AfriForum wants assurance on Sanral facial recognition tech

AfriForum has written to the South African National Roads Agency (Sanral) seeking assurance that its highway cameras will not be used for facial recognition and the invasion of privacy.

On Friday it questioned the agency about its use of machine learning (ML) technology meant to reduce congestion and improve road safety in the country.

This follows Sanral’s statement this week that its Technical Innovation Hub (TIH) is investigating the extent to which ML technology can be utilised in SA.

ALSO READ: Sanral lists legal steps it can take against e-toll delinquents

AfriForum wants Sanral to provide detailed information about the implementation of ML technology. The organisation has raised questions over the potential violation of citizens’ constitutional right to privacy through the use of facial recognition technology.

AfriForum also wants to know what steps Sanral will take to prevent such abuse.

“AfriForum does not oppose the harnessing of the latest technology to facilitate road safety, law enforcement or emergency response. However, this progress cannot be pursued at the cost of people’s constitutional rights,” AfriForum legal and risk manager Marnus Kamfer said in a statement.

“AfriForum is concerned that Sanral cites China as an example of where ML technology has been successfully implemented – especially since the Chinese government has a poor record of human rights violations and infringing on the privacy of its citizens.

“Technological progress cannot be pursued by the government to increase surveillance abilities that violate the privacy of its citizens or, to a degree, would enable rampant abuse.”

Potential privacy risks

Sanral on Tuesday admitted the use of the technology also came with significant risks.

But the agency said all efforts were being made to understand how to effectively use ML technology while maintaining strict compliance with legislation relating to the privacy of road users.

“Some of the ways to mitigate these potential privacy risks are strict security and access controls. Data can also be anonymous at the point of capture.

“After all, the intention is not to observe individuals, but to identify trends and incidents to inform appropriate response and interventions.”

Mechatronics engineer in the TIH Ruan van Breda explained how the technology works.

“Machine learning can be used to detect and segment objects within a camera frame (each frame of a video is analysed as a still image). These objects can then be classified based on pre-trained image classifiers. In road environment this allows one to detect and classify different type of vehicles, pedestrians, different types of animals, cyclists and so on.”

READ NEXT: Sanral mum on ‘dodgy’ R40m payment in e-tolls contract