San Francisco forbids itself face recognition
On 21 May 2019 the San Francisco Board of Supervisors gave final passage to Ordinance No. 103-19 by 10 votes to 1. It forbade city departments, the police among them, to obtain, retain or use face recognition technology or information derived from it, and for every other surveillance technology it required prior approval by the Board with an impact report and an annual audit by the Controller.
Why it matters
The city gave up a technology before it had started using one, and did so not by statement but by a norm with a text and a sanction. Wire coverage reports it as the first such ban by a city in the United States. In the atlas the line of harm had so far begun with measurements; this is the first record in which a measurement turned into law.
The ordinance added Chapter 19B to the city's Administrative Code. The prohibition sits in subsection 19B.2(d): no Department may obtain, retain, access or use face recognition technology on City-issued software or a City-issued product or device, nor any information obtained from it. The rest of the chapter requires a department to hold a Board-approved Surveillance Technology Policy and to file a Surveillance Impact Report before acquiring or newly using surveillance technology, and requires the City Controller to audit compliance each year. There were two votes: a first reading on 14 May 2019, 8 to 1 with two supervisors excused, and final passage on 21 May, 10 to 1. Catherine Stefani voted no both times. The sponsor was Supervisor Aaron Peskin. The record gives no count of the departments the ban reached: the ordinance states no such number. The words of subsection (d) are quoted from the code as it now stands, amended in 2021 by Ordinance 116-21; the city's copy of the 2019 text is a scan with no text layer.