Flock’s Secret Weapon Against Police Misconduct: The Unexplained AI

Key Takeaways:

• Flock Safety announced new policies and its “Audit Assistance” tool to curb misuse of its automated license plate reader (ALPR) systems by law enforcement, including shorter data retention and mandatory auditing.
• While Flock touts the tool’s effectiveness, the company has provided scant technical details on how “Audit Assistance” works, leading to significant skepticism from privacy advocates and experts who demand independent audits and transparent data.
• Critics argue that technological fixes alone are insufficient to prevent abuse without robust legal frameworks, severe consequences for misuse, and public accountability for surveillance technology.

In a move designed to address escalating concerns over privacy and potential abuse, surveillance technology firm Flock Safety has unveiled a series of new policies and tools for its automated license plate reader (ALPR) camera systems. The announcement, made Thursday, signals a public effort by the startup to enhance accountability among its law enforcement clientele and mitigate misuse of its powerful tracking technology, which has increasingly come under scrutiny from civil liberties groups and the public alike.

Flock’s ALPR systems, deployed in thousands of communities across the U.S., capture and analyze vehicle data, offering a potent tool for crime solving but also raising profound questions about mass surveillance and potential civil rights infringements. The company’s latest initiatives aim to walk a tightrope: preserving the crime-fighting capabilities of its technology while attempting to build trust by bolstering oversight and transparency.

Proactive Steps: Data Retention and Sharing Controls

Central to Flock’s revised strategy is a recommendation for police departments to significantly reduce data retention periods. Previously, the company suggested a 30-day retention window; this has now been trimmed to a more conservative seven days. This change is intended to limit the scope and duration of data collection, thereby reducing the potential for long-term tracking and database aggregation. However, recognizing the practicalities of ongoing investigations, Flock also introduced an “Evidence Mode.” This feature allows customers to extend data retention beyond the standard seven days in “exceptional cases,” provided a specific case number is supplied. This mechanism aims to balance rapid data purging with legitimate law enforcement needs, albeit with a required justification.

Furthermore, Flock is empowering its customers with greater control over data sharing. New functionalities will enable departments to set limits on the information they share with other customers, specifically based on the type of offense under investigation. This granular control is a direct response to concerns that broad data sharing agreements could create vast, interconnected surveillance networks accessible for a wide array of inquiries, potentially unrelated to serious criminal activity.

Audit Assistance: The Core of Flock’s Accountability Push

Perhaps the most significant of Flock’s announced changes is the expansion and mandatory adoption of its “Audit Assistance” tool. Launched in April, the tool has already been voluntarily enabled by “more than one-third” of Flock’s customers. The company is now requiring all customers to activate it by the end of the year, underscoring its perceived importance in their accountability framework.

According to Flock, Audit Assistance is designed to “detect abnormal activity and flag it for Administrator review.” This includes the ability to automatically lock out users flagged for suspicious behavior until an administrator intervenes. Executives have been vocal in their praise for the tool. Ashley Haber, Flock’s head of trust and compliance, stated in a video that Audit Assistance can flag “what may look like an odd search history from a specific user,” providing insights into user operations. Co-founder Paige Todd echoed this, noting the tool “actually flags unusual patterns early. Think a user searching the same plate repetitively for more than 30 days.” Even a customer, Major Patrick Krieg of the Dunwoody, Georgia Police Department, described it as “an algorithm that has notified us of any type of bias.”

Despite these glowing endorsements, the precise mechanics of Audit Assistance remain frustratingly opaque. Flock describes it as a feature that can “surface atypical activity early and review it quickly with a clear, documented workflow,” but the technical details of how this “algorithm” functions, what data it’s trained on, and whether it employs large language models have not been fully disclosed. TechCrunch specifically pressed Flock for clarity on these points, including what kinds of patterns trigger flags and what precisely constitutes “abnormal activity.”

In response, Flock clarified that Audit Assistance is “not based on [Machine Learning] or AI.” Instead, the company explained, “It is a data tool that flags atypical search patterns for further review. One example that would trigger a flag is if a user searches for the same plate using multiple different case codes, as this could potentially indicate abuse. A flag does not mean that it is definitely abuse — it only means it warrants further investigation.” While this provides a sliver of insight, numerous critical questions remain unanswered. Flock has yet to provide statistics on the tool’s effectiveness, its false positive and false negative rates, or whether it has undergone any independent third-party audits—all crucial metrics for assessing the reliability and efficacy of such a system.

Privacy Advocates Remain Unconvinced

The lack of transparency has fueled significant skepticism among privacy experts and long-standing critics of Flock’s technology. Chad Marlow, senior policy counsel for the American Civil Liberties Union (ACLU), expressed profound doubt regarding the tool’s consistency. “There is no evidence that the tool works consistently to address abuse,” Marlow stated in response to Flock’s announcement. He pointed out the fallacy of using arrests for misuse as proof of the tool’s effectiveness: “While dozens of officers have recently been arrested, fired, or otherwise disciplined for misusing Flock for personal reasons, Flock claims these arrests are proof its auditing tool works. However, unless we know the number of officers misusing the system, we cannot conclude if Flock and its auditing tools are catching 95% of violators or 5%.” Marlow unequivocally called for an independent evaluation: “Flock needs to have its auditing tool analyzed by an independent evaluator to determine its actual effectiveness. Until then, we don’t know if the tool is a real security measure or just window dressing.”

Cooper Quintin, a security researcher and senior public interest technologist at the Electronic Frontier Foundation (EFF), took an even more cynical stance, suggesting that the tool’s efficacy might be irrelevant. “If it does work, law enforcement will figure out ways to get around it,” Quintin told TechCrunch. For Quintin, the ultimate measure of accountability isn’t the tool itself, but whether abusers face real consequences. “If there are no consequences for abuse and whoever [the tool] reports to is the same police agency that is doing the abuse, it’s just a fig leaf for Flock. The best way to ensure accountability is to pass laws constraining the use of this technology and requiring a warrant for its use.”

Rachel Levinson-Waldman, director of the Brennan Center’s Liberty and National Security Program, acknowledged that making Audit Assistance mandatory is a “step in the right direction” and that “in theory” it’s “a great idea.” However, she echoed the call for external validation, emphasizing that the tool itself “needs to be audited” due to the dearth of information on its practical operation.

Real-World Abuse and the Accountability Question

Flock’s latest announcements come against a backdrop of years of reported incidents where its customers have allegedly abused the technology. Just this week, three former Georgia sheriff’s deputies were arrested, accused of using Flock cameras to stalk individuals with whom they had “personal relationships.” Significantly, in this particular case, Bibb County Sheriff David Davis specifically credited Flock’s new Audit Assistance tool with uncovering the misuse. This incident, while cited by Flock and its partners as evidence of the tool’s utility, paradoxically highlights the very problem it seeks to address and reinforces the urgent need for a robust, independently verifiable accountability framework.

Bottom Line

Flock Safety’s new policies and its mandatory “Audit Assistance” tool represent an acknowledgment of the critical need for greater accountability in the deployment of powerful surveillance technologies. While these steps are framed as genuine efforts to curb abuse, the enduring opacity surrounding the tool’s technical workings and the absence of independent audits mean that skepticism from civil liberties advocates is well-founded. Until transparent data, verifiable effectiveness, and stronger legislative safeguards are in place, the debate over whether such technological fixes are truly effective or merely “window dressing” will continue to cast a long shadow over the future of ALPR systems and public trust.