Top Australian police continued to meet with the CEO of controversial facial-recognition software company Clearview AI after both the company’s and police’s use of its technology was found to have broken Australian privacy law.
Crikey revealed in June that Australian Federal Police (AFP) had secretly met with Clearview AI staff in 2022, but new documents obtained by this masthead provide further insight into how police have continued to work closely with a company that violated the privacy of huge numbers of Australians while defying calls for more transparency about their relationship.
Clearview AI is a US-based company that gained global notoriety for scraping billions of images of people off the internet without permission to create a tool that allows users to scan a photo and find matching photos and information about them — such as their name and location — from across the web.
The company, led by Australian co-founder and CEO Hoan Ton-That, offered this to hundreds of law enforcement agencies around the world, including the Australian Federal Police whose staff trialled it in 2020.
The unauthorised gathering of people’s data attracted the attention of regulators and policymakers around the world. Clearview AI has been the subject of countless lawsuits and adverse legal findings.
In 2021, Australia’s information and privacy commissioner Angelene Falk found that Clearview AI had broken Australia’s privacy law — citing its collection of information without consent and by unfair means, and its failure to take steps to notify users of this collection or ensure it was accurate — and ordered the company to stop collecting information on Australian users and delete previously obtained data. Falk also slapped down the AFP for breaching Australians’ privacy by using the technology.
Despite these findings, emails and minutes released through a freedom of information request show that AFP staff repeatedly met with Ton-That and invited him to present at a high-level meeting of Australian and New Zealand police leaders. AFP staff also agreed to work with an Australian academic with the hope of encouraging the use of his technology in Australia.
On May 5 2022, Ton-That emailed Jon Rouse, operations manager at the time of the AFP’s Australian Centre to Counter Child Exploitation (ACCCE), to set up a chat to encourage him to meet with an Australian academic working on facial recognition technology regulation.
“I think it can be a positive thing for ensuring the use of FRT [facial recognition technology] by law enforcement in Australia,” Ton-That wrote.
“Was great to meet you in France,” Rouse replied. “Definitely open to a call”.
When Crikey first reported on an email regarding this study in June, both Clearview AI and the AFP refused to answer questions about the meeting, only stressing that the two did not have a commercial relationship.
In an email later that day, Rouse proposed arranging for Ton-That to brief Operation Griffin, a group of Australia’s and New Zealand’s head of law enforcement child protection units, so that “they get some education on the Clearview issue”.
“Would you be open to me setting up a video conference where you could basically do what you did in Lyon?” Rouse asked.
“Thanks for asking, I would love to,” Ton-That replied.
Meeting minutes from the July 7 meeting of Operation Griffin show that Ton-That had a 30-minute presentation on facial recognition technology.
“The chair welcome [redacted] CEO Clearview AI, to the meeting to present on facial recognition technology and its use in contributing to law enforcement outcomes,” the minutes read.
An AFP media person acknowledged Crikey’s questions about its relationship with Clearview AI and whether it knew if the company continued to break Australia’s privacy law, but did not respond by deadline. Ton-That and Rouse have been contacted for comment.
Rouse, who was just last week appointed Queensland’s interim victims’ commissioner, has been in the media this week advocating for the use of Clearview AI.
“Imagine if we never used DNA technology to solve cases. This technology is a similar kind of revolution in law enforcement,” he told the Herald Sun.
Greens Senator for NSW and digital rights spokesperson David Shoebridge said the continued meetings were deeply troubling.
“The fact that AFP leadership continued to meet with the company’s CEO even after the information and privacy commissioner’s directive to halt the use of Clearview AI’s controversial technology shows how little accountability and oversight there is in this space,” he said.
Shoebridge said the news should ring alarm bells to the regulator.
“Instead of doing their jobs to tackle issues like foreign interference, which are having significant impacts on Australia’s diaspora communities, the AFP is collaborating with controversial companies in an attempt to normalise the use of facial recognition technology and broaden its surveillance powers. We need stronger regulations that protect Australians as a matter of urgency,” he said.
