Metro Detroit · Wayne · Oakland · Macomb
Your everyday driving shouldn’t become a searchable government record.
Flock cameras and other automated license plate reader systems can create searchable records of where vehicles have been seen, when they were there, and how they move through our communities.
DeFlock Metro Detroit is an independent, nonpartisan, DeFlock-aligned local civic effort working to find these systems, uncover the contracts and public records behind them, and help residents organize for complete removal and nonrenewal.
Local · Lawful · Nonpartisan · Evidence first
Sterling Heights decision watch · October 20, 2026
Council is considering whether to continue a 24-camera Flock system.
The official 7:00 PM agenda lists an $84,100 annual agreement funded with state forfeiture money. The page below links directly to the city agenda and will be updated when the vote result is verified.
What an ALPR actually does
Driving past one can create a searchable record.
A Flock ALPR is not simply recording ordinary traffic video. When a vehicle passes, the system can create a searchable observation tied to a plate or vehicle description, together with the time and camera location.
One read is a point. Many reads across a network can reveal patterns of movement.
Privacy is not secrecy. We use doors, curtains and passwords because reasonable limits on access are normal. Being visible once in public is different from having many observations automatically indexed for later search.
Yes, ALPRs can help police locate wanted or stolen vehicles and develop investigative leads. We do not dispute that. The policy question is whether those benefits justify routinely creating searchable location records for the much larger number of people who are not suspected of a crime.
Usually records video of a place for later human review. It may be isolated and may not automatically index vehicles into a location-search system.
Extracts and indexes vehicle observations with time and location so authorized users can search them later and, depending on settings, across shared networks.
The useful investigative feature—rapid, scalable search—is also what can turn ordinary driving into a reconstructable movement history.
More than a plate reader
One camera isn’t the whole story.
“License plate reader” undersells what modern platforms can become. Flock publicly markets searches by visible vehicle characteristics, natural-language descriptions, cross-jurisdictional sharing and newer tools that help connect vehicle sightings and patterns. Separate video products can search people using visible, non-biometric characteristics. A competing vendor, Leonardo/ELSAG, now markets SignalTrace sensors that detect broadcast Wi-Fi, Bluetooth and supported RFID signals and correlate recurring device groups with time, place and—when available—plate observations.
That does not prove every feature is enabled locally. We separate vendor capability, local configuration, written policy and actual use. But technical capability still matters: software, integrations and permissions can change without replacing the roadside hardware, and the public may not know unless current settings, changes and audits are disclosed.
The plate may be what gets recorded first. The person behind it is what the information can ultimately reveal.
How much can these systems actually see? →
How SignalTrace follows wireless-device signatures →
Compare the claims with the public record →
Watch the video library →
Local cameras, wider reach
The poles are local. The network can extend far beyond them.
Metro Detroit communities have already purchased and operated automated license plate reader systems. But a city boundary does not necessarily mark the edge of the searchable information. Depending on sharing choices and permissions, observations can be available to participating agencies elsewhere in Michigan or across a nationwide network.
Flock says agencies control whether they share and can choose one-to-one, geographic, statewide or nationwide access. Its National Lookup feature allows a known full plate to be searched across cameras whose agencies opted into that network. That does not mean every local agency enabled every form of sharing. It means current settings, partner lists, agreements and audit records matter.
First you notice one camera. Then you find the local contract. Then you learn that outside agencies may be able to search across many cameras. The question changes from “What did this pole record?” to “Who can assemble the trip?”
If we don’t know something yet, we’ll say so.
See what’s happening near you →
Understand sharing, access and nationwide reach →
Documented failure and misuse
The risks are not only hypothetical.
Three different incidents show why both technical accuracy and authorized access matter. One occurred here in Metro Detroit. None proves every user acts improperly; each documents a failure mode officials should address before purchase or renewal.
Sherwood, Arkansas
Reporting documented a Flock plate misread that contributed to a high-risk stop involving an innocent couple and their infant before the mismatch was discovered. The lesson is narrow: an automated result must be independently verified before consequences escalate.
