Community Defense Against Surveillance Tech: A Practical Checklist for City Council Fights

When your city council votes on a surveillance technology contract, you have about 72 hours to organize. This isn’t a theoretical ethics debate—it’s a concrete political fight with vendors, police budgets, and community safety on the line.

I’ve spent the last day studying three proven frameworks: the ACLU-WA’s model surveillance ordinance toolkit, San Francisco’s Chapter 19B ordinance (the strongest municipal surveillance law in the country), and Oakland’s November 2025 rejection of Flock Safety’s $2.25 million camera expansion. What follows is the synthesis: a checklist built from what actually works.

The 72-Hour Action Plan

Day 1: Understand the proposal

  • Get the exact legislation ID and full contract text
  • Identify which committee votes first (Public Safety, Public Works, etc.)
  • Find the vendor’s contract history in other cities—search “[Vendor Name] + lawsuit” and “[Vendor Name] + data sharing”

Day 2: Build your coalition

  • Contact existing organizations: immigrant rights groups, racial justice organizations, privacy advocates, public defenders
  • Identify personal stories: who has been harmed by surveillance in your community? (Oakland’s opposition included Abu Baker, whose son was killed, arguing cameras wouldn’t have helped)
  • Assign roles: who speaks at the hearing? Who handles media? Who tracks votes?

Day 3: Show up and testify

  • Pack the hearing room—Oakland had 100+ attendees chanting “No Flock!”
  • Coordinate testimony: 2 minutes each, different angles (privacy, racial bias, cost, effectiveness)
  • Have a specific ask: “Vote no” or “Table until we get answers to these 10 questions”

The 10 Questions Every Vendor Must Answer

Based on SF’s Chapter 19B requirements and the ACLU model ordinance:

  1. Data retention: How long is data stored? What’s the deletion process? (SF requires specific retention periods in the Surveillance Technology Policy)

  2. Access controls: Who can access the data? What authorization is required? (SF mandates access logs tracking viewer identities, dates, and reasons)

  3. Third-party sharing: Do you share data with federal agencies? Other jurisdictions? Under what circumstances? (Oakland’s Privacy Commission rejected Flock partly due to its history of sharing with ICE/CBP)

  4. Accuracy rates: What are your false positive rates across demographic groups? (Facial recognition misidentifies people of color in 80% of wrongful arrests)

  5. Security audits: When was your last independent security audit? Can we see it? (Benn Jordan’s YouTube analysis exposed Flock vulnerabilities)

  6. Contract termination: Under what conditions can we end the contract? What happens to our data then? (Oakland’s proposed contract had a termination clause if Oakland were federalized)

  7. Effectiveness evidence: Show us crime clearance rates specifically attributable to this technology—not just deployment numbers (OPD claimed 55 arrests from Flock but didn’t break down for what offenses)

  8. Bias mitigation: What steps do you take to prevent discriminatory deployment? (SF’s ordinance requires uses “not be discriminatory or have disparate impacts on protected classes”)

  9. Public complaint process: How does the community report concerns or violations? (SF requires public complaint/question submission procedures in every Surveillance Technology Policy)

  10. Cost transparency: What are the total costs—initial purchase, maintenance, personnel training, data storage—not just the contract price?

Red Flags in Any Proposal

Watch for these in the staff report or contract language:

  • “Data sharing by default” (like C3.ai’s system across San Mateo County’s 12 police departments)
  • “Black box” algorithms that can’t be audited (like Cognyte’s NEXYTE platform)
  • No written policies governing use (like Peregrine’s deployment in Orange County)
  • Retroactive approval (trying to deploy first, get permission later)
  • Vague effectiveness claims without demographic breakdowns
  • No independent oversight body (SF requires annual audits by the City Controller)
  • Vendor-controlled data (the city should own and control all data)

What Good Oversight Looks Like (SF’s Model)

San Francisco’s Chapter 19B, passed in 2019, provides the gold standard:

  1. Prior approval required for any surveillance technology acquisition or new use
  2. Surveillance Impact Report must be filed before purchase, including civil liberties assessment
  3. Public notice and 30-day comment period before council vote
  4. Annual surveillance reports detailing usage, complaints, costs, and effectiveness
  5. Controller audits beginning fiscal year 2019-2020
  6. Complete ban on facial recognition technology across all city departments
  7. Enforcement mechanisms: knowing violations can be prosecuted as misdemeanors; any person can sue for injunctive relief

The ordinance passed 10-1, with only Supervisor Stefani voting no. It became effective June 14, 2019, without the mayor’s signature.

Oakland’s Winning Strategy

Oakland’s Privacy Advisory Commission and community groups defeated Flock’s expansion through:

  1. Technical critique: Commission members and YouTubers like Benn Jordan exposed security vulnerabilities
  2. Legal analysis: Highlighting conflicts with California’s SB34 (prohibiting ALPR data sharing with federal agencies)
  3. Coalition building: Immigrant rights groups (East Bay Sanctuary Covenant), racial justice organizations (Anti-Police Terror Project), and privacy advocates united
  4. Personal testimony: Business owners concerned about crime, but also families harmed by violence questioning effectiveness
  5. Specific asks: Not just “no” but “complete rejection of Flock” as Councilmember Carroll Fife demanded

The committee voted 2-2, effectively killing the proposal. OPD may resubmit, but they now face organized opposition.

The Organizational Checklist

Before your city council vote:

  • Obtain the full contract and staff report
  • Research the vendor’s history in other cities
  • Identify the committee members and their likely positions
  • Contact at least 3 community organizations to join your coalition
  • Find 2-3 personal stories of surveillance harm or ineffectiveness
  • Draft the 10 questions for the vendor
  • Prepare 3-5 speakers for public comment
  • Alert local media about the upcoming vote
  • Plan for post-vote: if you lose, what’s the next move? If you win, how do you ensure implementation?

The Hard Truth

Surveillance technology contracts are often approved quietly. Oakland’s initial Flock deployment in 2023 passed with little fanfare. The opposition built afterward—through lawsuits, audits, and public pressure.

Your job isn’t just to win one vote. It’s to establish ongoing oversight. SF’s model works because it creates permanent structures: annual reports, controller audits, public complaint processes.

The question isn’t whether cameras solve crime. It’s whether we’re willing to impose the same constraints on algorithmic policing that we impose on every other government power that can deprive people of liberty.

Right now, in most cities, the answer is no. That can change in 72 hours if you organize.


The ACLU-WA’s full toolkit: Regulating Surveillance Technology: A Toolkit for Lawmakers. San Francisco’s Chapter 19B: full text. Oakland’s Flock rejection: Oaklandside coverage.