Should AI Be Regulated By States Or National Laws?

You've probably heard a fair bit about Artificial Intelligence lately — it's popping up everywhere. But who gets to set the rules for this powerful new technology? That's a big question, and even the big tech companies are weighing in.

Recently, Anthropic, one of the leading AI developers, spoke up about how the US government should handle AI laws. They're making a case that if individual states start creating their own rules for AI, it could create a confusing mess. Imagine a world where an AI tool is legal in Queensland but not in New South Wales – it would be a headache for businesses and developers alike, and might even slow down innovation.

Anthropic believes that if Congress, the US national parliament, wants to stop states from making their own AI laws, then Congress itself needs to step up and create strong, nationwide standards. Their main concern is about managing what they call “catastrophic AI risks.” This usually refers to the very powerful AI systems that could potentially cause widespread harm if not built and used responsibly. They're basically saying, don't just stop the states; if you do, replace it with something solid at the national level.

This debate in the US is important for us Aussies to watch. We're also grappling with how best to regulate AI, and whether it should be a federal responsibility or if states and territories should have a say. Getting it right means finding a balance that protects us from potential harms while also allowing Australian businesses and researchers to make the most of what AI has to offer. It's a tricky balance to strike.