Source: Institute for Justice incident database ↗Savannah, Georgia
A police officer was fired after an internal investigation found she gave her Flock login to her husband, an ICE agent. Her account was used for 983 searches involving 97 identifiable plates across six states. The case shows that an audit trail can expose misuse after access has already occurred.
Source: WJCL reporting based on department records ↗Eastpointe, Michigan
The police chief said two employees misused Flock—one for a personal purpose and another in a policy violation. Council voted 4-1 to terminate the agreement, and police discontinued using the system.
Read the Eastpointe evidence →What this does not prove: that most reads are wrong or most officers misuse access. It demonstrates documented risks the system’s design and local safeguards must be evaluated against.
See the full documented evidence file →
Understand false matches and human verification →
Removal is possible
Cities are saying no—and cameras are coming down.
Westland let its Flock arrangement expire and began physical removal. Eastpointe terminated after two employee-misuse findings. Since the last site update, South Lyon and Milford rejected renewal, Madison Heights announced nonrenewal after deactivating its cameras, and Lincoln Park ordered its remaining cameras off while reviewing the contract.
These communities used different paths: nonrenewal, termination, immediate deactivation and hardware-removal demands. What they share is proof that an installed system does not have to become permanent.
A contract has an end date. A council has a choice. Residents can make that choice visible.
The obvious question
But don’t these cameras help solve crimes?
Sometimes they help investigations. We do not need to pretend otherwise. Real life is more complicated than a slogan.
The better question is whether that investigative benefit justifies routinely creating searchable location records about ordinary traffic, including people who were never suspected of wrongdoing.
Police still have witnesses, physical evidence, targeted surveillance, warrants, security footage, public tips, public records, lawful digital evidence, open-source research and detective work.
Ordinary traffic is recorded first. Suspicion can come later.
We do not have to choose between safety and privacy.
But there are rules against misuse, right?
Rules don’t make the database disappear.
Audit logs can help detect abuse. Retention rules can reduce how long information remains available. Access controls can reduce who can search it.
Those are real safeguards. They do not undo the underlying collection.
The strongest protection against misuse of a movement database is not creating it in the first place.
See documented misuse, failures & harm →
Answer the common objections with evidence →
Our position
Removal. Not better mass surveillance.
Shorter retention, stronger audits, tighter access and warrant requirements may reduce harm while a system exists.
They do not solve our underlying objection: information about ordinary people is still collected first.
DeFlock Metro Detroit works for complete removal and nonrenewal of covered mass vehicle-surveillance systems.
The investigative benefit does not justify routinely tracking ordinary people.
So what do we actually do?
We turn concern into something useful.
A surprising amount of this work is PDFs, meeting packets and asking, “Wait, when does this contract actually renew?” We find systems. Find contracts. Find deadlines. Find policies. Find votes. Then we translate the paperwork and help residents contact officials, attend meetings, request records, verify cameras and organize locally.
The goal is not endless research. The goal is to identify the decision point and organize before the next renewal.
Follow the contract and renewal window →
Understand who can search the data →
Start small. Help where you can. You can always do more later.
A note from D
This started with a pretty simple question.
When did ordinary driving become something government should routinely record and search later?
The cameras multiply. Cities join networks. Contracts renew quietly. Infrastructure starts to feel normal.
That did not sit right with me.
I started DeFlock Metro Detroit to make it easier to see what is happening, understand what the records actually show, and make these systems harder to renew quietly.
I go by D publicly and organize DeFlock Metro Detroit. I use a public handle because I want this work judged by the evidence and what local governments actually do, not by a personal brand.
You do not need to make this your whole personality, become a privacy expert or start reading procurement law for fun. Start with what your own community has approved. Look at the evidence. Decide what you think. Then take one useful step when it matters.
- D
Organizer, DeFlock Metro Detroit
Ready to do something useful?
Start with your own community.
You do not have to call yourself an activist or take on a second job. Join for local alerts, find one record, verify one camera, attend one meeting or contact one official when a real decision is coming up. Small actions matter most when they connect to the decision that can actually stop the next term.