To its supporters, the system is a more effective and less obtrusive alternative to the age-old metal detector, making events both safer and more pleasant to attend. To its critics, however, Evolv’s effectiveness has hardly been proved. And it opens up a Pandora’s box of ethical issues in which convenience is paid for with RoboCop surveillance.
“The idea of a kinder, gentler metal detector is a nice solution in theory to these terrible shootings,” said Jay Stanley, senior policy analyst for the American Civil Liberties Union’s project on speech, privacy, and technology. “But do we really want to create more ways for security to invade our privacy? Do we want to turn every shopping mall or Little League game into an airport?”
Evolv machines use “active sensing” — a light-emission technique that also underpins radar and lidar — to create images. Then it applies AI to examine them. Data scientists at the Waltham, Mass., company have created “signatures” (basically, visual blueprints) and trained the AI to compare them to the scanner images.
Executives say the result is a smart system that can “spot” a weapon without anyone needing to stop and empty their pockets in a beeping machine. When the system identifies a suspicious item from a group of people flowing through, it draws an orange box around it on a live video feed of the person entering. It’s only then that a security guard, watching on a nearby tablet, will approach for more screening.
— Steven Zeitchik in AI may be searching you for guns the next time you go out in public