Municipal Transportation Agency Biannual Surveillance Report 2026

General Description

1. Please (a) describe the technology or technologies and (b) make a clear notation of which policy each technology corresponds to.

Red Light Cameras and No Right-Turn Surveillance Technology Policy: This technology enforces traffic law by photographing the license plates and drivers of vehicles that run red lights or make illegal turns, and it issues citations by mail.

Automated Speed Enforcement Camera Surveillance Technology Policy: This technology enforces speed limit laws by photographing the license plates of vehicles that speed, and it issues citations by mail.

License Plate Readers and Transit-Only Lane Enforcement (TOLE) Camera Surveillance Technology Policy: License plate readers use vehicle mounted cameras to identify vehicles for parking management and enforcement and to identify vehicles reported on a hotlist database. Transit-Only Lane Enforcement uses bus-mounted cameras to identify violations of the Transit-Only Lane and parking in bus stops for issuance of citations to the registered owner.

Security Cameras Surveillance Technology Policy: SFMTA security cameras are fixed and mobile video surveillance systems deployed across transit vehicles, facilities, and public areas to capture real-time and recorded video. These systems are designed to support departmental operations by enabling live monitoring, recording images and video footage, and reviewing recordings as needed. The technology includes high-resolution cameras that operate continuously (24/7) and are capable of storing and retrieving footage for authorized purposes.

Taxi Dashboard Camera Surveillance Technology Policy: This technology uses passenger compartment dashboard cameras owned by taxi permit holders to investigate customer complaints and ensure compliance.

Video Analytics for SFMTA passenger vehicles Surveillance Technology Policy: This technology uses video and audio event recorders together with proprietary, vendor-owned algorithms to record and identify certain behavior-based safety events, such as operator looking at cell phone while driving. The event recorders are triggered by collision impacts, abrupt braking, and similar events, and capture eight seconds of video/audio prior to the trigger, and four seconds after the trigger, for a total of 12 seconds of video and audio. These events are then categorized by the vendor, reviewed and corrected by the SFMTA, and if appropriate, action is taken to address safety and/or training issues.

2. How was the surveillance technology or technologies used by your department during the reporting period? Provide a 3-4 sentence description for each technology listed.

Red Light Cameras and No Right-Turn: During the reporting period the technology was used at 19 enforced approaches located at 13 intersections. The technology photographed license plates and drivers that ran red lights or made illegal turns. It also issued citations by mail.

Automated Speed Enforcement Camera: During the reporting period the technology used electronic automated detection equipment to detect a violation of speed laws and cameras to obtain a clear photograph of a speeding vehicle's rear license plate. These cameras are only triggered by speeding vehicles. They do not record data unless triggered by a speeding vehicle.

License Plate Readers and Transit-Only Lane Enforcement (TOLE) Camera: During the reporting period, the technologies were used for parking management and enforcement. They were also used to identify vehicles on hotlists for referral to SFPD and to enforce Transit-Only Lanes. In addition, de-identified occupancy data from the cameras was used to inform the Department's demand-responsive parking program and to develop changes to curb regulations.

Security Cameras: During the reporting period, SFMTA security cameras were used to support public safety, operational awareness, and enforcement activities across vehicles, transit facilities, and public right-of-way locations. The system enabled live monitoring as well as the recording and review of video footage to investigate incidents, respond to complaints, and document safety or security-related events. Authorized personnel reviewed footage and, when appropriate, provided video evidence to law enforcement or other authorized entities following incidents. Cameras were also used to support enforcement of parking and driving violations and to improve customer and employee safety conditions.

Taxi Dashboard Camera: During the reporting period, the passenger-compartment dashboard cameras were used to investigate customer complaints and to ensure taxi permittees' compliance with the requirements and conditions of Article 1100 of the Transportation Code. The video and associated data were reviewed in response to complaints received from the public about on-board incidents. The technology was also used to verify that taxi drivers completed publicly funded accessible rides — such as those under the paratransit or wheelchair taxi incentive — before the companies or taxi drivers were paid for those rides.

Video Analytics for SFMTA passenger vehicles: During the reporting period, the video and audio event recorders were used to address safety issues and to support driver training and remediation. The technology was also used for driver recognition for collision avoidance and for notification of mechanical issues affecting a bus. In addition, it supported possible driver discipline and the identification of infrastructure issues affecting safety, such as roadway sinkholes, overhanging trees, and blocked signage.

