Top cyber spy warns against dependence on artificial intelligence we don’t understand
Source – smh.com.au
Australia’s top cyber spy says the world needs to think more about risks created by over-reliance on artificial intelligence so that people don’t “sleepwalk” into dependence on machines they don’t actually understand.
In an rare public speech, Mike Burgess, Director-General of the Australian Signals Directorate, also said spy agencies around the world – including his own – were increasingly using artificial intelligence to carry out their covert activities.
“It is right the world embraces artificial intelligence, but we must embrace this with our eyes wide open. We should not sleepwalk into this, where we suddenly find ourselves in the world that is controlled by software and very few people understand how it works,” Mr Burgess told a conference on artificial intelligence hosted by the Australian Strategic Policy Institute.
“How much of our world will be outsourced AI? How much of our brain power and decision-making will we hand over?
“It’s great … for productivity and the economy and society, but with those same great benefits come serious risks that require serious thinking … and I don’t think we’ve done enough of that thinking to date.”
Artificial intelligence, or AI, includes machine learning, data analytics, data mining and automation. The main uses revolve around solving problems by analysing large amounts of data to pick patterns that a human either couldn’t spot or would take much longer to do.
Mr Burgess said he had seen “unsupervised machine learning” operate very effectively.
His agency gathers electronic or “signals” intelligence, protects Australian government computers and attacks foreign networks to support military operations or to counter terrorism and serious cyber crime. It has its own AI specialists, he said.
“ASD will use artificial intelligence to maintain our capability edge, to defend Australia from rogue threats and help advance Australia’s national interests.
“In the spirit of transparency, there wouldn’t be an intelligence agency on this planet that would not be thinking today about how AI could be exploited, ASD included.”
Mr Burgess said artificial intelligence could be used to defend computer networks but also to “find vulnerabilities or find chinks in the armour” of rival networks.
Peter Singer, a world-renowned expert in technology and national security at the United States think tank New America, told the conference that out of the Fortune 500 companies in the US, 244 listed AI as “the core to their future business strategy”.
“Besides being a business arms race, it’s a governmental arms race.”
China had declared its ambition to become the world leader in AI by 2030, whereas the US government was leaving progress to the private sector.
Mr Singer said the rapidly growing “Internet of Things”, made up of web-connected devices, would triple in the next five years.
Most countries were woefully underprepared for the consequences, he said.
“The responsibility for security is unclear. There is, at least in the United States and most western nations, no regulation. Even basic liability hasn’t been figured out. Too often the devices lack even basic security features,” he said.
These ranged from “hackable trucks to hackable barbie toys”.
But he said the politics and legal questions would likely sharpen quickly once cyber attacks – which in the past have previously tended to revolve around data – would start to affect real-life objects.