is theft bad? The analogy
34. 26:39How do you stop an AI that doesn't want to be found?
35. 26:59Use AI to stop AI
36. 27:49The countries that opt out
37. 29:36Four things you can do now
38. 30:19What Blu Sky Factory actually teaches
39. 32:49Final thought: do your life purpose

What actually happened

Two levers, and no easy answer

On the show I described it as two levers Australia is wrestling with. "If we regulate too slowly, Aussies may be exposed to general and safety, privacy, employment risks, and that sort of thing. If we regulate too aggressively, Australia could fall further behind the countries building their technology."

There's an uncomfortable truth underneath that. Governments don't build this technology, and they don't understand it as well as the people who do. "Let's face it, governments don't have the systems and infrastructure that technology companies do in these sort of spaces." When a government asks the technology companies to handle online safety for it, it's quietly admitting the problem is bigger than it is.

Why a kill switch won't save us

Travis took the kill switch apart, and not from the side you'd expect. His objection isn't that it's too strong. It's that it only works once: "once you hit that kill switch, it's done. It's like everything's gone. Then you're leaving yourself totally vulnerable."

His alternative: "I think that Australia needs to develop an AI that is purely designed to stop other AI from interfering." He made the point with a simple question. Is theft bad? Yes. Have we outlawed it? Yes. Do people still do it? Of course. Banning a technology doesn't stop the people prepared to misuse it. In his words, "We need to use AI to stop AI."

The same logic applies to scams. Rather than only writing rules the scammers will ignore, Travis's proposal was to be proactive: an AI that actively goes out and hunts down scam operations before they reach anyone.

What it means for Australian jobs and business

Here's where the conversation turned practical, because AI is no longer confined to technology companies. It's in customer service, marketing, content, accounting, financial analysis, software, administration, legal research, healthcare support and business planning.

And it's changed who can build things. As I said on the show, "if you could imagine it, there's a good probability that you can create it." I gave an example from my own business. Setting up a custom payment plan for a client "used to take me about half an hour to set up for a client." Over a weekend I built a tool that suggests four or five options in front of the client and sets the plan up on the spot.

Travis uses AI every day at Edge Radio, from building playlists for new presenters to turning a sales proposal around: "20 seconds later, it's done."

The real dividing line

This is the line the whole episode turns on. People ask whether AI will take their job. On the show I put it this way: "That's the dividing line. People are worried, am I going to lose my job or not? And we're going, look, it's going to go to those who know how to operate it and use it."

Travis hears the same thing in the city. Talking to lawyers and business people, he's "yet to find someone who's worried about their job, but they're worried about how their job's going to change." That's the right worry to have.

Why small business has the advantage

Big companies have the budgets. Some spend thousands of dollars a day on AI computing alone. But smaller businesses have something money can't buy. As Travis put it, "if you're smaller, you can pivot quicker. If you're bigger, it takes longer for the ship to turn around." AI also gives a small business "the ability to appear much larger than what they are."

That's why I said "A lot of the thing about AI is mindset." It's another tool in the toolbox. The question is simply: what can this do for me?

Four things to do now

I closed the show with four practical steps:

1. Educate yourself, and understand the structural change that's going on.
2. Avoid reacting solely from fear. That emotion can stop you taking the next step. "Please, whatever you do, do not go, oh, AI doesn't affect me, so I'm not going to learn about it."
3. Build relevant skill sets.
4. Be prepared before the consequences become unavoidable. It's better to be three steps ahead than seven steps behind.

And the thought behind all four: "Don't be fearful of these things, these changes that are going on. It's important to go, okay, the world's changing. How do I pivot? How does AI give me the opportunity to do my life purpose better and with more efficiency?"

Three questions to ask this week

1. Which task in my week takes half an hour that AI could do in minutes?
2. If a competitor half my size started using AI tomorrow, what could they offer that I can't?
3. What am I waiting for before I learn it properly?

If the honest answer to the third is "nothing", the free masterclass is a good place to start.

Education only. Nothing here is personal financial, tax or legal advice.