Surveillance Technology Goal

3. Has the surveillance technology been effective at achieving its identified purpose?

Yes

4. In 4-10 sentences, please explain how the technology has or has not been effective.

Red Light Cameras and No Right-Turn: The technology has been effective at identifying and citing red light and illegal turn violations. The Automated Enforcement Program is part of the Department's Vision Zero commitment to eliminate traffic fatalities, and enforcement cameras are placed at the intersections with the highest collision totals so that enforcement is directed where the safety risk is greatest. By reducing red light running and illegal turns, the technology helps make intersections safer for pedestrians, bicyclists, and other vehicles and decreases the risk of collisions that cause serious injuries or fatalities. The technology also enforces these violations without bias and removes the potential for escalation associated with in-person traffic enforcement. In addition, associated data collected by the system — including vehicle counts, vehicle speeds, and violation counts — is used for engineering analysis to assess traffic safety and the effectiveness of the cameras at reducing red light running and illegal turns.

Automated Speed Enforcement Camera: The surveillance technology has efficiently enforced vehicle speed laws, helping staff remotely identify speeding violations at multiple locations and improving the effectiveness and efficiency of speed enforcement. Speed enforcement supports the Department's Vision Zero goal of eliminating traffic-related fatalities, as enforcement is one of the “three E's” of Vision Zero — education, engineering, and enforcement — and speed is the leading contributor to traffic collisions that cause serious injuries and fatalities. The technology is intended to reduce vehicle speeding and, by extension, the likelihood of speed-related collisions. Because enforcement is automated, it allows speed limits to be enforced without in-person traffic stops, improving staff safety and removing bias and police contact from routine enforcement. The technology also improves the accuracy of data on vehicle speeds and provides more immediate data to demonstrate the impact of various traffic control measures on streets over time.

License Plate Readers and Transit-Only Lane Enforcement (TOLE) Camera: The technology has effectively identified vehicles in violation of parking regulations, vehicles impacting Transit-Only Lanes, and vehicles on hotlists referred to SFPD for appropriate action. It has also been useful for planning purposes. In particular, it has provided de-identified data on the utilization of curb space to inform potential changes to parking regulation. It has also supported operational management of the demand-responsive parking program.

Security Cameras: The SFMTA security camera system has been effective in enhancing public safety and supporting day-to-day operations. The ability to perform live monitoring and review recorded footage has improved incident response and allowed staff to investigate safety or security-related events more efficiently. The system has also been effective in supporting enforcement activities, including identifying violations and sharing relevant footage with law enforcement when appropriate. Additionally, the cameras' continuous 24/7 operation has increased situational awareness while reducing the need for constant physical monitoring. Overall, the technology has contributed to improved safety for both employees and customers by deterring misconduct and providing reliable documentation of incidents, demonstrating its effectiveness in meeting its intended purpose.

Taxi Dashboard Camera: The technology has been effective for investigating complaints, verifying appropriate use of accessibility subsidies, and ensuring compliance with permit conditions. It allows the Department to review recordings of on-board incidents in response to public complaints and at appeals hearings, and to confirm taxi permittees' compliance with the requirements and conditions of Article 1100 of the Transportation Code. The technology is also used to verify that rides paid for with public funds were completed before drivers are paid — for example, confirming under the paratransit or wheelchair taxi incentive that drivers picked up individuals with qualifying disabilities for trips funded through paratransit programs — and to investigate transportation code violations and incidents involving taxis. Reviewing a recording of an incident improves data quality by helping the Department determine whether corrective action is warranted, rather than relying solely on witness interviews, which can be time-consuming and unreliable. The presence of the cameras also serves as a deterrent to drivers who might otherwise operate a vehicle in an unsafe manner.

Video Analytics for SFMTA passenger vehicles: The technology has been effective at improving driver safety, reducing collisions, and identifying problems leading to passenger falls. It allows the Department to identify collision dynamics and causation and investigate passenger fall events and explore potential safety improvements. The technology also identifies driver training issues, misconduct, or negligence and allows the Department to commend drivers who demonstrate outstanding defensive driving skills. Because the event recorders are triggered by excess g-forces and capture a short 12-second clip that the vendor categorizes and reviews before sending it to the Department, the technology lets the Safety Division target training to the specific driver behaviors that trigger safety events and work to minimize those behaviors over time. It also enables the Department to identify infrastructure and signage issues — such as damaged bus stop shelters, hazardous trees, or obscured signs — that affect transit safety. Overall, the technology provides these safety, training, and engineering insights without the hundreds of additional personnel that would otherwise be required to obtain them.

