# U.S. Flock Safety Issue, Misuse, Litigation, and Governance Research Index

**Snapshot verified through:** 2026-07-21  
**Incident/case/policy records:** 280  
**Source documents and articles:** 88  
**States represented:** 40  
**Legal/court records:** 19

## What this is

This is a broad, source-auditable public-interest index of reported issues involving Flock Safety and its automated license-plate-reader or related camera ecosystem in the United States. It covers documented or reported misuse, audit-log findings, false alerts and wrongful stops, security exposures, leaked product plans, data-sharing disputes, lawsuits and court rulings, local contract actions, legislation, permitting/safety questions, efficacy disputes, and material company or law-enforcement counterpoints.

The project does **not** assert that every allegation is true or that every deployment is unlawful. Each record has a `claim_status`, `source_type`, and `evidence_strength` field so allegations, government findings, court decisions, municipal actions, investigative reporting, advocacy claims, and company statements remain distinguishable.

## Important limitation: “every single” cannot be guaranteed

No public web scan can prove completeness. Flock deployments and controversies change rapidly; many audit logs, contracts, disciplinary files, settlements, and lower-court records are not indexed; news sites can be paywalled or deleted; and some incidents are described only in local meeting packets or records requests. This release is therefore a **wide-ranging best-effort snapshot**, not a claim that no other incident exists.

The source library intentionally avoids counting every syndicated rewrite as a separate finding. The main CSV records underlying incidents/cases; `flock_safety_sources.csv` retains the collected source trail, including corroboration, responses, legal documents, and context.

## Coverage in this release

- **Discrimination / profiling:** 82
- **Mission creep:** 45
- **Protest / First Amendment:** 30
- **Personal / insider misuse:** 25
- **Immigration / data sharing:** 23
- **Local action / regulation:** 23
- **Wrongful stop / false alert:** 21
- **Litigation / court ruling:** 19
- **Security / product / installation:** 7
- **Company / stakeholder context:** 3
- **Reproductive-health surveillance:** 1
- **Governance / efficacy debate:** 1

Most represented states by record count: IL (50), CA (40), TX (15), MI (12), VA (12), FL (11), SC (11), OH (11), CO (11), WA (9).

Nationwide audit-log investigations generate many agency-level records. For example, one source may identify dozens of agencies that used ethnicity-related terms, protest-related search reasons, low-level administrative searches, or immigration hotlists. Those agencies are mapped as separate records because the underlying conduct and jurisdiction are distinct.

## Files

- `flock_safety_us_issues.csv` — main research table, one underlying incident/case/policy finding per row.
- `flock_safety_sources.csv` — source index, one article/document/docket/tool per row.
- `flock_safety_data_dictionary.csv` — field definitions.
- `flock_safety_us_issues.geojson` — GIS-ready points with record properties.
- `flock_safety_us_issues_map.html` — interactive filterable map and record browser.
- `flock_safety_quality_report.json` — validation counts and warnings.
- `build_flock_research.py` — reproducible build script used for this release.
- `eff_flock_roma_searches.xlsx` — public EFF source spreadsheet used to create the agency-level Roma-search records.

## Inclusion rules

A record was included when a collected source materially tied Flock to at least one of the following:

1. Alleged or documented misuse, discriminatory/protest/sensitive searches, mission creep, or unauthorized access.
2. A false alert, mistaken stop, detention, arrest, or systemic data-quality problem.
3. A lawsuit, court ruling, government enforcement action, public-record dispute, or constitutional challenge.
4. A security exposure, leak, product-governance issue, contractor-access concern, installation/permit question, or physical-safety concern.
5. A municipal contract cancellation, shutdown, restriction, replacement, expansion controversy, or state legislative response.
6. A material company, police, or stakeholder response necessary to understand a contested claim.

Purely promotional deployment announcements and routine crime-arrest stories were excluded unless they supplied a necessary counterpoint or directly related to an issue in the index.

## Deduplication

The record unit is the **underlying event or agency finding**, not the headline. Multiple sources about the same LAPD contract expiration, for example, are linked to one policy record; a separate systemic false-alert investigation remains a different record. Conversely, an umbrella audit that identifies many agencies produces separate jurisdictional records.

## Evidence-strength rubric

- **High:** court document, government audit/ruling, source spreadsheet or audit logs, detailed documentary investigation, or strongly corroborated local action.
- **Medium:** credible news or advocacy compilation with clear attribution but limited underlying documentation in the collected set, or a newly raised concern not yet adjudicated.
- **Low:** company self-report, marketing/impact claim, or contextual policy statement included to preserve counterevidence and stated safeguards.

Evidence strength does not mean moral severity, and it is not a prediction of who will win an active lawsuit.

## Mapping and privacy

Coordinates are **display points only**. City/county centroids or state/national centroids receive small deterministic jitter so overlapping records can be clicked. They are not exact camera locations, homes, traffic stops, plaintiffs’ addresses, or surveillance targets. Exact plate numbers and private target identities are excluded.

For crowdsourced camera-location research, see the separate DeFlock resource (source S084). For searchable audit-log reasons, see Have I Been Flocked? (source S085). Their content is not copied into this project.

## Updating the index

To extend the project:

1. Add a source to the source table with publication date, type, perspective, and URL.
2. Add or amend an underlying record; do not create a duplicate just because another article covered it.
3. Preserve attribution words such as “alleged,” “reported,” “audit found,” or “court held.”
4. Record material agency/company responses and later court dispositions.
5. Re-geocode only to a jurisdiction centroid; never publish exact camera or target coordinates in this map.
6. Update `last_verified` and rerun validation.

## Validation notes

- None.

## Editorial note

This tool is designed to support investigation, accountability reporting, legal research, policy analysis, and public debate. It should not be used as a substitute for checking the cited source, docket, or current law. Active cases and local policies can change after the verification date.