The Morrison government was warned of serious problems with a tool designed to predict future crime before using it on 14 occasions over three years, including to justify the use of extraordinary preventative detention powers for terrorist offenders who had finished their sentences.
Guardian Australia can also reveal that, in three separate cases, the government failed to disclose to defendants that it knew of deep flaws with the tool being used to assess their likelihood of future offending.
Last week, the independent national security watchdog called for the abolishment of a regime allowing for the detention of convicted terrorists to prevent future crime, which relies heavily on a risk assessment tool known as VERA-2R.
In May 2020, the government received an independent report it had commissioned into the risk assessment tool, which found the “lack of evidence” underpinning it had “serious implications for [its] validity and reliability”.
“It is not reasonable to anticipate that the instruments are able to predict their specified risk with anything other than chance,” the report prepared by Dr Emily Corner for the home affairs department said.
But the government confirmed on Monday that, despite the report findings, the commonwealth continued to use the tool on 14 occasions across almost three more years.
“The VERA-2R risk assessment tool has been used as part of the preparation of reports by commonwealth-appointed experts in relation to commonwealth post-sentence order legal proceedings on a total of 14 occasions from June 2020 to 1 March 2023,” the department confirmed in a response to a question on notice from the Greens senator David Shoebridge.
The attorney-general’s department has also confirmed that in three cases, the contents of Corner’s report was never disclosed to the defendants.
“There have been three individuals who were the subject of concluded post sentence order proceedings,” a spokesperson said. “The Corner report was not disclosed to the court or those individuals’ legal representatives at the relevant time.”
In one of the cases, the government has now had to provide a formal explanation to both the court and the defendant for its non-disclosure.
The tool was used on multiple occasions in three cases to justify the use of continuing detention orders (CDOs) and extended supervision orders (ESOs), which allow for the imprisonment or control of individuals deemed to pose an unacceptable risk of committing a future terror offence.
The tool was also used in cases where the government did not end up applying for a CDO or ESO.
The department said it was now “working with the department of home affairs and other agencies to consider the circumstances of the non-disclosure of the Corner report”.
“The commonwealth has provided affidavits to the supreme court of Victoria and [one of the individual’s] legal representatives explaining the circumstances of the non-disclosure of the Corner report,” a spokesperson said.
Offenders subject to CDOs or ESOs have been detained in prison for years after completing their sentences, or are required to comply with strict conditions upon their release, such as regularly reporting to police or electronic monitoring.
The Guardian has found another eight cases in New South Wales where the risk assessment tool was used after Corner’s report was given to the federal government.
In the eight NSW cases, the state government relied on the VERA-2R tool to argue for either preventive detention or the imposition of strict post-sentence supervision, using the state’s Terrorism (High Risk Offenders) Act 2017.
In all cases, the VERA-2R risk assessment tool was used to help convince a court that the defendants posed an “unacceptable risk of committing a serious terrorism offence” and must either be detained or subjected to supervision orders to prevent such crimes.
The home affairs department is understood to have shared Corner’s report with a group of government officials, known as the VERA-2R community of practice, in December 2022. Some of those officials worked for the NSW government.
In many of the eight NSW cases, the courts did receive some warning of the limitations of trying to predict the risk of terror offences.
But none of the cases appear to have specifically referred to Corner’s report or its findings that the lack of evidence had “potentially serious implications” for the “validity and reliability” of the VERA-2R tool.
Shoebridge told the Guardian that the government had to come clean on whether the NSW government had been told about the Corner report, and when.
“We know that this assessment tool can have a significant impact on criminal law proceedings, and procedural justice would surely require the defendants to have this material,” he said.
“If you allow proceedings to go on and you don’t tell the court about this critical report, and allow these assessments to be put before a judge, there’s a real question about whether this is misleading the court.
“This goes to the very ethics of the administration of justice and the disregard that commonwealth agencies have for it.”
The independent national security legislation monitor, Grant Donaldson, said in a report released last week that Corner’s report “should have been provided in all [court] applications where relevant experts make a risk assessment using the VERA-2R tool. There is no excuse for not doing so.”
It is understood that Donaldson was referring only to the commonwealth’s use of the powers, not cases brought by the NSW government.
The Guardian revealed in July that the Australian Human Rights Commission and a peak body for Muslims had urged Donaldson to investigate the reliability of VERA-2R.
Corner declined to comment on the government’s handling of her report, other than to share a statement from the Australian National University confirming that she and another researcher were engaged by the Department of Home Affairs for the work. They commenced work in June 2018 and concluded in May 2020.
