The Facts

This page collects the core factual claims behind DeFlock Wylie's campaign. Claims are dated because the situation changes: contracts, court cases, and camera counts move over time. Where a fact is time-sensitive, verify it against the cited primary source. Corrections: [email protected].

About Flock Safety

• Flock Safety is the brand of Flock Group Inc., a surveillance company founded in 2017 that sells ALPR cameras and audio-detection devices to police, businesses, and HOAs, connected through a national searchable network.
• Flock ALPR cameras record every passing vehicle's plate, make, model, color, and distinguishing features with time and location, not only vehicles tied to a crime.
• The company shares a major early investor (Founders Fund) with Palantir, a data-analytics and intelligence contractor.

The Wylie contract

• The City of Wylie contracts with Flock Safety to operate ALPR cameras in the city at $27,000 per year. The contract was procured through a sole-source process (no competitive bid), approved on a consent agenda, and was not put to a public vote.
• The current term runs October 14, 2025 through October 13, 2026 and auto-renews for 24 months unless either party gives written notice of non-renewal at least 30 days before the term ends; the no-cost non-renewal deadline is September 13, 2026. As of August 19, 2026, the renewal is Item 1 on the August 25, 2026 City Council regular agenda.
• The contract, sole-source letter, ALPR policy, council minutes, and network audit logs were obtained by residents under the Texas Public Information Act.
• Wylie's program has been positioned for expansion (additional cameras), which raises the stakes of the surveillance footprint rather than reducing it.

Documented security failures

• In 2025, technologist Benn Jordan and other researchers demonstrated that roughly 60+ Flock AI cameras were exposed on the public internet with no password, streaming live and archived footage that anyone with a browser could view, in some cases reaching admin-level functions.
• Security researchers have documented numerous vulnerabilities in Flock hardware and software, including ways to access unencrypted images and credentials.
• Law-enforcement and government system credentials (including some tied to Flock) have been reported for sale on dark-web markets, where "access brokers" trade stolen logins.

Audio surveillance

• Some Flock devices include microphones marketed to detect gunshots and other "sounds of distress," raising the prospect of recording ordinary public conversations.

Effectiveness and accuracy

• When the ACLU reviewed scan data from nearly 300 police departments, its example agencies showed hit rates as low as 0.01% (Rhinebeck, NY) and 0.08% (High Point, NC). In Maryland's statewide data, only 47 out of every million plates read (0.005%) were potentially associated with a stolen car or a person wanted for a serious crime. (Source: ACLU, "You Are Being Tracked")
• The most complete real-world test to date: Atlantic City installed fixed ALPRs on every road on and off the island. The peer-reviewed evaluation (co-authored with an Atlantic City police official) found the expansion did not reduce violent crime (it found some association with reduced vehicle theft and shootings). The authors note rigorous ALPR research "has not kept pace" with the technology's spread. (Source: Justice Evaluation Journal, 2025)
• A police-run study in Vallejo, California found up to 37% of fixed-position ALPR hits were erroneous: stale or wrong entries the cameras faithfully matched. (Source: The Police Chief study, via ObscureIQ accuracy review)
• In July 2026, the LAPD Inspector General audited 498 Flock hot-list alerts and found 161 (roughly 1 in 3) were false. LAPD let its Flock contract expire the day after the audit published. (Source: LAPD OIG audit coverage; TechCrunch)
• By Flock's own published numbers (96% character-level read accuracy "under optimal conditions," across a network running roughly 20 billion scans per month), the arithmetic yields up to 800 million potential misread records per month: wrong plates attributed to real vehicles in a searchable police database. (That is the conservative reading: 96% is character-level accuracy, and per-plate accuracy across a roughly seven-character plate is necessarily lower.) Field audits suggest the real rate is worse: Roseville, California found 71% of reads erroneous over two years. (Sources: Flock's 96% claim vs. Roseville audit; network scan volume)
• The studies reporting large clearance-rate gains from Flock cameras (e.g., a claimed 9.1% clearance increase) were commissioned by Flock Safety. The independent literature is thin and mixed. That's itself the point: warrantless mass collection is being justified by evidence that has not been independently established.

National context

• Thousands of Flock and other ALPR cameras are deployed across the United States; privacy advocates map them at deflock.me.
• As of August 21, 2026, DeFlock's national tracker counts 106 communities that have canceled, rejected, deactivated, paused, or banned Flock or ALPR programs; at least a dozen more exits are documented on this site's main page that the tracker has not yet logged, putting the true number above 115. Communities in 23 states acted in 2026 alone.
• Documented misuse: this site names 27 officers across 18 departments in nine states caught using Flock or similar systems for personal reasons (most often tracking women they knew), with national investigative tallies running from 50 to more than 70 cases.
• In documented cases elsewhere, local Flock data has been searchable by outside agencies, including federal immigration enforcement, through the national network.