Flock Safety, the surveillance technology company known for its automated license plate reader (ALPR) systems, issued a major policy and product overhaul on Thursday, introducing a suite of changes designed to address growing concerns about police abuse of its technology. The centerpiece of the announcement is a tool called “Audit Assistance,” which the company touts as a way to detect and stop misuse by law enforcement officers. Yet, despite its prominent billing, Flock has provided remarkably little detail about how the tool actually functions, leaving privacy experts, civil liberties advocates, and the public to question whether this is a genuine accountability mechanism or a carefully managed public relations effort.
What Did Flock Announce? The New Policies and Tools Explained
Flock’s Thursday blog post outlined several key policy shifts and technical features. The most significant policy change involves data retention. The company now recommends that police departments retain license plate data for just seven days, a sharp reduction from the previous standard of 30 days. This change is accompanied by a new feature called “Evidence Mode,” which allows customers to retain data for longer periods only in what Flock describes as “exceptional cases.” To activate Evidence Mode, a specific case number is required, ostensibly creating a clear audit trail for any data that is kept beyond the new seven-day window.
Additionally, Flock is introducing more granular data-sharing controls. Customers can now set limits on the data they share with other law enforcement agencies based on the type of offense under investigation. This is intended to prevent the kind of widespread, indiscriminate data pooling that critics argue enables mass surveillance.
However, the most heavily promoted feature is “Audit Assistance.” Flock launched this tool in April 2026. The company reported that “more than one-third” of its customers have already enabled it, and it is now requiring all customers to turn it on by the end of the year. According to Flock’s blog post, Audit Assistance “detects abnormal activity and flags it for Administrator review.” When such activity is detected, the tool can automatically lock out the flagged user until an administrator intervenes. Flock claims the tool has already helped customers spot and stop abuse.
How Does Flock’s Audit Assistance Tool Work? The Vague Answers
The central tension in Flock’s announcement lies in the gap between the tool’s stated importance and the lack of technical transparency. The company has not explained in detail how Audit Assistance actually works. While some official comments provide a general sense of its capabilities, the specifics remain shrouded in ambiguity.
Ashley Haber, Flock’s head of trust and compliance, said in a video that Audit Assistance can flag “what may look like an odd search history from a specific user,” helping customers understand how their users are operating the system. Flock co-founder Paige Todd provided a more concrete example in a separate video, stating that “Audit Assistance actually flags unusual patterns early. Think a user searching the same plate repetitively for more than 30 days.” Major Patrick Krieg of the Dunwoody, Georgia Police Department, a Flock customer, described the feature in a press release as “an algorithm that has notified us of any type of bias.” Flock’s own website describes the tool as a feature that can “surface atypical activity early and review it quickly with a clear, documented workflow.”
These descriptions, while illustrative, prompt a series of deeper, unanswered questions. How does this “algorithm” actually work? What data is it trained on? Was it developed using a large language model? What kind of patterns is it designed to surface? What exactly gets flagged as “abnormal activity”?
When pressed by TechCrunch, Flock provided a direct but limited answer: Audit Assistance is not based on machine learning or AI. The company explained that “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.”
This explanation, while clarifying that the tool is rule-based rather than AI-driven, still leaves numerous questions unanswered. Does Flock have any statistics or data showing the tool’s effectiveness? What are its false positive and false negative rates? Has it been audited by a third party? The company has not provided this information.
What is an “Atypical Search Pattern” According to Flock’s Audit Tool?
For users seeking a clear, standalone definition, an “atypical search pattern” as defined by Flock’s Audit Assistance tool is a rule-based data flag. The system is designed to detect specific, unusual user behaviors, such as an officer searching for the same license plate using multiple different case codes or searching the same plate repetitively over an extended period. These patterns are identified not through machine learning but through a pre-defined rule set. The flag itself is not an accusation of abuse; it is a trigger for human review by an administrator. The tool can also automatically lock out the flagged user until a supervisor intervenes, creating a forced pause in the investigation of the user’s activity.
Flock’s Track Record: A History of Abuse by Law Enforcement Customers
The context for Flock’s announcement is a long and well-documented history of its customers abusing the company’s technology. There have been numerous reported incidents of law enforcement officers using Flock cameras for personal reasons, including stalking and harassment. The problem has been persistent and widespread, with officers repeatedly getting arrested for misusing Flock systems.
