Ending a surveillance contract should be as simple as signing one. The reported trouble one Minnesota city had in removing Flock Safety cameras shows that cities need clear exit terms before installation, not after.
Key takeaways
- The Verge reports that Plymouth, Minnesota, voted in September to end its contract with Flock Safety, but the company did not answer the city’s request.
- According to that report, the cameras in Plymouth stayed in place for more than a month after the council decided to end the contract.
- Licence plate cameras usually belong to the vendor rather than the city, so removing them depends on the vendor’s cooperation.
- Contracts that set firm deadlines for removing hardware and deleting data would make a council’s decision to withdraw take effect promptly.
Leaving should be as easy as joining
The argument of this article is simple. When an elected council decides to stop using a surveillance system, that decision should take effect quickly and predictably. It should not depend on a private supplier choosing to reply. Plymouth’s experience, as The Verge describes it, suggests this is not guaranteed. Cities considering automated licence plate recognition should therefore negotiate exit terms as carefully as they negotiate price and coverage.
Flock Safety sells automated licence plate recognition systems to local governments, police departments and private customers. Its cameras sit beside roads, take images of passing vehicles and turn those images into searchable records of plates and vehicle features. Police use the records to look for stolen cars, to follow up on alerts and in investigations. Supporters see the cameras as a cheap way to extend police capacity. Critics worry about the scale of data collection, how long records are kept and who can search them.
The Verge reports that more cities are cancelling their Flock contracts. It also reports that some of them are finding cancellation hard. This article does not try to settle whether licence plate cameras are good or bad policy. Its claim is narrower. Whatever a community decides, the decision should actually happen, and a contract is where that can be secured.
The position is easy to disagree with. Some will say existing contracts already handle this well enough, that one case proves little, or that the delay had ordinary causes. Those objections are set out below. The evidence available so far still favours putting stronger exit protections in writing.
Plymouth’s council voted, and the cameras stayed
The clearest evidence is the case at the centre of The Verge’s report. According to the publication, Plymouth’s city council voted in September to end its contract with Flock. The city then asked the company to act on that decision, and the company did not respond. The cameras stayed up for more than a month.
Several details are not public in the material available here. It is not known how many cameras were installed in Plymouth or where. It is not known what the contract said about termination notice, removal of equipment or what happens to data. It is also not known whether the cameras kept recording during that month, whether anyone could still search the data, or whether Flock has explained the delay. The Verge’s summary is cut short just after it says the cameras stayed up, so it is not clear from this material how the matter was eventually resolved.
Even with those gaps, the basic sequence matters. An elected body made a formal decision. It told the supplier. The infrastructure stayed where it was. In public procurement, a council’s vote is meant to settle a question. If carrying out that vote depends on a contractor’s reply, the city has given up some control over its own policy. That control is worth protecting whether or not residents supported the cameras in the first place.
A month may sound short. For a device that records every vehicle passing a fixed point, it is not trivial. If the cameras kept working, a month of records was produced after the community’s representatives had decided they no longer wanted that collection. If the cameras were switched off but left standing, residents would have had no way of telling, because a dormant camera looks the same as a working one.
Vendor-owned equipment puts control with the vendor
The second piece of evidence is about how these systems are commonly supplied. Licence plate camera programmes for local governments are often sold as a service, not as equipment the city buys outright. Under that kind of arrangement, the vendor usually owns, installs and maintains the cameras. The city pays a recurring fee for access to the system and its data.
This model has real benefits. Cities avoid large up-front costs, need no in-house technical staff and get maintenance and software updates included. It also has a consequence that matters when a city wants out. If the vendor owns the cameras, city workers may have no clear right to take them down. Removing equipment they do not own could create legal or contractual risk. The practical route is to ask the vendor to come and collect it, so the timing of removal depends on the vendor’s schedule and willingness.
Data raises the same problem. In a hosted system, the records sit on the vendor’s servers. Deleting them, returning them or confirming they have been destroyed is something the vendor does, not the city. A council can vote to end a programme, but proving that the data is gone means relying on the supplier’s statement unless the contract provides for independent checks.
