The power of artificial intelligence will shape how all of us live and work, yet this breakthrough technology is still nascent in Australia and the promises and benefits are yet to be seen.
As with all new technologies, there has been an enormous amount of hype with AI and with few companies demonstrating any benefits from its use, a clear path for success hasn’t emerged here.
Australia is dramatically underinvested in AI technology, compared with the rest of the developed world, says AMP Capital investment manager Alistair Rew, who will speak on this topic at the Investment Magazine Investment Operations Conference, to be held on February 26 in Sydney.
In Rew’s view, global the adoption of AI has reached an inflection point, moving from innovators and early adopters to the early mainstream – and there is a danger that Australia will be left behind.
“The Australian market hasn’t made as many attempts to make AI consumer-facing products,” the investment specialist notes.
Daisee founder and chief executive Richard Kimber, who will also speak at the conference, adds: “The US has a…head start with its tech giants. Amazon and Google DeepMind are driving the front-end consumer aspect of AI.”
Kimber’s message is clear: AI technology is important because, increasingly, it enables human capabilities – understanding, reasoning, planning, communication and perception – to be undertaken by software, effectively, efficiently and at low cost.
To him, Australia’s reluctance to commit to AI is somewhat ironic since, he notes, Australia has been at the forefront of numerous new technological waves over the years. In fact, he noted, wifi was invented and commercialised here.
But over time, he went on to say, while Australia’s academic research has been strong, it has struggled to turn the research into a commercial application.
“We have a lot of people dabbling in AI without many people committing – I think what we are seeing is risk aversion.”
Daisee’s figures show a lack of funding for early-stage tech businesses; for example, According to Daisee research, the UK is generating about 400 AI start-ups a year, compared with about a dozen or so in Australia.
It is estimated Canberra allocates just $29 million annually to AI. This is a small fraction of the ₤100 million Westminster allocates each year, and even less of the US$1 billion Beijing reportedly spends.
But the story is much bigger than how well Australia is investing its resources. From where Kimber sits, corporate Australia faces a major structural challenge.
While companies have a massive opportunity to boost competitiveness with new technology, the investment community has failed to recognise AI’s importance.
“I don’t think people appreciate the quantum of breakthrough brought about by a confluence of events – the cost of computing, the cost of storage and the access to data,” Kimber says.
Typically, a company’s data analytics team sits in finance or operations and, as such, the data scientists are fixated on the past, rather than looking at the future, he argues.
“The power of AI is prediction. It has enormous potential in the revenue drivers of business segmentation, marketing and sales, yet very few companies are seizing the benefits.”
For instance, AI could have a massive impact on the superannuation and wealth sector by completely changing the scale of information that can be synthesised by several orders of magnitude, Kimber says.
“Investment management firms that are able to create information assets will be able to spot trends earlier, identify anomalies more easily and predict market movement far more accurately and quickly than firms that don’t embrace the powers of machine learning and data models.”
As he sees it, ultimately, investment management is an information business and machine learning offers a tool set that will fundamentally change the way people work with unstructured and structured data.
He cited earnings calls – where management give verbal updates on companies – as an example.
“AI-powered systems will be able to analyse the vast number of these calls and overlay them with many other data sets to identify cues and insights that others would simply not be able to process.”