EDITORIAL: Don’t let AI Medicare scare knock us off path to future

Fear of the rise of the machines is nothing new.

It is a storyline that has been extensively mined by filmmakers at least as far back as 1968 when Stanley Kubrick’s 2001: A Space Odyssey, unleashed HAL, the computer which turned on the astronauts on board the spaceship Discovery.

And there are regular displays of the progress being made in other futuristic fields like robotics — such as was illustrated extraordinarily just last month when a Chinese humanoid robot ran 100m faster than sprint great Usain Bolt’s world record time.

It’s hilarious footage too, ending in an undignified tangle of robot bits when two of them crashed into a barrier to bring the race to a halt.

But what’s not funny is the news that a rogue Open AI agent accessed “non-public files” in Australia’s universal health insurance agency, Medicare.

The US-based AI firm confirmed the unauthorised access to Service Australia’s data portal after an AI agent acted independently while conducting health research.

The agent also interacted with sites belonging to the Victorian Department of Health, NSW Bureau of Crime Statistics and the Australian Institute of Health and Welfare.

The tech company has admitted its models “took actions we did not intend”.

The incident raises a number of questions.

They include the timelines around the company learning of the breach and alerting the Government, and the Government making it public.

The company claimed it only found out about the June 18 breach in August before contacting the Government through a public facing general email address on September 10.

The Government says Services Australia checked the inbox the following day and picked up the email.

And yet the information was only made public as Prime Minister Anthony Albanese was in New York this week pushing his quest to regulate AI and social media algorithms — a concept slapped down vigorously by President Donald Trump.

So the timing of the information release looks political, to say the least.

This seems to be backed up by the fact that Mr Albanese expressed Australia’s “extreme concern” to OpenAI, while back on home soil, Defence Minister Richard Marles insisted the impact was “relatively minor”.

While we don’t need to buy into any scare campaigns or catastrophising, the point needs to be made about what seems to be a glaring unpreparedness of government technology and systems.

Where was our cyber security defence?

It seems archaic that, as revealed by Government Services Minister Katy Gallagher, the email address used by OpenAI to notify the Government was checked “once a day”.

Really? In 2026?

What other areas might be similarly vulnerable?

This is not a good sign as we move quickly into the AI era, which offers great possibilities.

The Federal Government’s Intergenerational Report — ”Australia’s future to 2066” — released this week, said AI would “be a defining influence on the economy over the next 40 years”.

Let’s face it, with Australia’s dire productivity performance, AI is our big hope for the future.

We have to extract the best out of this emerging technology and work out how best to use it.

And fast.

Responsibility for this editorial comment is taken by Editor-in-Chief Christopher Dore