Flock is not a camera. It is a searchable, cross-agency database with a camera on the front end.
What is Flock actually protecting?
Every "just a camera" comparison depends on you not asking this. So let's ask it.
Every other camera in your day-to-day life protects something specific, owned by whoever installed it. That is the whole point of installing a camera. Flock is the only one where the answer doesn't fit.
- Store surveillance cameraProtects the store's inventory and staff. The store owns it.
- Ring / Nest doorbellProtects the resident's household. The resident owns it.
- DashcamProtects the driver in the event of a collision. The driver owns it.
- Bank / ATM cameraProtects the bank's cash and customers. The bank owns it.
- Red-light cameraEnforces a specific traffic violation at a specific intersection. Authorized by state statute.
- Flock ALPR cameraProtects nothing you can point to. Owned and operated by Flock Safety, a private vendor headquartered in Atlanta.
What Flock is not.
These are the eight comparisons that come up in every real conversation about Flock. Here is what actually separates each one from the ALPR network on OKC's streets.
Flock vs. a Ring doorbell
- One property, one owner, one storage. A Ring covers a single doorway on a single house. The homeowner decides who sees the footage.
- The homeowner controls access. Police need a subpoena or the owner's consent to pull the recording. There is a legal process.
- The homeowner can turn it off. Unplug it, opt out of cloud storage, delete the app. Done.
- Flock is thousands of cameras on public roads, owned by a private vendor, feeding a database no one in OKC city government audits. There is no doorway you can move to that is not covered.
Flock vs. a store surveillance camera
- The store owns the camera, the footage, and the storage. They can hand you the recording tomorrow or tell you no. That is a private property decision.
- Store footage is not queryable across a national network of other stores. A camera at Target does not talk to a camera at Walgreens across the country.
- When police want the footage, they subpoena the store. The store's lawyer looks at the request. Legal process exists between "police want to see this" and "police see this."
- The store camera protects specific merchandise. Flock protects nothing you can point to, and its output is available to any agency on the network without the store's-lawyer step in between.
Flock vs. a dashcam
- A dashcam records only what one driver drives past. One vehicle, one perspective, ephemeral. Most overwrite themselves in hours or days.
- No searchable database. No cross-driver query. Nobody can query "show me every dashcam that saw this plate last Tuesday."
- No state actor operating it. The driver owns the footage. If police want it, they ask, and the driver decides.
- You installed it and you own it. Flock was installed on a pole in your neighborhood and you did not consent to it. Different category.
Flock vs. red-light and traffic cameras
- Red-light and speed cameras have a single narrow purpose defined by statute. Enforce a specific traffic violation at a specific intersection. That's it.
- The data flow, retention, and use are limited by law. If the city wants to use the footage for something else, that is a policy fight the public gets to have.
- The footage is not aggregated into a national vehicle-movement database. A red-light camera in OKC does not share with an agency in Kansas or Texas.
- Flock captures every vehicle at every location, not just the ones committing a specific violation, and feeds a queryable history the operator can search retroactively across the network.
Flock vs. "you're on camera everywhere anyway"
- Being seen is not being tracked. A camera that records a hallway records that a person walked through a hallway. Flock records where every plate has been at every intersection with a Flock camera, so a single search produces a person's movement history.
- This is the aggregation problem, and the Supreme Court has already addressed it. Carpenter v. United States (2018) held that persistent, aggregated location data is a Fourth Amendment search even when each individual data point was public.
- You retain some expectation of privacy in public. Katz v. United States (1967) set that standard. Being visible to a passerby is not the same as being logged in a searchable database.
- The comparison collapses time. "On camera" today usually means "recorded, then overwritten." Flock is "recorded, then searchable across every camera in the network for the retention window."
Flock vs. Google, Facebook, and the tech giants
- Google and Facebook operate under (weak, but present) opt-in consent. You clicked a Terms of Service. In theory you can use different products, block cookies, or delete the account.
- They are not the state. They cannot arrest you, pull you over, or execute a warrant. That is a bright legal line.
- Flock is used by police. The output triggers stops, searches, and arrests. Fourth Amendment protections apply to state actors, not to private companies.
- "Google is worse" is not an argument for expanding state surveillance. It is an argument for regulating private surveillance too. That is a separate conversation, and one worth having.
Flock vs. "license plates are public"
- True. And irrelevant. Your face is public too. That does not mean the government can build a nationwide database of every location where every face was seen.
- The plate is public. The searchable history of where it has been is not. Those are two different pieces of information governed by two different sets of expectations.