This is why the Plymouth report matters beyond one city. If removal and deletion are actions only the supplier can take, the city’s ability to leave is only as good as the contract’s terms and the supplier’s responsiveness. Where the terms are vague, a supplier that is slow, distracted or reluctant to lose a customer can stretch out the exit without breaking any rule.
A wave of cancellations makes clear terms more urgent
The third piece of evidence is the trend The Verge describes: more cities are cancelling their Flock contracts. The publication’s summary does not give a number, list the cities or explain the reasons for each decision. Those details are not known here. Still, the direction is significant for the argument.
When few customers leave, exit procedures get little attention. Nobody tests them, and any weaknesses stay hidden. When cancellations become more common, those procedures carry real weight. Each city leaving runs into whatever its contract does or does not say. The Verge reports that as many cities are finding cancellation difficult as are going ahead with it. If that is accurate, problems with exits are a recurring issue, not an isolated one.
Cancellations also tend to follow public debate. A council may vote to withdraw after residents raise privacy concerns, after questions about who can see the data, or after a review of whether the system has been worth the cost. In those situations, how quickly the cameras come down affects public trust in local government as well as the vendor. A city that announces a decision and then cannot carry it out may find residents doubting its other commitments.
Taken together, these three points support a practical conclusion. Contracts for licence plate recognition should include a fixed period for removing hardware after termination, written confirmation that data has been deleted, a right for the city to remove equipment itself if the vendor misses the deadline, and penalties or withheld payments as an incentive. None of this needs new legislation. It needs procurement officers to treat leaving as a normal part of the contract.
The case against: one report does not show a pattern
The strongest objection is that the evidence is thin. The article rests mainly on one reported example, and the rest of that account is not available here. It is not known whether Plymouth’s contract had a notice period that had not yet run out when the cameras stayed up. Many service contracts require notice of 30, 60 or 90 days, or tie termination to the end of a billing cycle. If Plymouth’s contract had a term like that, a month with cameras still in place might have been fully within the agreement. In that case the lesson is about councils reading their contracts, not about vendors ignoring them.
A second objection is about logistics. Taking down roadside equipment takes crews, scheduling and sometimes coordination with utilities or highway authorities. A supplier dealing with many cancellations at once might fall behind without bad intent. Not replying to a request is poor practice, but it could reflect administrative overload rather than a deliberate attempt to keep a customer.
A third objection is that stricter exit terms come at a price. Suppliers who face penalties for slow removal or who must allow city crews to handle their equipment may charge more, ask for longer minimum terms or decline to bid. Smaller cities with little negotiating power may simply be offered the vendor’s standard contract. Seen this way, the market should sort the problem out: a supplier that makes leaving painful will lose reputation and future customers.
Finally, The Verge’s wording is cautious. It says cancellation is difficult “in some instances”, not in all. Many cities may have ended their contracts with no trouble at all, and those cases rarely make the news. Coverage naturally favours the difficult cases. These objections are serious. They do not show that firm exit terms are unnecessary, but they do mean the scale of the problem has not been established.
Evidence that would change this conclusion
Several kinds of information could weaken or strengthen the argument. The most useful would be the termination clauses in Plymouth’s contract and in contracts held by other cancelling cities. If those clauses already set short, enforceable deadlines and the delays happened inside them, the case for new terms would be weaker. If the clauses say nothing on removal and deletion, the case would be stronger.
A second source would be a consistent record of how long removal takes across many cancellations. If most cities see cameras taken down within days of their notice period ending, Plymouth looks like an outlier. If long delays are common, the problem is structural.
A third source would be Flock’s own explanation for what happened in Plymouth, and any general policy it has on handling cancellations. A clear, published policy with firm timescales would answer much of the concern. Lastly, confirmation of whether the cameras recorded data after the council’s vote, and whether that data was deleted, would show whether the delay had any real effect on residents’ privacy.
Sources and further reading
- The Verge, technology news coverage reporting on cities cancelling Flock Safety contracts and the Plymouth case
- City council records and meeting minutes, the primary documents for local votes on surveillance contracts
- Civil liberties organisations’ published guidance on automated licence plate recognition and local procurement
- Public procurement guidance from municipal associations on service contracts, termination clauses and data handling
Surfaced from the rss:verge signal “surveillance contract cancellations”. AI-assisted draft, editorially reviewed.
