Member Article

EU ponders ban on facial recognition technology

According to an early draft of a whitepaper, the EU could provisionally ban the use of facial recognition technology in public places such as sports stadiums, shopping centres and train stations, due to mounting public concern over the surveillance of European citizens. By imposing a three to five year ban, Brussels would attempt to manage this risk and slow the speed at which this software is being adopted.

The technology currently allows faces captured on CCTV to be monitored and cross-checked in real time against watchlists, which are often compiled by police. Late last year, politicians and campaigners urged the UK police and other organisations to stop using live facial recognition for public surveillance, claiming that the technology is intrusive, inaccurate and infringes on privacy. However, advocates claim it actually helps to protect the public as it can catch people in a way police can’t. Taking both sides into consideration, the EU has stated that the proposed ban would allow new rules to be developed to bolster existing regulation surrounding privacy and data. These guidelines would impact both developers and users of artificial intelligence, and some exceptions to the ban could be made for specific security projects as well as R&D efforts.

“If the European Commission’s decision to ban facial recognition in public areas goes ahead, the technology clearly loses any security or safety benefit that it intends to deliver,” commented Steven Furnell, senior member of the IEEE, associate dean and professor of information security at University Plymouth. “If we are to continue using facial recognition, then it comes down to responsible use and how this is governed. There is guidance on police use of AFR that was published last year by the Surveillance Camera Commissioner, and a related data protection code of practice for surveillance cameras and personal information from the ICO. So, being able to demonstrate compliance with these principles may ease concerns in some cases. Nonetheless, having guidance is not the same as having specific regulation around the use of the technology and the handling of the data concerned.”

This proposed ban is not unique to Europe. Last year, areas across the US also started to prohibit the use of facial recognition technology. In May, it was banned in San Francisco, with Oakland following suit shortly afterwards. Somerville and Brookline in Massachusetts also imposed a ban, with a potential state-wide ban to follow. Lastly, San Diego suspended a facial recognition program in advance of a new state-wide law (which declared it illegal) coming into effect.

“Generally speaking, there is a lack of general public understanding about the technology and its capability. Indeed, many will have first become aware of facial recognition as a result of coverage of related concerns and bans in the past, and so the resulting impression is very likely to be negative. If it is to be accepted, there needs to be effective awareness-raising around the intended benefits, the accuracy of the technology, and how it will be governed to prevent misuse. Otherwise many will continue to see it as another potentially uncontrolled genie escaping from the technology bottle.”

The outcome of this potential ban remains to be see, once the final version of the EU whitepaper is published in February as part of a wider overhaul of the regulation of artificial intelligence.

This was posted in Bdaily's Members' News section by D Baker .