- Oklahoma statute already recognizes the distinction. 47 O.S. § 7-606.1 defines when and how license plate reader data may be collected, retained, and used - precisely because "the plate is public" was never a full authorization to build a movement database.
- The pattern is the private thing. Where you were on Tuesday, then Wednesday, then the following Sunday, is not a public fact just because each individual data point was.
Flock vs. "you already carry a phone"
- The state forces the plate it tracks. A phone is something you chose to buy. A license plate is something the government ordered you to bolt to your car and keep clearly visible every time you drive (Oklahoma law, 47 O.S. § 1113), and covering or obscuring it is itself a citable offense. The only way to stop Flock from logging your tag is to break that law or stop driving. You cannot consent to a mandate, and a choice you are forbidden to decline is not a choice.
- A phone is a thing you can put down. Whatever your phone or connected car reports, you still have doors: airplane mode, leaving it at home, declining the connected-services subscription, driving an older vehicle. Flock has no door. It photographs every car that passes, offers no opt-out and no setting, and will not even tell you which roads it watches.
- The Supreme Court already heard this argument and threw it out. In Carpenter v. United States (2018) the government argued that because you voluntarily carry a phone that logs your location, you have no privacy claim in where it has been. The Court rejected it, holding that a person does not "voluntarily assume the risk" of tracking just by carrying a device modern life has made unavoidable. "You have a phone" is the losing side of that case.
- Your phone company is more protective than your city. To pull your location history from your carrier, police generally need a warrant, because of that same Carpenter ruling. To pull it from Flock, they need nothing but a login. The company this argument points to guards your movements better than the government that bought the cameras.
- "You're already tracked" is not a permission slip. That a connected car sells its owner's driving data is a scandal, a Republican attorney general sued General Motors over exactly that, not a reason to let the government build a second tracker on a public pole. The remedy for too much surveillance is less of it, not a new system aimed at everyone.
- Turn it around. If there is truly no privacy in public, the rule should start with the officials who voted yes. Let them post their own daily driving logs, and let the chief publish the 109 agencies he refused to name to the council. It stops being "public" the moment it is their car.
Every reason they give, answered.
These are the arguments councilmembers and proponents actually make for keeping the cameras. Here is what each one leaves out.
But the department is hundreds of officers short
- $270,000 buys zero officers. The Flock contract costs $270,000 a year and hires not one cop. That money leaves the department and goes to a private vendor. If you are short hundreds of officers, a surveillance subscription is not a paycheck, and it will never put a badge on the street.
- That's Flock's sales pitch, not a fix for the shortage. Flock markets its cameras as a "force multiplier" for understaffed departments, and it features Tulsa PD's staffing shortage as a sales case study on its own website. Chiefs in East Chicago and Mount Juliet say on camera the cameras do the work of officers they no longer hire. A councilmember using Flock to cover a hiring gap is confessing the business model, not solving it.
- The tools that actually solve crimes already exist. OKCPD already runs a real-time crime center that does not depend on Flock, and officers have been able to run a plate on a suspect since 2014, by Chief Bacy's own words. A camera does not answer a 911 call, investigate a case, or testify in court. Every one of those still takes an officer.
- Bad alerts burn the officers you cannot spare. An LAPD Inspector General audit found Flock stolen-vehicle alerts were wrong nearly one in three times. Every false hit sends a scarce officer chasing the wrong car. When you are already short-staffed, wasting your people on the vendor's mistakes is the last thing you can afford.
But it solves crimes and recovers stolen cars
- Their own paperwork says the record does not exist. OKCPD told a records requester in writing (OCPD-10088) that it keeps no records linking a Flock search to a case outcome. You cannot call a role documented when the department admits it documented nothing. The chief's numbers, 60 murders, 1,280 vehicles, are a slide, not a case file. Ask for the receipts and the city has none to give.
- Anecdotes are not a clearance rate, and the clearance rate is falling. American police solved about 9 in 10 murders in the 1960s with no cameras at all. Today, armed with every surveillance tool ever built, they clear barely half, 52.3% in 2022. Whatever solves murders, mass surveillance is demonstrably not the driver. A camera does not answer a 911 call, canvass a block, or testify in court.
- It is wrong often enough to put innocent people at gunpoint. An LAPD Inspector General audit found Flock stolen-vehicle alerts were false nearly 1 in 3 times, 32.3%. Police work a stolen-car hit as a high-risk stop, so a wrong alert aims that stop at an innocent driver. Recovering some cars does not erase the traffic stops Flock manufactures on people who did nothing.
