The Backlash Against Flock Safety ALPR Cameras
Update as of late July 2026
Scale of the Pushback
|
![]() |
Why Cities Are Pulling Out
Unauthorized federal and ICE access
This is the single biggest trigger. A San Francisco compliance audit found that the Northern California Regional Intelligence Center ran 299 improper queries on the city's Flock network on behalf of federal and out-of-state agencies over roughly one year. More than 50 federal, state, and local agencies have run searches tied to protest activity, and reports that local data was reaching ICE and CBP despite state privacy laws prompted several of the cancellations.
Officer misuse
Multiple cases have surfaced of police officers using the system for personal reasons unrelated to law enforcement — including stalking ex-partners and monitoring activists and protesters.
Lawsuits
- A class action was filed in San Francisco Superior Court (February 2026) alleging that ICE, CBP, the FBI, and the ATF queried the city's cameras without authorization.
- The city of Eugene, Oregon has been served a second public-records lawsuit for withholding Flock-related records.
A Supreme Court wrinkle
A recent SCOTUS ruling is seen by legal analysts as potentially weakening Flock's legal footing, though it is not expected to stop ALPR-based mass surveillance broadly.
Grassroots and Direct Action
- Citizens have sawed down camera towers, sabotaged solar charging panels, and paint-bombed camera sensors — the ACLU's campaign against this technology is literally named “Get the Flock Out.”
- Vandalism has become common enough that Flock has reportedly started disguising cameras inside other objects; one was found hidden inside a speed-radar box operated by police in Volusia County, Florida.
- A tracking project called DeFlock maintains a public map and running count of camera locations and contract cancellations.
AI Industry Response: Anthropic vs. the Rest
Anthropic is the one major AI lab that has staked out a public, principled stance specifically against domestic surveillance — and it has cost the company dearly.
Anthropic's stance*
Anthropic's usage policy bars use of its models for mass domestic surveillance. When the Pentagon pushed to loosen Claude's restrictions for broader "lawful use" by the Department of Defense, CEO Dario Amodei refused, saying the DoD made "virtually no progress" on ruling out use for mass surveillance of Americans or fully autonomous weapons, and that Anthropic "cannot in good conscience" allow unrestricted DoD use. Anthropic's public reasoning: "mass domestic surveillance of Americans constitutes a violation of fundamental rights," and that such surveillance is only legal today because the law hasn't caught up with AI's capabilities — the same concern at the heart of the Flock backlash.
The government's response
- February 27, 2026 — President Trump directed federal agencies to stop using Anthropic technology.
- Defense Secretary Pete Hegseth designated Anthropic a "Supply-Chain Risk to National Security."
- June 12, 2026 — the government forced Anthropic to disable its two most powerful models, Fable 5 and Mythos 5, for every customer worldwide, without explanation. (Fable 5 is back!)
How other providers compare
Amazon's Ring (owned by Amazon, which also builds its own AI models) announced a partnership to feed Ring doorbell footage into Flock's network — already used by ICE, the Secret Service, and police. That deal was only canceled after a Super Bowl ad triggered public outcry, not because of any pre-existing policy against domestic surveillance use. There is no public evidence that Flock's "natural language search" features (Flock Nova) are built on a major outside LLM provider with restrictive usage terms; Flock appears to run its own models and analytics pipeline rather than being gated by an outside provider's policy the way DoD use of Claude is gated by Anthropic's stance. The broader pattern: other major AI providers have pulled back surveillance-adjacent products only after public backlash, not on a stated principle — Anthropic is the outlier in drawing the line in advance on principle.
Sources
- The backlash against Flock cameras is spreading — Malwarebytes
- Map Shows Flock Camera Locations in Every State Amid Growing Backlash — Newsweek
- Fight Creepy ALPR Cameras — ACLU
- So Many People Are Smashing Flock's Cameras That It's Disguising Them Inside Other Objects — Futurism
- Get the Flock Outta Here: Backlash Grows Over Video Surveillance Tool — U.S. News
- Inside the growing police use and misuse of Flock's license-plate reader cameras — CNN
- People Across The US Are Tearing Down Flock's Traffic Cameras — Yahoo
- San Diegans Push Back on Flock ALPR Surveillance
- How Cops Are Using Flock Safety's ALPR Network to Surveil Protesters and Activists
- Flock Safety Crosses 100,000 Cameras as 53 Cities Cancel Over Unauthorized Federal Data Access — Tech Times
- Cities Are Giving Up on Flock Safety Cameras — Townhall
- Why Flock Safety Finds Itself in a Surveillance Backlash — GovTech
- SCOTUS Ruling Could Be Bad News for Flock — Truthout
- Denver City Council rejects extension of Flock Safety cameras — CBS News
- South Pasadena to remove Flock Safety cameras over privacy concerns — AOL
- Another public records lawsuit filed against Eugene over Flock records — AOL
- What Everyone Is Missing About Anthropic and the Pentagon — Internet Governance Project
- U.S. Government Bans Use of Anthropic Products — Taft Law
- Anthropic Refuses Pentagon Demand to Remove AI Security and Safety Guardrails — ASIS
- Is the Pentagon allowed to surveil Americans with AI? — MIT Technology Review
- Anthropic–United States Department of Defense dispute — Wikipedia
- Scaling Laws: The Pentagon Goes to War With Anthropic — Lawfare
- Amazon's Ring cancels partnership with Flock — TechCrunch
- Ring and Flock cancel camera partnership amid outcry over Super Bowl camera ad — AOL
- Flock Safety launches new AI-powered tools to accelerate police investigations — Police1
-------------
* Disclaimer: To save myself a ton of time I actually used Anthropic's Claude to assemble this information. AI is not good or bad on it's own. It is just a tool what can be used for good or evil depending an man, not machine.
/
