Ithaca-Area Flock Camera Rollback: Cayuga Heights Joins The License-Plate-Reader Backout

BadPD source-check, June 20, 2026: the current Ithaca-area accountability lead is not just one crime call. The sharper public-records story is that Cayuga Heights has now joined the local rollback of Flock Safety license-plate-reader cameras, after the City of Ithaca and Tompkins County already moved away from the vendor earlier this year.

What Happened In Ithaca-Area Government

607 News Now reported that the Cayuga Heights Board of Trustees voted unanimously on June 17, 2026 not to renew the village’s Flock Safety contract. The outlet reported that four Flock cameras had been operating in the village since May 2025 and that the contract runs out in September. The same report placed the decision in a local pattern: Trumansburg, Ithaca, and Tompkins County have also moved away from Flock after privacy concerns.

The official Cayuga Heights board page lists the June 17 meeting and links to the agenda packet. Inside that packet, Exhibit 2027-033 is titled as a resolution to decline renewal of the Flock Safety camera system contract. That is the primary receipt. It says the village maintains a contract for a license plate reader camera system scheduled to expire on or about September 27, 2026. It also says the village operates four Flock cameras that have been in service since May 2025.

The official resolution language is unusually useful because it does more than say yes or no. It identifies the public concerns in policy terms. The board packet references residents’ concerns about searchable databases documenting vehicle movements, and it names the kind of sensitive inferences that can come from location patterns: medical visits, religious visits, and private associations. It also raises the public-records problem: even if Flock says the data belongs to the village, the resolution says access is controlled by Flock and quotes contract language giving Flock control over the method, timing, format, and medium of access or delivery.

That is the BadPD hinge. A public agency cannot responsibly tell residents that data belongs to the public body if the public body cannot practically retrieve, audit, disclose, or restrict that data on public terms. Ownership language is not enough. Public control needs operational control.

This Was Not A One-Town Fluke

WSKG reported on March 4, 2026 that the Ithaca Common Council passed a resolution ending the city’s relationship with Flock Safety. That report described a public meeting, a protest outside city hall, and a debate between privacy concerns and law-enforcement utility. It also quoted Ithaca’s police chief defending the cameras as a tool while reporting that the mayor voted for cancellation because of civil-liberties concerns.

WBNG reported on April 16, 2026 that the Tompkins County Legislature voted to terminate the county’s Flock Safety contract. WBNG reported that the vote was 12-1, with one legislator opposed, and that community concerns and lack of confidence in the vendor helped drive the decision. WBNG also reported that the Tompkins County Sheriff’s Office, other county partners, and the Ithaca police chief had submitted a next grant-cycle application that did not involve Flock.

607 News Now’s April 20 county follow-up adds useful context. It reported that Flock license plate readers and gunshot detection devices had been operating in the county as part of the Gun Involved Violence Elimination program, commonly known as GIVE. It also reported the public-safety argument in favor of the system: proponents pointed to arrests and help locating a vulnerable missing adult. BadPD keeps that context attached because accountability journalism should not pretend a tool has zero legitimate use when the record says otherwise.

But legitimate use does not end the question. A tool that sometimes helps police can still be a bad public bargain if data access, outside sharing, search auditing, data retention, FOIL compliance, contract renewal, and abuse penalties are weak. The public does not have to choose between public safety and public control. It can demand both.

What The Cayuga Heights Resolution Gets Right

The official Cayuga Heights resolution is worth reading closely because it states the core surveillance problem in ordinary-government language. First, it recognizes that ALPR systems are not just cameras. They are databases. A camera sees one moment. A database connects many moments. A searchable database can reconstruct habits, relationships, routines, vulnerabilities, and private associations.

Second, the resolution points to FOIL and transparency. In New York, public agencies have public-records obligations. If a private vendor controls how and when the data can be accessed or delivered, the public body may struggle to answer records requests, preserve records, or prove that it searched fully. FOIL does not become optional because a vendor stores the data.

Third, the resolution identifies third-party access concerns. That is a key ALPR issue nationwide. Residents may be told a camera is local, but the database may connect to broader networks, hot lists, outside agency searches, vendor workflows, or emergency exceptions. The question is not only who placed the camera on a pole. The question is who can query the results.

Fourth, the resolution considers operational value. It says the local network is expected to decline as neighboring jurisdictions end or reduce Flock use. That matters because even if a board accepts some privacy tradeoff for a tool, the cost-benefit calculation changes when the network shrinks. A village should not renew a controversial surveillance contract out of habit if the original regional value has changed.

The Public-Safety Argument Still Deserves Receipts

BadPD is not publishing this as an anti-police technology ban. Police agencies can and do use ALPR systems to locate stolen cars, suspects in serious cases, and missing vulnerable people. If a department wants to defend a system, it should publish the receipts in a way that the public can evaluate without revealing active investigative details.