- It cannot be both the modest plate-checker and the crime-fighting engine. Either it is the simple tool they sold the council, in which case OKCPD ran plates on suspects since 2014 and runs a crime center without it, or it is a database tracking where everyone drives, which is the aggregated-movement problem Carpenter v. United States left wide open. They pick whichever answer wins the room. Pin them to one.
You cannot audit what they will not show you
- The oversight is Flock's product, not the city's policy. By Chief Bacy's own words on August 18, the retention change and the audit feature are things Flock "created" and "added." Those are vendor product settings, and Section 2.4 of the contract lets Flock change the platform whenever it wants. The city did not write a rule. It accepted a software update, and what a vendor adds, a vendor can take back.
- The audit is a claim you are not allowed to check. A resident asked the city for the audit log so the public could see it working. The city denied the request (OCPD-10716). An audit nobody outside the department can inspect is not oversight, it is a press release. If the safeguards were real, the log would be the easiest thing in the world to hand over.
- It all appeared only after residents forced it. The March 10 memo (OCPD-2885) admitted in writing that there were no access controls, no audit procedures, no discipline standards, and no transparency reporting. The fixes arrived after people filed records requests and packed the chamber. Oversight that shows up only when the public is watching is not a standard, it is damage control.
- The councilman who defended the safeguards is still asking whether they work. Explaining his yes vote, Carter called the renewal the responsible decision because of its oversight, then in the same statement said he would "continue asking questions about access, retention, data sharing, auditing, misuse, and whether the safeguards surrounding this technology are actually working" (KOKH). A safeguard you have to keep asking about is not one anyone has verified.
- The claim that officers use it only on real active cases is a promise the system is built to break. By Chief Bacy's own confirmation on August 18, officers are not required to enter a case number to run a search, so nothing requires a lookup to be tied to an active investigation. In the same meeting, he conceded the department had already had misuse "incidents with two employees." A tool that requires no case number, and that the department admits has been misused, cannot be vouched for as always tied to real police work.
- Even a perfect audit answers the wrong question. "Show the guardrails" is a line from Flock's own leaked coaching guide for police. And the best audit imaginable only polices how the dragnet gets used. It never asks whether a private company should be running a plate-tracking net over every driver in the city in the first place. That is the question, and the audit is built to change the subject.
"Nothing to hide"
- It reverses who owes whom an explanation. In a free country the government justifies watching you; you never justify your privacy. A permanent movement database logs the innocent by design, because the innocent are the ones driving to church, the doctor, the gun range, and the recovery meeting every single day. That is not catching wrongdoers. That is cataloging everyone.
- "I don't care" is not your call to make. The cameras do not log only the people who shrug at surveillance. They log everyone. Saying you do not care about the privacy issue does not waive your privacy, it waives your neighbors', your family's, and everyone's who never agreed to be in the database. One person's shrug is not consent for a whole city.
- The danger is not you. It is whoever holds the record. "Nothing to hide" assumes the watcher is always honest. Flock stolen-vehicle alerts were wrong nearly 1 in 3 times in an LAPD Inspector General audit, so innocent drivers can get stopped over a machine's mistake. And when an OKC resident asked to see the audit log that supposedly guards against abuse, the city denied the request (OCPD-10716). You cannot check the checkers.
- You are not handing this to the police. You are handing it to a vendor. Contract Section 4.1 gives Flock an irrevocable license to the data and Section 4.3 a perpetual right to train its AI on it. Your daily movements become a private company's searchable, permanent asset. "Nothing to hide" was never about hiding wrong. It is about not surrendering a free person's whole life to a corporation you never voted for.
- Ask the person saying it to go first. Would they bolt a GPS to their own car and publish their daily driving log for anyone with database access to read? Nobody accepts that, and for good reason. Privacy is not a confession. It is the baseline every Oklahoman already lives by until a camera network takes it without asking.
Three things that make Flock different from every "just a camera" comparison.
If you take nothing else from this page, take these three sentences.
Flock is not a camera. It is a searchable, cross-agency database.
The camera is the front end. The database is the product. Every "just a camera" comparison misses that Flock's value to a police department is the ability to search history across a network the department itself does not operate.
The searches are automated and retroactive.
Traditional cameras produce footage a human has to review after something happens. Flock produces a queryable history of vehicle movement that any authorized user on the network can search at any time, going back through the entire retention window, without notice to the person being searched.
OKC does not own or fully control it.
The June 15, 2026 rewrite of the OKCPD Operations Manual, made only after the program came under public scrutiny, handed the "System Administrator" sole approval over cross-agency data sharing. The manual itself defines that role as "manages the technical aspects of the server." That is not a security-camera problem. That is a data-governance problem.
What Flock actually is comparable to.
