Written in my capacity as Technology Chair, Restore The Fourth. This is a personal technical brief, not legal advice, and does not constitute a formal position of Restore The Fourth national organization without separate ratification. Expert reviewed.

Abstract

The Fourth Amendment doctrine developed for physical searches is structurally inadequate for AI-powered mass surveillance. Automated License Plate Readers (ALPRs), predictive analytics platforms, and commercial data fusion are creating searchable records of population movement, association, and behavior at a scale and resolution that no warrant system was designed to govern. This brief documents the technical architecture, identifies the doctrinal gaps, and proposes technical and legislative remedies.

1. The Technical Infrastructure

1.1 Automated License Plate Readers

Flock Safety deploys fixed and mobile ALPR cameras that capture license plates, vehicle make/model/color, and associated metadata (timestamp, GPS coordinates, direction of travel) for every vehicle that passes within range. Data is uploaded in near-real-time to a cloud platform accessible to subscribing law enforcement agencies.

Key technical properties:

• Retention: Default 30-day retention; configurable per agency.
• Sharing: Cross-agency sharing is a feature, not an edge case. A plate hit in Albany can generate an alert in a subscribing agency three states away within seconds.
• Enrichment: Flock's "Vehicle Fingerprint" feature captures features beyond the plate — roof racks, bumper stickers, damage patterns — enabling tracking of vehicles with obscured or stolen plates.
• Network scale: As of 2025, Flock operates in over 5,000 communities across 42 states.

1.2 Video Intelligence Fusion — Fusus

Fusus aggregates private camera feeds (businesses, residential complexes, institutions) into a unified law enforcement interface. Participating private parties install a Fusus "core" device that streams their camera feed to the platform. Law enforcement accesses the unified feed through a dashboard.

The constitutional significance: private parties are voluntarily sharing surveillance capability with law enforcement, potentially circumventing the warrant requirement that would apply to direct government surveillance of the same locations.

1.3 Predictive and Behavioral Analytics

The combination of ALPR data, video feeds, and commercial data broker records enables behavioral pattern analysis: who goes where, when, with what frequency, with whom. This is not a theoretical capability — it is the marketed feature set of platforms sold to law enforcement today.

2. The Doctrinal Gap

2.1 The third-party doctrine

Smith v. Maryland (1979) established that information voluntarily shared with a third party loses Fourth Amendment protection. In 1979, "voluntary sharing" meant calling a phone number — a discrete, chosen act. In 2026, driving on a public road generates data in Flock's platform whether you choose to or not. The third-party doctrine, applied literally, eliminates constitutional protection for the movement patterns of every person who drives, parks, or is photographed on a street with camera coverage.

2.2 Carpenter and its limits

Carpenter v. United States (2018) held that seven days of cell-site location information requires a warrant, reasoning that "seismic shifts in digital technology" require updating Fourth Amendment doctrine. The Court explicitly limited its holding and declined to address "other collection techniques involving foreign affairs or national security."

The Carpenter majority did not address: ALPR data, commercial data brokers, video surveillance fusion, or predictive analytics. Lower courts have split on whether Carpenter's reasoning extends to ALPR data. Several have held it does not, treating license plate readers as equivalent to a police officer observing traffic — a comparison that ignores the persistent, searchable, cross-jurisdictional nature of ALPR records.

2.3 The mosaic theory

Justice Sotomayor's Jones concurrence (2012) articulated a mosaic theory: that aggregation of individually non-sensitive data points can create a picture whose collection would violate the Fourth Amendment even if each individual data point would not. The theory has academic support but has not been adopted by a majority of the Court. It is, however, the correct framing for AI-powered surveillance — the harm is in the aggregate, not the individual observation.

3. What Can Be Done

3.1 Technical countermeasures

For individuals: ALPR evasion is legally and practically limited. What is available: understanding coverage geography (public records requests for camera placement), minimizing predictable route patterns, awareness that rental vehicles and ride-share trips generate ALPR records under the fleet account rather than personal identity.

For communities: The most effective technical countermeasure is political — demanding data retention limits, cross-agency sharing restrictions, and audit logs as conditions of municipal ALPR contracts before they are signed.

3.2 Legislative approaches

Several states have enacted ALPR-specific legislation. The strongest frameworks share common elements:

• Mandatory retention limits (30 days or less without judicial order)
• Prohibition on cross-agency sharing without specific legal process
• Mandatory audit logs of all queries
• Annual public reporting requirements
• Civil cause of action for violations

New York has not enacted comprehensive ALPR regulation as of this writing. The NYCLU Capital Region chapter is a coalition partner on this work.

3.3 Litigation strategy

The strongest current litigation theory combines Carpenter's "seismic shift" reasoning with mosaic theory to argue that persistent, cross-jurisdictional ALPR record aggregation constitutes a search requiring a warrant. The argument is stronger in circuits that have engaged seriously with Carpenter's implications.

A secondary theory — that Fusus-style private/public surveillance fusion constitutes state action subject to Fourth Amendment constraints — has not been tested at the appellate level but is legally plausible under agency theory.

4. Conclusion

The surveillance infrastructure being deployed in American cities today was not designed to comply with constitutional doctrine — it was designed to avoid triggering it. Closing the doctrinal gap requires parallel action: litigation to extend Carpenter, legislation to impose data minimization requirements, and technical work to make surveillance infrastructure legible to the communities it operates in.

Restore The Fourth's Technology Working Group is actively working on all three tracks. If you are a technologist, lawyer, or organizer who wants to contribute, contact us at restorethe4th.com.

Expert reviewed. This brief reflects the author's analysis as of the publication date. The legal landscape changes. Verify current case law before relying on doctrinal statements for any legal purpose.