Who’s talking
2 speakersTranscript
37 turns · 15 minWelcome to Fear and Greed Q&A, where we ask and answer questions about business, investing, economics, politics, and more. I'm Sean Aylmer. The AI boom is creating extraordinary demand for computing power, but that comes with its own set of challenges. AI infrastructure needs electricity, space and cooling. at a time when Australia is already grappling with constraints around power, water, electricity, etc. My guest today is David Keane, the co-founder and CEO of SCX.AI, which listed on the ASX last month, becoming Australia's first ASX-listed pure play AI infrastructure company. Rather than building new mega data centres, SCX installs specialized AI technology inside existing commercial data centers to extract more computing power while using fewer resources. SCX.AI is a supporter of fear and greed. David, welcome.
Great to be
here. Thank you
so
much for having me. You're going to have to dumb it down for me. All right, let's give it a go. What exactly does SCX.AI do?
Right. So we are building the sovereign AI inferencing capacity for Australia. Okay, what does that mean? Yeah. Well, everybody today that's listening to this is an AI user. I'm sure. You're all using it. People are using it every day. They're using it in their business. They're using it in their personal life. And when you use AI to perform work, our industry has created this name, inferencing. So inferencing is simply the word we use to describe AI doing work. So if you write an email to your boss asking for a pay rise, you're an inferencer. If you're asking it to look at a document and give you an opinion on a report, you're an inferencer. Just like Oprah said, we're all inferences. You're an inferencer, you're an inferencer, and you're an inferencer. So it is reality. That word has caused a bit of challenge for people to come across. But the best way I think about it is if you compare it with the other word our industry uses, which is training. People have heard about this concept of an AI training workload. So training is where you actually build one of those AI models. So again, you're creating the chat GPT or you're creating a cloud. That's where you get a whole lot of data. and you put that together with an algorithm, and you put that in a big building with tens of thousands of GPUs in it, and you close the door, and you come back three months later, and you have an AI model. But when you use the AI model to perform work, that's inferencing. So what does SCX do? So we are building the machines, the software, and the capability to allow Australians to perform AI work or inferencing here in Australia. So real world examples. If you're an organization and you want to use AI to perform an analysis of something happening in your business, you want to automate a workflow, that's an inferencing load. So you would come to SCX and you would buy, another lovely word, tokens. We'll talk about tokens. You'd come and buy tokens from SCX. And you'd use those tokens in your application or AI agent to perform work. And SCX is one of the first organizations to do that here in Australia. Because we believe Australians want to have their intelligence here. They want to control their
Episode notes
The AI boom has triggered enormous investment in data centres and the infrastructure needed to train artificial intelligence. But the next opportunity could be what happens every time we actually use AI – a process known as inferencing.
Sean Aylmer speaks with David Keane, co-founder and CEO of ASX-listed SCX.ai, about why he believes Australia needs its own AI inferencing capacity, the privacy implications of sending data offshore, and the possibility of Australia becoming an exporter of AI computing power.
SCX.ai is a supporter of Fear & Greed.
This is general information only, and you should seek advice tailored to your circumstances before making investment decisions.
Find out more: https://fearandgreed.com.au/
See omnystudio.com/listener for privacy information.