If Flock is not a store camera or a Ring doorbell, what is it? These are the honest comparisons.
Four things Flock actually resembles.
Naming what Flock is comparable to sharpens what it is not. We are not saying Flock is uniquely evil. We are saying it belongs to a category of surveillance the country already regulates under different rules.
- Cell-site location dataThe Supreme Court held in Carpenter v. United States (2018) that seven days of cell-site location data is a Fourth Amendment search because the aggregation is so revealing of a person's movements. Flock's ALPR pool works on the same aggregation logic. It has no equivalent warrant framework.
- A national fingerprint or DNA databaseCross-agency. Indexed. Searchable. Retroactive. Every state and federal agency has rules governing who can search these systems and under what conditions. The rules exist because the aggregation is powerful. Flock has no equivalent framework.
- Utility smart meter aggregationEach reading is mundane. Aggregate them and you can tell when someone is home, awake, or on vacation. Same "each data point is public" logic, same aggregation problem, and a growing patchwork of state regulations trying to catch up. Flock is not in that patchwork.
- State-pooled license plate reader systemsSome states operate ALPR pools at the state level with statutory retention limits, audit requirements, and public reporting. Flock is a private vendor operating a national network with none of those constraints, sold agency-by-agency.
The two arguments that sound unanswerable.
Every council chamber eventually hears these two. They sound like conversation-enders. Read closely and each one is an admission.
"But these cameras help solve murders."
- The argument proves too much. Warrantless searches of every home would also help solve murders. So would a GPS tracker on every ankle. "It would help catch criminals" is true of every surveillance measure ever proposed - which is exactly why the Fourth Amendment exists. The question was never whether it helps. It is what it costs everyone who will never be a suspect.
- The brag is a confession. A camera that checks plates against a stolen-car hotlist cannot solve a murder. The only way Flock helps a murder case is by reconstructing where a specific person traveled, retroactively, across weeks of stored data. Every murder anecdote is an admission that the system is not the plate-checker that was sold to council. It is a movement-tracking database - the exact thing Carpenter v. United States warns about.
- The numbers do not back the story. In 1965, American police cleared roughly 90 percent of homicides - with zero cameras, zero databases, zero DNA. Today, with 100,000 Flock cameras and every surveillance tool ever built, barely half of murders get solved, a record low. Criminologists cite many causes for the decline. But one thing the sixty-year record makes plain: whatever solves murders, mass surveillance is demonstrably not it.
- What actually solves murders is witness cooperation. Research on high-violence neighborhoods finds witnesses are aware of 70 to 80 percent of shootings but cooperate with police in fewer than 20 percent of cases. Murders get solved by trust, witnesses, and detectives. Blanketing a community in surveillance builds the opposite of the trust that gets witnesses talking.
- And in Oklahoma City, the anecdote is unverifiable by the department's own written admission. On July 10, 2026, OKCPD's response to OCPD-10088-2026 confirmed that no records exist linking Flock to case outcomes. Even Flock's own statement to KOCO 5 only claims the cameras were "contributing to" the Midtown arrests. An anecdote from an agency that keeps no records on the question is a story, not evidence.
"But the police say they can't do their job without it."
- That is a dependency confession, not an endorsement. This technology has existed at scale for roughly four years. Policing existed for two centuries before it. A chief who cannot imagine doing the job without a private vendor's subscription is describing how fast operational competence atrophies once a department outsources it.
- Dependency is exactly what the vendor's pricing model needs. Section 6.2 of Oklahoma City's own Flock contract lets the vendor change fees at every renewal with 60 days' notice. "I wouldn't know what to do without it" is the sentence that lets a vendor charge anything it wants, forever.
- By the chief's own math, the cameras replace officers. He puts his technology budget at $600,000 a year - "the equivalent of six officer salaries" - and says the cameras do "the job of probably 15-20 officers." That is the actual pitch: fewer sworn, trained, accountable human officers; more subscription hardware owned by a company in Atlanta. Flock's CEO accuses critics of wanting to defund the police. His own customers describe the product as doing precisely that.
- The largest departments prove the alternative exists. In July 2026, the Los Angeles Police Department - one of Flock's biggest customers - declined to renew its contract over data-ownership concerns. Departments can police without Flock. One of the largest in the country just chose to.
This is why the answer is no.
Flock's marketing depends on the "just another camera" framing staying uninterrogated. Once you look at what Flock actually is - a searchable, cross-agency, retroactive database of every vehicle at every intersection where a Flock camera is bolted to a pole - the comparisons stop working. And once the comparisons stop working, there is only one honest position left to take on the contract: end it.