Scribd
Upload
20 views1 page

NZ Herald Reported Two Tertiary Institutions

Facial recognition technology has quietly integrated into many areas of everyday life beyond just security and surveillance, including retail to track customer behavior. Its use is also growing in New Zealand, such as by a major supermarket chain and tertiary institutions. However, there are key privacy issues with facial recognition regarding the collection and use of facial image databases without proper consent, as well as the technology's unreliability and potential for misidentification due to biases, which could result in inaccurate personal information in violation of privacy principles.

Uploaded by

Aman Gupta
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
Download as docx, pdf, or txt
20 views1 page

NZ Herald Reported Two Tertiary Institutions

Facial recognition technology has quietly integrated into many areas of everyday life beyond just security and surveillance, including retail to track customer behavior. Its use is also growing in New Zealand, such as by a major supermarket chain and tertiary institutions. However, there are key privacy issues with facial recognition regarding the collection and use of facial image databases without proper consent, as well as the technology's unreliability and potential for misidentification due to biases, which could result in inaccurate personal information in violation of privacy principles.

Uploaded by

Aman Gupta
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1/ 1

Facial recognition technology has quietly integrated into a number of areas of everyday

life. In addition to the more obvious security, public safety and surveillance applications
(which have led, for example, to Hong Kong protesters wearing masks and felling
lampposts equipped with facial recognition CCTV cameras in order to avoid
identification), the technology is increasingly being used as a retail tool to track customer
behaviours and movement to generate predictive consumer analytics. And New Zealand
is following suit. Last year the NZ Herald reported on a major supermarket chain having
deployed facial recognition CCTVs in some of its stores and more recently on two
tertiary institutions trialling the technology to monitor student attendances in New
Zealand. But how does this increasingly common technology fit within global trends
towards greater privacy, control and transparency over data collection and use?
There are three key privacy issues that tend to arise in facial recognition privacy
discussions:
First, facial recognition software can only work alongside a rich database of facial
images, so the recognition algorithm can be trained to detect faces, and to then match a
detected face to an identity in the database. Populating and using facial image
databases raises all sorts of questions regarding the source of those images, the extent
of any consent or authorisation obtained, the potential uses to which the images will be
put and how they will be protected through storage and security. There have been
reports that some facial recognition software products already in use have been trained
on images of individuals mined from the internet and obtained without consent. In the
absence of appropriate consents, these products may simply breach the law.
Second, independent tests of facial recognition technology have repeatedly shown that it
is far from perfect or reliable, presenting a meaningful challenge to compliance with
Information Privacy Principle 8 under New Zealand’s (current) Privacy Act and
equivalent legislation around the world. Principle 8 requires that “An agency that holds
personal information shall not use that information without taking such steps (if any) as
are, in the circumstances, reasonable to ensure that, having regard to the purpose for
which the information is proposed to be used, the information is accurate, up to date,
complete, relevant, and not misleading.” If an agency was collecting images and using
unreliable technology to match that image to an individual it is difficult to see how it could
comply with this principle. Facial recognition technology relies heavily on the quantity
and quality of the data fed into it. The potential for misidentification and error rates due to
gender and racial biases in algorithms are known to be high, even with the best software
of its kind, with disproportionately higher error rates among certain ethnic and racial
groups. Depending on the function, flaws in facial recognition software could have
catastrophic impacts.

You might also like