News · 2026-10-04

Sanders, Ocasio-Cortez, and Merkley propose federal restrictions on license-plate surveillance

Bernie Sanders, Alexandria Ocasio-Cortez, and Jeff Merkley announced the Ban Flock Act on October 2, proposing restrictions on federal access to automated license-plate surveillance and federal funding that supports it. The draft targets an entire technology category, including vehicle characteristics and derived records, rather than only Flock Safety’s cameras; it remains a proposal, not an enacted ban.

Key facts

• The three lawmakers announced the proposal on October 2, 2026.
• The draft would restrict federal use and condition assistance administered by five federal departments.
• Future congressionally authorized uses would face safeguards, including a retention ceiling of 48 hours.
• Primary source: Sanders’s official announcement and its attached draft.

The name is memorable, but the mechanism is the news. A law can regulate a technology by directly forbidding conduct, by controlling who may buy it, or by attaching conditions to money. This proposal uses several of those levers. Understanding them prevents the headline from turning into a claim that every local camera would be switched off the moment the proposal appeared.

The released Senate text, titled “Ban Flock Act,” would prevent federal agencies, officials, and people acting for the federal government from acquiring, possessing, accessing, or using covered systems and captured plate data. That includes records another entity collected. The point is to regulate access to an existing network as well as ownership of a camera.

Congress could authorize a particular use later, through legislation specifying who may use the data, for what purposes, with what standards and safeguards. The draft includes audit, accuracy, and enforcement requirements for that route, with a retention ceiling of 48 hours. A separate exception covers toll systems used solely for defined tolling functions, with purpose and deletion restrictions.

The funding provisions extend the proposal’s practical reach. Starting with the first fiscal year after enactment, certain assistance administered by Justice, Transportation, Health and Human Services, Housing and Urban Development, or Homeland Security would depend on a state or local government adopting restrictions at least as strong as the federal standard. Another section would bar federal money from covered system and data activities, subject to exceptions.

Think of this as a set of gates around a transport network. One gate controls federal entry. Another controls the money that helps other operators maintain or use the network. The gates could alter local incentives without making every independently funded, nonfederal use directly illegal. That distinction matters both for privacy advocates assessing the proposal’s reach and for agencies assessing what it would require.

The connection to AI is specific. Flock’s own frequently asked questions describe machine learning that categorizes vehicles and makes characteristics such as body type, make, color, racks, and aftermarket wheels searchable. That changes the search key: an investigator may narrow records by visible attributes as well as a known plate. The verified product description uses selectable filters. It does not establish a conversational search interface or every colorful example circulating online.

The draft’s definition includes persistent visual vehicle identifiers, photographs, locations, timestamps, confidence scores, and information derived from plate records. It therefore reaches more than a text string identifying a plate. The operative language does not single out AI as a separate technique. Its focus is the collection, linkage, access, and use of surveillance information.

The proposal arrived alongside reporting about an Oklahoma federal judge suppressing evidence obtained after a historical vehicle-location search in United States v. Kyle. 404 Media’s report quotes the decision and Flock’s response. The dossier could not directly read the signed opinion, so that reported ruling is context here, not the primary verification basis for this article. It should not be elevated into a nationwide warrant rule.

Flock’s strongest counter-position is that plate cameras create point-in-time observations useful for public safety, and that the Kyle decision is limited to its facts. The company told 404 Media it considered that decision contrary to other authority and expected reversal. Its public legal argument emphasizes favorable court decisions. Those are relevant company positions, not a resolved answer to the privacy impact of querying many observations together.

The analytical issue is aggregation. A single sighting on a public road and a searchable historical trail create different practical powers. Machine learning can increase the ease of finding relevant records, while network access can increase their geographic reach. Neither capability alone decides the legal standard, but both help explain why a data-access bill can matter more than a debate about individual cameras.

The proposal extends the policy debate already visible in LAPD’s decision to let its Flock contract lapse. The next questions are legislative support, the final text, and how funding conditions would work in practice. The published draft establishes a concrete policy design; it does not establish passage, implementation, or an immediate change in anyone’s access.

Primary source, verified: read the paper →

Ground Truth. (2026, October 4). Sanders, Ocasio-Cortez, and Merkley propose federal restrictions on license-plate surveillance. Ground Truth. https://groundtruth.day/news/ban-flock-act-targets-federal-surveillance-access-and-funding.html

@misc{groundtruth:ban-flock-act-targets-federal-surveillance-access-and-funding,
  title  = {Sanders, Ocasio-Cortez, and Merkley propose federal restrictions on license-plate surveillance},
  author = {{Ground Truth}},
  year   = {2026},
  month  = {oct},
  url    = {https://groundtruth.day/news/ban-flock-act-targets-federal-surveillance-access-and-funding.html}
}

Topics: policy · privacy · surveillance · computer-vision · civil-liberties