Just this week, three former sheriff’s deputies in Georgia were arrested and accused of using Flock cameras to stalk people with whom they had a “personal relationship at the time,” according to a press release from the Bibb County Sheriff’s Office. In a notable twist, Bibb County Sheriff David Davis specifically credited Flock’s new Audit Assistance tool for uncovering the misuse in this particular case. This anecdote provides a powerful, real-world example of the tool’s potential, but it also underscores a fundamental question: how many other incidents go undetected?
Critical Reactions from Privacy Experts and Civil Liberties Groups
Privacy experts and critics of Flock are not impressed by the Audit Assistance announcement. Their skepticism centers on the lack of independent verification and the inherent limitations of a self-policing system.
Chad Marlow, senior policy counsel for the American Civil Liberties Union (ACLU), wrote a pointed response to Flock’s announcement, stating that “there is no evidence that the tool works consistently” to address abuse. He argued that Flock’s claims of success are misleading. “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,” Marlow wrote. “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%. 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), offered an even more fundamental critique. He told TechCrunch that it may not even matter whether Audit Assistance works. “If it does work, law enforcement will figure out ways to get around it,” he said. He emphasized that the ultimate measure of accountability is whether abusers are caught and face 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, offered a more tempered but still skeptical view. She acknowledged that making Audit Assistance mandatory is a step in the right direction and that “in theory” the tool is “a great idea.” However, she echoed the core concern: the tool itself “needs to be audited” because there is not enough information about how it works in practice. Without independent verification, the tool’s effectiveness remains an article of faith rather than a proven fact.
The Deeper Problem: Self-Policing vs. Independent Oversight
The debate over Audit Assistance highlights a deeper structural issue in the relationship between surveillance technology companies and their law enforcement customers. Flock is essentially asking the public to trust that its software will police the police. This model has inherent limitations. The audit tool is deployed and managed by the same police departments that may be abusing the system. The administrator who reviews the flagged activity is an employee of the police department, not an independent third party.
This creates a conflict of interest. A police department has a strong incentive to minimize reports of abuse, both to protect its public image and to avoid legal liability. An internal audit tool, no matter how well-designed, cannot fully overcome this institutional pressure. The EFF’s Quintin captured this dynamic perfectly, describing the tool as a potential “fig leaf” that allows Flock to claim it has addressed the problem while deflecting calls for more meaningful reform, such as warrant requirements or outright bans on certain uses of ALPR technology.
What Flock Still Needs to Explain
For Audit Assistance to be taken seriously as a genuine accountability mechanism, Flock must provide far more transparency. The following questions remain unanswered:
- What are the specific rules? Flock must publish the complete rule set that governs what constitutes “abnormal activity.” Without this information, the public cannot assess the tool’s scope or potential biases.
- What are the effectiveness metrics? Flock should disclose the tool’s false positive and false negative rates. How many users are flagged? Of those, what percentage are found to be actual abusers? How many abusers are not being flagged?
- Has the tool been independently audited? If Flock is serious about accountability, it should commission an independent third-party audit of Audit Assistance and publish the results in full.
- What happens after a flag? Flock should provide a standardized workflow for what happens after a user is flagged and locked out. Is there a requirement for the case to be referred to an external oversight body, or is it handled entirely within the police department?
- How does Flock ensure the tool itself is not abused? Could a malicious administrator use the tool to target a whistleblower? What safeguards are in place to prevent the audit system from being used as a weapon against legitimate users?
Flock’s decision to reduce data retention from 30 days to 7 days is a meaningful concession to privacy advocates. The introduction of Evidence Mode and granular data-sharing controls also represents a tangible improvement over the previous, more permissive framework. These changes demonstrate that Flock is listening, at least partially, to the outcry over its technology.
However, the company’s opaque handling of Audit Assistance undermines the credibility of these other reforms. By promoting a self-policing tool while refusing to explain its inner workings, Flock risks creating the perception that it is more interested in managing public relations than in ensuring genuine accountability. The ultimate test of Audit Assistance will not be how many officers it flags, but whether it can withstand independent scrutiny and whether it leads to a verifiable reduction in abuse. For now, the burden of proof remains firmly on Flock. Until the company provides the detailed technical and operational information that critics are demanding, the most generous interpretation of Audit Assistance is that it is a promising but unproven experiment. The less generous, and more likely, interpretation is that it is the latest chapter in the long and troubled story of technology companies selling surveillance tools without adequate safeguards, leaving the public to suffer the consequences.