4.a. Provide quantitative data to support your response. This should include crime statistics for the radius where the technology operates if that was a motivating factor in acquiring the surveillance technology.

Red Light Cameras and No Right-Turn: In the full year of 2025, between January and December, the Program issued 10,385 citations.

Automated Speed Enforcement Camera: There were 553,617 warnings issued since March 2025 program launch, including the initial 60-day warning period and 163,906 citations issued since August 2025 when citation issuance began.

License Plate Readers and Transit-Only Lane Enforcement (TOLE) Camera: The technology supported up to 215,037 meter violations, 244 hotlist reads, and 7,716 transit-only lane enforcement violations.

Security Cameras: Since 2024 the technology has provided 26,994 video pulls. The Departmentbelieves that approximately 40% of these pulls were to investigate crime, including 419 assaults on operators. We are unable to give a precise breakdown due to technology limitations.

Taxi Dashboard Camera: The technology has been used to verify an average use of 85 wheelchair users transported subsidies per month.

Video Analytics for SFMTA passenger vehicles: Data is only available for any preceding 180-Day inquiry, but in the previous period from this date (May 1, 2026) there were 21,363 video events forwarded to the MTA by the vendor (Lytx/Drivecam). Among that number close review and action was taken on 1835 video events, including 26 collision avoidance recognitions, 26 possible red-light violations, and 17 speed policy violations.

Complaints, Concerns and/or Accidental Receipt of Face Recognition Technology

5. How many complaints or concerns has your department received from the public about the surveillance technology?

Red Light Cameras and No Right-Turn: The Department has not received any complaints or concerns.

Automated Speed Enforcement Camera: The department has received five complaints or inquiries, four via the program email address and one via 311.

License Plate Readers and Transit-Only Lane Enforcement (TOLE) Camera: The Department has not received any complaints or concerns.

Security Cameras: The Department has not received any complaints or concerns.

Taxi Dashboard Camera: The Department has not received any complaints or concerns.

Video Analytics for SFMTA passenger vehicles: The Department has not received any complaints or concerns.

6. Please summarize the complaints or concerns which your department received about the surveillance technology.

Red Light Cameras and No Right-Turn: The Department did not receive any complaints or concerns about this technology during the reporting period.

Automated Speed Enforcement Camera: Four complaints received by the Department represent that Speed Cameras are an invasion of privacy and overreach of government which cause distress to San Francisco residents. One inquiry sought to confirm whether the program shares data with law enforcement. Technical safeguards make loss of dignity (embarrassment and emotional distress) unlikely because the cameras photographs only the rear license plate (not drivers or occupants); any incidental images of nearby pedestrians or cyclists are purged by the vendor; DMV records associated with license plate numbers are used only to identify the vehicle for a speed violation; access is limited to authorized users; the data is confidential under state law (CVC §22425); retention is short; and data is anonymized where shared

License Plate Readers and Transit-Only Lane Enforcement (TOLE) Camera: The Department did not receive any complaints or concerns about this technology during the reporting period.

Security Cameras: The Department did not receive any complaints or concerns about this technology during the reporting period.

Taxi Dashboard Camera: The Department did not receive any complaints or concerns about this technology during the reporting period.

Video Analytics for SFMTA passenger vehicles: The Department did not receive any complaints or concerns about this technology during the reporting period.

7. How many complaints or concerns has your department received from the public about the receipt of information from face recognition technology?

All six technologies: The Department received no complaints or concerns, because no face recognition technology was used.

8. Please summarize the complaints or concerns which your department received about the receipt of information from face recognition technology.

All six technologies: This is not applicable, because no face recognition technology was used.

Violations

9. Were there any violations of the Surveillance Technology Policy or Surveillance Impact Report, reported through community members, non-privileged internal audits, or through other means in the last year?

No

Requested Modifications

10. Is your department requesting to modify the Surveillance Technology Policy or Policies covered in this report?

No