In January, NYPD’s intelligence and counter-terrorism bureau told officers to “conduct thorough inspections of all eyewear permitted within inmate cells.” The memo said arrested people could record inside police buildings with glasses that look like ordinary prescription frames. Released footage of cell layouts, camera placements, or patrol routines, it said, would be a security problem. The bureau pointed to social-media videos as examples.
One clip posted to TikTok in July 2025 was shot on smart glasses inside a South Carolina detention center that had already been under federal review for conditions. The wearer filmed showers, cells, a crowded cafeteria, and other detainees. Other people in the frame, the Guardian reported, did not appear to know they were on camera. US jails normally confiscate phones. Some glasses have passed as regular Ray-Bans or Oakleys. A small blinking light on the frame is the main giveaway when the camera is on.
Fusion centers and ICE drew the same line
Maine’s fusion center circulated a June 2026 memo: smart glasses used by the public for “surveillance and other nefarious activities” add risk inside corrections and police sites. The same note said some glasses can run AI face recognition and tie an officer’s face to public records, which the memo called a doxing risk.
Similar warnings came from DHS’s joint counter-terrorism assessment team and fusion centers in San Diego, Tennessee, Virginia, Maine, and New York.
In August, ICE classified smart glasses as body-worn cameras and barred them from the “federal workspace.” Acting Director David Venturella wrote that Meta glasses or similar devices “could unintentionally capture, record or transmit sensitive information.” New York State courts, and courts in England and Wales, have banned recording eyewear in judicial buildings.
A January 2025 NYPD memo also told officers to watch for people in smart glasses acting in a way the department called suspicious, after the FBI said the perpetrator of the New Year’s Day 2025 New Orleans attack had worn smart glasses while moving through the French Quarter late in 2024. Meta’s line, through spokesperson Dina El-Kassaby, was that the Quarter is a tourist zone and a phone could have filmed the same streets.
The camera already pointed the other way
Public backlash against the same product has been running in parallel. Wearers can record people who never agreed to be in a video. Wired and others have documented harassment cases. Australia’s eSafety commissioner has separately argued that glasses should blur faces while filming.
Ryan Shapiro of Property of the People put the police memos in that frame: agencies film the public as routine, then treat a civilian camera as a threat. “When only the government is allowed to hold the camera, that’s not security. It’s impunity.”
Evan Greer of Fight for the Future said she was not surprised police worry about the same hardware being used against them. “There’s no such thing as Meta glasses that are only for the good guys.”
Meta says the glasses “are designed to be noticed.” Every pair, El-Kassaby said, has a bright white LED that blinks when a photo or video goes to the gallery. The light cannot be turned off. Covering or tampering with it disables the camera.
That design choice is the company’s answer to covert recording. The NYPD and Maine memos treat the LED as easy to miss in a booking room or a dim cell. Detention staff are now being told to look at frames, not only pockets.
Two privacy problems, one product
Law-enforcement operational security is real. A jail’s camera map and a courthouse’s holding pattern are not public records because someone walked in wearing Oakleys. Searching permitted eyewear after arrest is a facility-control decision, not a theory about the First Amendment on a sidewalk.
The other problem is the one civilians already have. A device that looks like sunglasses and stores video is a consent problem on a bus, in a clinic waiting room, and on a protest line. Face recognition on the same hardware turns a casual recording into identification. Maine’s fusion center named that risk for officers. It is the same risk for everyone else whose face is already in a scrapeable database.
An August 2025 NYPD paper in the same document set worried that public fear of police AI could feed unrest, and that “anti-tech violent extremist activity” might follow job loss and deepfakes. Civil-liberties lawyers have flagged that phrase as a way to smear ordinary opposition to surveillance tools. The glasses memos sit in that pile: agencies want AI for investigations and want limits when the public holds the same class of sensor.
ICE’s workspace ban is the cleanest operational fix for agencies. It does not answer street recording, store listings that still sell the glasses as fashion, or whether a blinking LED is notice enough under state wiretap and consent-to-record statutes. Those statutes were written for phones people hold up, not frames people already wear for vision.