That means annual counts. How many alerts? How many confirmed hits? How many false positives? How many stolen vehicles recovered? How many missing-person or endangered-person recoveries? How many arrests followed from the system? How many cases were later dismissed? How many searches were unrelated to hot-list alerts? How many user accounts had access? How many outside agencies searched the local data? How many searches were audited? How many searches were rejected, flagged, or disciplined?

Without those numbers, the public is stuck arguing from vibes. Supporters say the tool helps. Critics say the tool chills civil liberties. Both can be partly right. A democratic renewal process should force the agency and vendor to show the actual record, the actual contract, the actual audit controls, the actual retention schedule, the actual sharing list, and the actual incidents of misuse or false alerts.

Why Ithaca Is A Useful Test Case

The Ithaca area is now a useful test case because several nearby public bodies moved in the same direction within months. The city ended its contract. The county terminated its contract. Cayuga Heights is declining renewal. That creates an accountability laboratory: what happens to crime clearance, stolen-car recovery, missing-person response, grant funding, and community trust when local governments step back from a regional ALPR network?

If crime spikes because a genuinely useful tool was removed without a replacement, residents deserve to know. If crime does not spike, residents also deserve to know because that would undercut the claim that ALPR was indispensable. If agencies replace Flock with a narrower tool that has stronger controls, that is important. If they quietly route around the public vote through neighboring access, private cameras, retail networks, or state/federal databases, that is the next story.

The right answer is not to assume. The right answer is to measure. BadPD wants the before-and-after numbers: stolen-vehicle recovery, shots-fired response, violent-crime clearance, investigative search counts, mutual-aid requests, grants, vendor costs, outside-agency access, data-retention changes, and complaint records.

The Campus And Private-Camera Lane

The Cornell Daily Sun reported in March 2026 on student concerns about Flock camera usage on campus. BadPD is treating that as a concern receipt, not as proof that any municipal action occurred on campus. Still, it is important because Flock is not limited to traditional police departments. The broader ecosystem can include universities, private customers, homeowners associations, businesses, and other institutions.

That creates a public accountability gap. A city can end its contract, but residents may still pass cameras controlled by private entities or other institutions. A county can terminate a contract, but data may still flow through nearby jurisdictions, campus systems, commercial lots, or regional sharing agreements. The public needs a map that distinguishes municipal cameras, county cameras, campus cameras, retail/private cameras, and state or federal access.

If a public body leaves Flock but law enforcement keeps access to private or campus Flock feeds, the public deserves that disclosed in plain language. If law enforcement does not retain such access, the public deserves that assurance in writing. Either way, no one should have to guess.

Routine Crime Calls Are Not The Same Story

The Ithaca search also surfaced normal police items: a June 20 stolen-vehicle public-assistance notice from 607 News Now and a June 11 WBNG stabbing report on Elmira Road. Those are real local public-safety items, but they do not by themselves prove a police-accountability failure. BadPD is not turning every crime blotter item into a misconduct post.

The relevant connection is narrower. Departments often defend ALPR systems by pointing to stolen vehicles, violent crime, and urgent missing-person cases. Critics often respond that public agencies can investigate those crimes without building broad movement databases. The Ithaca-area Flock rollback should therefore be tracked against exactly those categories. Did stolen-vehicle recovery worsen after cameras leave? Did violent-crime investigations lose a critical tool? Or did agencies adapt with narrower, more accountable methods?

That is where the follow-up reporting should go. Not slogans. Not panic. Numbers, contracts, audit logs, public records, meeting minutes, and real case outcomes.

What Residents Should Ask For Next

Residents and reporters should ask Cayuga Heights, Ithaca, Tompkins County, and any participating law-enforcement agency for the same core records. Ask for the Flock contract, renewal notices, cancellation letters, data-retention settings, sharing settings, hot-list sources, agency-user list, outside-agency access list, audit logs, annual search counts, false-positive reports, grants that paid for the cameras, and any communications with Flock about FOIL requests or data export.

Ask whether any data will be deleted when contracts end. Ask who confirms deletion. Ask whether the vendor provides a deletion certificate or export. Ask whether local agencies keep historical copies. Ask whether any outside agency keeps copies. Ask whether any private entity in the jurisdiction still feeds Flock data to police. Ask whether future public-safety grants will fund replacement technology, and if so, what guardrails will be different.

Ask for the cost ledger too. Hardware, subscription, installation, maintenance, staff time, legal review, grant management, public meetings, and any exit costs should be part of the public debate. A surveillance tool should not be judged only on the sticker price or on one success story.

How To Judge Any Replacement Tool

Cayuga Heights’ resolution directs staff to explore alternatives that better align with village needs and the concerns identified in the packet. That is reasonable, but it cannot become a loophole. A public body should not leave one surveillance vendor because residents are angry, then buy a different system with the same data problems under a softer name.

