Flock Safety Tightens ALPR Safeguards After Dozens of Officers Accused of Stalking
Flock Safety on August 14, 2026, unveiled a package of privacy and access controls for its automatic licence-plate reader (ALPR) platform, responding to reporting that at least 50 US law enforcement officers have been accused, charged, or convicted of using the systems to spy on individuals, including romantic partners and exes (Al Jazeera).
The Atlanta-based company operates approximately 120,000 cameras across the United States that capture vehicle images and licence plates for storage in a searchable database. A Washington Post investigation published in early August 2026 found that at least 46 officers had been accused of unauthorised tracking through Flock's cameras, including stalking wives, girlfriends, and former partners. As of August 13, 2026, that figure had risen to at least 50 accused, charged, or convicted officers across Flock's system and other licence-plate readers (Good Morning America). Cases include a police officer who allegedly used Flock cameras to track a boyfriend's ex-wife, according to a police chief (ABC News).
The changes Flock announced target three pressure points: data retention duration, access governance, and inter-agency data sharing. The company is reducing its recommended default retention period for captured images from 30 days to 7 days. An "evidence mode" will allow longer storage windows in exceptional cases. Flock will also require case codes for all law enforcement searches by the end of 2026, and is making an audit feature mandatory, one that documents all searches and flags abnormal activity. Users exhibiting abnormal behaviour will be locked out of the system for further review (Al Jazeera).
On the inter-agency front, Flock will let law enforcement agencies add offence-based filters that restrict which data other agencies can access. For example, an agency could configure its data to permit searches only for stolen vehicles, missing persons, or violent crimes while blocking immigration-enforcement-related queries (Al Jazeera).
The American Civil Liberties Union welcomed the shortened retention window but was dismissive of the remaining measures, calling them superficial or recycled versions of inadequate safety features (Al Jazeera). The ACLU's reaction points to a gap that has dogged ALPR oversight: the distinction between a technical capability and a binding constraint. Flock's own public statements have long maintained that every search of its system is recorded and reviewable, and that the platform cannot recognize, identify, or track individuals (Flock Safety). The company has also stated that more than 30 courts have ruled its licence-plate readers do not constitute mass surveillance, on the grounds that they capture images on public roadways rather than tracking movement (Flock Safety). Flock describes its platform as a network of individual cameras that submit vehicle data to a user interface (Flock Safety).
Yet the investigative record shows that internal safeguards of the kind Flock and other ALPR providers have long cited did not prevent misuse. As of April 2026, the Institute for Justice had documented at least 14 stalking cases involving licence-plate readers in recent years, with providers maintaining that internal controls were sufficient (Institute for Justice). The escalation from 46 accused officers in early August to at least 50 by August 13 suggests the investigative picture is still expanding.
The architecture of the new measures reveals where Flock sees the systemic weakness. Retention reduction shrinks the window in which misuse can occur, and the 30-to-7-day cut is the change the ACLU credited. Mandatory audits with automated flagging and lockout address the accountability gap that allowed officers to conduct searches without immediate detection. Offence-based filters on inter-agency access respond to a different concern: the proliferation of data-sharing agreements among departments, which broadens the universe of users who can query a given agency's camera feeds without that agency's direct oversight.
Whether these measures satisfy civil-liberties advocates or merely prompt the next round of scrutiny is the open question. The ACLU's characterization of most changes as recycled suggests that the debate will turn on enforcement and verification, not on feature announcements. Flock is also hosting "Flock Forward 2026," an Atlanta public safety conference whose listed topics include surveillance, stalking, and planned burglaries (Flock Safety), a programming slate that signals the company's awareness of the reputational stakes even as it contests the mass-surveillance label in court.
The broader context here is the tension between networked surveillance infrastructure designed for investigative utility and the access-governance regimes needed to prevent that same infrastructure from functioning as a tool of personal targeting. Flock's platform, with 120,000 cameras feeding a searchable database, gives individual officers a capability that historically required dedicated surveillance teams or physical tailing. The announced safeguards, structurally, attempt to reinsert friction into a system whose operational value depends on low-friction access. The retention cut, audit mandates, lockout triggers, offence-based filters, and case-code requirements each add a checkpoint where previously there may have been none, or only optional ones. Whether optional safeguards were the failure mode, or whether the problem runs deeper into the culture and oversight of the law enforcement agencies themselves, is a question the company's software changes can mitigate but not resolve.
