Face biometrics force a deeper recognition Down Under

A white paper from Australia’s Monash University about the increasing use of biometric facial recognition cuts directly to an anti-privacy startup that is proving problematic for the technology, and to a lesser extent, the nation itself.

But the publication put a light on a societal concern darker than who has the right to use a person’s face.

One hundred words into the report, the authors bring up Clearview AI. That is the startup, co-founded by native Australian Hoan Ton-That, that scrapes facial pictures from every crevice of the internet that is not behind security or pay walls. It also includes images that are uploaded so that they can be matched against the entire database.

Hoan Ton-That, his backers and fans and subscribing police agencies worldwide feel if someone’s face can be found online, it might as well have been found on a milk carton. It is available to be used.

In Clearview AI’s case, the image can be used in a facial recognition database containing 3 billion to 4 billion images which, for example, law enforcement officials can check as part of routine police tasks. Controversy and court challenges have ensued.

According to a survey commissioned as part of the white paper, 49 percent of respondents said using face scanning in public spaces is an invasion of privacy. A third of survey takers said the risks of scanning countered any benefits.

The nationwide questionnaire was paid for by Monash University’s Automated Society Working Group.

Perhaps shifting their perspective to that of the victim of a crime, 61 percent said Clearview AI might be on to something. A role in public safety, these respondents said, might be the benefit that outweighs the risks.

Indeed, those responding to the facial recognition survey “were broadly supportive … even in cases where it was seen as inaccurate or biased around matters of race.”

In this case, it seems that respondents are saying they can live with some ambiguity when it comes to foreigners in the nation. They might perceive facial recognition as part of a heightened crime deterrence.

“Even though respondents noted concern, they accepted facial recognition in contexts where racialised bias in more significant,” according to the report.

Article Topics

accuracy  |  Australia  |  biometrics  |  Clearview AI  |  criminal ID  |  facial recognition  |  privacy