Any replacement should be judged before purchase, not after installation. First, it should have a narrow use case. If the stated need is stolen-vehicle recovery, the system should be configured around stolen-vehicle recovery, not general-purpose historical tracking. If the stated need is violent-crime follow-up, the search policy should say who can run a search, what case number must exist, what supervisor approval is required, and how the search is later audited.

Second, the retention period should be short by default. Public agencies often defend retention by saying old data might become useful later. That is true of almost any data, which is exactly why retention needs a rule instead of an instinct. If a record is not tied to a specific investigation, hit, warrant, missing-person case, or documented public-safety need, it should not sit around as a movement archive waiting for a future reason.

Third, outside access should be opt-in, written, and visible. If a sheriff, federal agency, neighboring department, task force, campus police department, private security team, vendor employee, or state fusion center can access local reads, the public should be able to see that access list. Emergency access should be logged and reviewed after the fact. If the vendor can disclose data under broad emergency, fraud, security, or legal-process language, the public body should say how often that has happened.

Fourth, audits should have consequences. Audit logs that nobody reads are decoration. The public should know how often supervisors review searches, whether random audits are performed, whether improper searches trigger discipline, and whether any employee has ever lost access. If a system can track everyone else, the public has the right to ask who is tracking the users.

The Records Ledger BadPD Will Keep Watching

This package should stay on the watch list because the most important receipts come after the vote. One set of records should show the exit: cancellation letters, non-renewal notices, vendor communications, removal schedule, hardware pickup, data export, data deletion, and any final invoice. Another set should show continuity: whether law enforcement keeps access to old reads, whether exported data remains in local systems, whether open investigations retain snapshots, and whether outside agencies had access before the shutdown.

A third set should show replacement planning. If Cayuga Heights, Ithaca, or Tompkins County pursues a different platform, residents should see the request for proposals, bids, vendor scoring, legal review, privacy review, cybersecurity review, fiscal note, insurance requirements, data ownership language, breach-notification language, and public hearing schedule. If officials say the next tool is different, the contract should prove it.

A fourth set should show public-safety results. Track stolen-vehicle recoveries, violent-crime case support, missing-person searches, false hits, stop outcomes, dismissals, and complaints for the period before and after the cameras leave. That is the only honest way to test the competing claims. If the cameras were essential, the data should show a measurable gap. If the cameras were over-sold, the data may show that local police can still do the job with narrower tools and better community trust.

BadPD will also keep this tied to the broader New York ALPR lane. The Westchester lawsuit raised one version of the question: how much vehicle-location data can a county collect and share before the public has a constitutional and statutory problem? The Ithaca-area rollback raises another version: when residents and local lawmakers see the risks early enough, can they unwind the system before it becomes permanent infrastructure?

BadPD Bottom Line

The Ithaca-area Flock rollback is publishable because it is source-backed, local, current, and accountability-relevant. Cayuga Heights’ own board packet says the village is declining renewal and explains why. Ithaca and Tompkins County had already moved away from Flock earlier in 2026. The public-safety arguments remain part of the story, but so do the civil-liberties and public-records failures that led multiple local governments to step back.

BadPD’s position is simple: if police technology scans public movement, the public gets to see the contract, the data rules, the sharing rules, the audit rules, the abuse penalties, the renewal vote, and the before-and-after results. Public safety cannot be a black box. Neither can civil liberties. Put the receipts on the table.

Source Trail

• 607 News Now: Cayuga Heights trustees vote not to renew Flock contract (June 18, 2026) – Current local report that Cayuga Heights trustees voted June 17 not to renew the Flock Safety contract, with four cameras operating since May 2025 and contract expiration in September.
• Village of Cayuga Heights Board of Trustees page (accessed June 20, 2026) – Official meeting page listing the June 17, 2026 meeting, agenda link, and board meeting records.
• Village of Cayuga Heights June 17, 2026 Board packet (June 17, 2026 agenda packet) – Official packet containing Exhibit 2027-033, a resolution to decline renewal of the Flock Safety camera system contract.
• WSKG: Ithaca Common Council votes to end contract with Flock Safety (March 4, 2026) – Regional public-media report on Ithaca Common Council ending the city Flock contract, including protest, public-safety arguments, civil-liberties concerns, and sanctuary-policy context.
• WBNG: Tompkins County terminates contract with Flock Safety (April 16, 2026) – Local TV report that the Tompkins County Legislature voted 12-1 to terminate the county Flock contract and that a next grant-cycle application did not involve Flock.
• 607 News Now: Tompkins County terminates Flock Safety contract (April 20, 2026) – Local radio/news report with additional context on GIVE funding, public concerns, law-enforcement arguments, and the county deadline.
• Cornell Daily Sun: student concern over Flock Safety AI camera usage on campus (March 2026) – Campus reporting used as a local concern receipt, not as proof of municipal action.
• Atlas of Surveillance: automated license plate reader data (accessed June 20, 2026) – Public surveillance database entry listing Cayuga Heights Police Department use of Flock automated license plate readers.