COVIDSafe: the failure of an app

Image: Pixabay

The returns have not been impressive. For an app promoted as a saviour for contact tracing purposes in the worst pandemic in a century, COVIDSafe is a lesson in exaggerated prowess and diminished performance.

It was billed as necessary and supremely useful. Prime Minister Scott Morrison was fashionably vulgar in linking the use of the app with an important goal: Getting watering holes opened.

“If that isn’t an incentive for Australians to download COVIDSafe, I don’t know what is”, Morrison claimed in May last year. The prime minister even equated the use of the app to wearing sunscreen: “If you want to go outside when the sun is shining, you have got to put sunscreen on. This is the same thing”.

Numerous organisations across Australia, private and public, jumped to the call in a concerted effort to embrace surveillance in order to suppress the virus.

The National Waste and Recycling Industry Council, to name one enthusiastic participant, claimed that the COVIDSafe App, “together with good hygiene, safe work practices and social distancing will help protect the health of waste and recycling staff and our committee”. Seven million people also seemed to agree in downloading the app.

But problems began to burgeon.

Identifying close contacts did not seem to be the app’s forte. There were issues with its performance on Apple devices. The iOS version, for instance, had a habit of trying to link with each device a user’s phone had ever been connected to.

New South Wales Health Minister Brad Hazzard had to concede in October last year that COVIDSafe had “obviously not worked as well as we had hoped”. To this could be added that hovering cloud of privacy concerns and trust in government.

The review by consultants Abt Associates, commissioned by the federal government and delivered in March this year, did not make for pretty reading.

It took the persistence of The Canberra Times to obtain a more complete version of the report which had initially been released in severely redacted form. (The redacted evaluation was rather more glowing in concluding that “the COVIDSafe app was the correct tool to employ".)

The app, it was found, had added to the workloads of state tracing teams “without optimisation of benefits”. This included, for example, the presence of false close contacts among 205 individuals flagged by the app (some 61%), with 30% already identified through standard contact tracing.

A mere 2% of total close contacts in NSW between March and November had been detected by the app, with none in Victoria and Queensland.

COVIDSafe was costing as much as $100,000 a month to continue its operation by May.

Randall Brugeaud, chief of the Digital Transformation Agency, explained to Senate estimates, that the amount could increase to $200,000 a month “to allow us to make future changes”.

These were additional totals to the $6.7 million the app had already cost till that point, most of which had been swallowed by development outlays.

Such costs are far from draw dropping, given the number of partners this crowded venture has involved. The Amazon Web Services (AWS) platform may well claim to have led the pack, but then came Delv, Boston Consulting Group and the Melbourne-based outfit Shine Solutions.

It was reported in November last year that Shine Solutions had received $350,000 in increased payments from the government, bringing the then total to $2 million for work on an app that had, as yet, to identify any new close contact anywhere other than in the state of New South Wales.

Despite an entrenched stubbornness, a Department of Health report in July was forced to admit that it was “rare” for contract tracers to use COVIDSafe. Moreover, a mere 779 people who tested positive for COVID-19 had availed themselves of the app.

COVIDSafe keeps company with several similar tracing apps that have been left behind in the wake of renewed infection waves and the discovery of vaccines. But that does not mean that their advocates have not stopped their fires of conviction.

A co-authored contribution in the Journal of Medical Ethics insists that we should not have too many hang-ups on overly centralised data in terms of what risks it poses to privacy. Decentralised systems, argue the authors, are also inefficient and risky.

Any identifiable moral here must lie in the risk posed by zeal. COVIDSafe never replaced the sleuthing efforts of human contact tracers and may have even inhibited them.

“The lure of automating the painstaking process of contact tracing is apparent,” a co-authored report from the Brookings Institute asserts. “But to date, no one has demonstrated that it’s possible to do so reliably despite numerous concurrent attempts”.

The authors even suggest that such apps can “serve as vehicles for abuse and disinformation, while providing a false sense of security to justify reopening local and national economies well before it is safe to do so”.

Even now, on its long feted deathbed (and keeping company with a range of other feats of government misspending) you can still go to the Department of Health website to see the following: “The COVIDSafe app is a tool that helps identify people exposed to coronavirus (COVID-19). This helps us support and protect you, your friends and family”.

The belief in the role of this tool remains steadfast. The eyes, in the Morrison lingo, remain on the pub prize.

[Dr Binoy Kampmark lectures at RMIT University, Melbourne. He can be emailed at]