Sounding the alarm on our terror alerts 

ASIO chief Mike Burgess recently revealed he is working with the Home Affairs department to overhaul our terror threat warning system introduced in 2014. Burgess says that the current threat level of “probable” underestimates the seriousness of the danger we face. ASIO has foiled 14 terror plots since Bondi.

Our five-level scale is not expected, possible, probable, expected, and certain. Any terror alert system isn’t a perfect science. A “not expected” alert isn’t a guarantee that a terrorist attack won’t happen. An “expected” alert is no guarantee that an attack will occur. Absolute security from terrorism through any national public warning system remains a pipedream: the intelligence driving it will always contain imprecision.

But our alert system can psychologically prepare the public by giving people an assessment at a particular time. It can increase the number of eyes and ears helping our security and law enforcement agencies. It can remind people to be vigilant and report suspicious activity. In Israel, there’s no formal alert system. The threat level would almost always be “imminent”, with a terrorist attack assessed as likely. Alerts tend to be in response to specific threats.

As far as Australia is concerned, our national threat alert system can be improved with some modest changes. While continuing to maintain a national approach — most state police websites on terror alerts refer the public back to the Australian government’s national security website for the alert level — if there’s a specific and credible threat, then an alert could be issued for that specific region, perhaps a state or a major city.

The Department of Foreign Affairs and Trade provides public information on risks in different countries where Australians might travel. The public is familiar with the “Smartraveller” system. We could extend that logic in our terror alert system inside Australia. Our security agencies and police would naturally, however, need to be mindful that an alert for a specific area may incline a terror cell to relocate to an area where the counter-terrorism focus might be lower.

We could consider changes to the duration of alerts. Authorities lowered our alert level to “possible” in November 2022. But unlike weather alerts, terror alerts aren’t backed by scientific data indicating that a threat has reached its peak and declined. Lowering an alert may run the risk of lulling the public into a false sense of security.

But a sunset clause – providing the mandatory expiration of a raised level after six months – would allow the level to be lowered objectively. It would provide enough time for intelligence and law enforcement agencies to assess the situation. This would reduce the risk that an elevated level just becomes the “new normal”. The mandatory lowering of an alert after a certain period, unless there’s evidence that it shouldn’t be changed, would reduce the costs associated with remaining at elevated alert levels, such as the costs to critical infrastructure operators and police.

There’s a need to give the public more narrative around each alert level. The description on the government’s website states that “probable” means “there’s a greater than fifty percent chance of an onshore attack or attack planning in the next twelve months”. But when you visit the official website for the “National Terrorism Advisory System” there’s no narrative provided at all on what each alert level means. “Probable’ is simply defined as “the government has concerns of a plausible threat”. Without explanation, the threat levels are likely to leave Australians alarmed but uninformed. Dealing with Burgess’s concern that we’re now at the upper end of “probable doesn’t require introducing another threat level to the scale. But it means we should raise the percentage chances of an onshore attack from fifty percent in the current description (a coin toss) to a higher percentage.

Finally, the advisory system website provides no guidance to citizens about what to do at each level. One way to test the language would be to look at warnings used by our meteorological advisory and even bushfire advisory systems. Obviously, terrorism warnings aren’t as scientific as weather alerts. They rely on classified information, which must be protected. It’s much more certain when a weather threat is looming and when it’s passed. But the public is used to acting upon natural hazard warnings.

We should be able to do better than our political leaders telling us to be aware, but not afraid. The government should be starting to think through what being secure from or during a terror attack might look like in terms of practical advice. We have bushfire plans. They will vary depending on location. Perhaps the government should be developing a terrorism personal and community safety plan giving the public specific steps they can take. Other nations, such as Singapore, do this. Our authorities should be able to do so.

Dr Anthony Bergin is an expert associate at the National Security College.