For years, **Flock Safety** has assured the public that its technology "cannot recognize, identify, or track individuals." However, a recent investigation by WIRED reveals the company has built a sophisticated artificial intelligence tool for police that precisely does both. ### OS Investigate: A New Era of Surveillance The AI tool, initially named **Nightshift** and now rebranded as **OS Investigate**, leverages **Flock Safety**'s extensive network of cameras, which log vehicle movements across more than 6,000 communities. What sets **OS Investigate** apart is its ability to identify potential witnesses based on how frequently their cars pass through a neighborhood or to surface a driver's "associates" by analyzing their shared camera sightings. This system also integrates with police case files, 911 dispatch logs, and commercial identity records, allowing license plates to be linked to names, home addresses, and even relatives. Crucially, the system can search for individuals within a geofenced area based solely on a physical description. ### Pre-Written Prompts and Deep Data Access **OS Investigate** ships with 69 prewritten prompts that officers can select, review, and edit before submitting to the AI. Officers can also create their own custom prompts. WIRED's investigation uncovered these suggested prompts within a cache of over 450 files directly on **Flock Safety**'s website, accessible via its login pages. These files detail 45 distinct tools at the AI's disposal, granting access to: * License plate scans and camera metadata * Arrest records * Case files * Dispatch logs * Ballistics results * Commercial databases containing Social Security numbers, dates of birth, phone numbers, email addresses, relatives, and associates ### From Vehicle Lookup to Behavioral Profiling **Flock Safety** built its business on a simple premise: cameras photograph passing cars, convert plates to text, and check them against a list of wanted vehicles. The original assurance was that non-wanted cars were recorded but otherwise ignored. **OS Investigate** represents a significant shift, inverting this arrangement. An officer no longer requires a specific plate, name, or even a crime to initiate a search. Instead, they can supply a location, a timeframe, and a pattern of behavior, and the system is designed to return individuals who fit the criteria. One preloaded prompt, for instance, reads: "Find me witnesses based on vehicles most seen in [neighborhood] during [last 14 days] during [daily timeframe] *(will not include whitelisted vehicles)." The system then converts the resulting list of plates into names and home addresses. Another prompt instructs the system to list individuals arrested more than twice in two years for "any offense" (excluding narcotics), map their residences, retrieve service calls at their homes, and "do a workup on the top three individuals." A "workup," in **Flock Safety**'s terminology, is a one-command background check that retrieves departmental records and commercial data, including vehicles, prior suspect listings, relatives, phone numbers, and online accounts. ### Searching for Patterns, Not Just Suspects Nineteen of the prompts focus on identifying patterns rather than searching for known records. Fourteen of these require no plate, no name, and no description—only a location, time window, and behavior. Examples include requests for vehicles that visited three or more retail locations in a city in three days, multiple banks in a week, or multiple gas stations between midnight and 5 AM. These searches specifically filter out commercial vehicles, focusing on ordinary drivers. Similar pattern-based searches apply to travel between cities, identifying vehicles that made repeat roundtrips or passed through specific areas in sequence. ### Constitutional Concerns and Company Response Privacy and legal experts are raising serious constitutional concerns, arguing that these tools provide police with broad, often warrantless access to extensive records of people's movements. **Jay Stanley**, a senior policy analyst with the **ACLU's Speech, Privacy, and Technology Project**, notes, "Most people imagine that license-plate readers are checking license plates and they’re looking if a vehicle is wanted. But we’re increasingly seeing the aggressive deployment of AI by **Flock**—and in some ways, like China, very little space between what can be done and what is being done." **Noel Pichardo**, a former Pawtucket, Rhode Island, police officer who reviewed the prompts, stated, "I don’t know how else to say this, but this sounds completely insane... I don’t know how anyone can argue against the idea that **Flock** literally tracks people." **Flock Safety** spokesperson **Paris Lewbel** characterized **OS Investigate** as a separate product from its license-plate-reader technology, designed to help investigators work across information agencies already possess. He indicated that the tool's capabilities and workflows might change significantly before a broader release. This development comes amidst growing bipartisan political pressure, an increasing record of officers misusing **Flock**'s platform, and a wave of vandalism targeting its cameras across the U.S.