Beyond your license plate: How deep can a warrantless search reach into your private life?
Your child’s school. Your doctor’s office. Your bank. Your place of worship. The job that pays your bills.
A camera sees a plate. A network can turn it into a history of where you have been. Accurate or not, that history can become “a lead”—and put you under suspicion—or worse, lead to incarceration and the loss of basic rights and freedoms.
You don’t have to be suspected of a crime for your car to be recorded. Connecting those sightings can expose routines you never chose to disclose—where your children spend their days, where you seek medical care, where you worship, when you leave work.
As cameras multiply and networks connect, where is your consent? What requires probable cause? What due process exists for a search you never know happened?
Someone searching the history can begin to piece it together—and get your life dangerously wrong.
A clinic visit mistaken for a diagnosis. A stop near someone’s home treated as a relationship. A coincidence turned into a connection to a crime.
The record shows where your car was—not why you went, whom you met or what you did.
Yet a false story built from those fragments can put your freedom, livelihood or family at risk. Once suspicion enters a file, it can outlast the investigation—and damage a reputation for a lifetime.
Ten chapters follow the system, the logs and cross-state access, VCSO’s acknowledgments, requested and missing records, the public-interest test, Axon/Fusus integration questions, local cases and national comparisons, and what residents can demand next.
Perfect auditing would not make mass location surveillance harmless. But when government chooses to exercise this power, being able to prove the rules were followed is the minimum.
Investigation navigator
Contents
10 chapters
Illustrative roadside loop. Not incident footage or evidence of local camera placement.
Loop paused until in view
Illustrative camera montage. Not incident footage, actual tracking or evidence of local camera placement.
Loop paused until in view
Illustrative data montage. Not incident footage, actual tracking or evidence of local camera placement.
Inside Axon and Flock’s Expanding Surveillance Networks—and the Illegal Searches, Cross-State Access, and Oversight Failures Exposing Their Risks.
Illustrative loop: a capture becoming a searchable record. Not footage from any incident or camera.
Loop paused until in view
What Happens After the Camera Records Your Plate?
A roadside camera captures a moment. Connected surveillance can make it searchable long afterward.
Axon’s ecosystem extends into video evidence management, body-worn and in-vehicle cameras, and connected camera networks. Its Fusus platform brings previously separate sources into a common working environment. Flock’s ecosystem includes license-plate readers, searchable location histories and connected camera networks. Across these ecosystems, investigative search and analytics can turn individual records into information a user can assemble.
These are vendor capabilities, not a verified inventory of VCSO deployments or proof that the two vendors’ networks are connected locally. The Ventura records discussed here document Flock use, later Axon testing and settings changes, and ALPR activity through Fusus Web / ECOM. They do not establish that every advertised camera feed, analytics tool or integration was enabled.
The Camera Collects the Data. The Network Determines Its Reach.
A plate image, a time and a camera location can become a searchable vehicle sighting. Repeated sightings can reveal patterns of movement—not necessarily who drove, why they traveled or a complete account of their day.
Retention settings determine how long records remain available in a system. Search permissions and cross-agency sharing determine who can reach them. Downloads, evidence files, software-created information and backups may have separate retention and deletion rules. Deleting the original picture does not necessarily delete every copy or everything learned from it.
The questions are direct: Who can search? Across which cameras, agencies and periods? For what authorized reason? Who verifies that the rules were followed?
ANALYSISVisual 1
How ALPR Data Moves
A camera captures a plate. Follow where that information can go—and who can search it.
This is a conceptual illustration, not a verified feature inventory for a particular ALPR device or a map of VCSO’s deployed system. People or objects visible in an image are not necessarily identified, indexed or separately searchable. VCSO-specific records are needed to establish video capture, enabled tools and actual search results.
When Surveillance Access Is Misused, the Consequences Become Personal.
Indianapolis officers were charged following allegations of personal ALPR tracking. In Braselton, Georgia, a police chief was charged after an investigation of alleged ALPR harassment and stalking. Surveillance access meant for public safety can become a way to intrude on someone’s private life.
Pending charges are allegations, not convictions; the supplied sources do not establish adjudicated guilt in these cases. Audits can—and did—help expose potential misuse: IMPD says its internal audits triggered investigations. The Georgia case involved an external GBI investigation. Neither case establishes misconduct by VCSO, and the Georgia announcement does not identify the ALPR vendor.
The Ventura audit records examined in this report contain organization entries associated with Texas, Georgia, Illinois and Florida, among other states. Those names raise questions about the reach of California vehicle data.
Audit entries are not necessarily unique searches, matches or successful retrievals. A state association inferred from an organization name does not, by itself, identify the individual user, establish what was returned or prove an illegal search. How was access authorized, how was it governed, and how can those answers be independently verified?
The accountability question extends beyond the cameras and the agency that owns them.
The lines show possible ways to get access, not completed searches, returned information or verified local connections between vendor platforms. Dated agreements and settings records are needed to check federal, out-of-state and international access. The California-agency label represents an attributed assurance, not a verified inventory of all current or historical sharing.
Oversight Should Be Demonstrable, Not Just Promised.
Policies and internal reviews matter. Verification needs a record of who searched, when, what they searched, why they searched, and who reviewed the activity. Review criteria, findings and corrective actions show more than a promise that searches are checked.
Omitted reason fields constrain public verification: an outside reviewer cannot check a search’s purpose from a record that does not supply it. That omission does not prove misconduct or that no internal reason or review exists. Activity logs are not the same as audit findings, and records not supplied are not necessarily records that do not exist.
The chapters that follow trace the changing system and February–March 2025 anomaly, interstate organization entries, public acknowledgments, records requested and received, unresolved oversight questions, the public-interest test, Axon/Fusus integration capabilities, attributed local and national examples, and the actions residents can demand.
The investigation keeps three questions separate: what systems can do; what Ventura County records document about use, settings and deployments; and what the available records prove about access, authorization, results and oversight. A vendor description is not a local deployment record. A recorded action is not, by itself, proof of lawful purpose or every downstream consequence.
Illustrative loop: from a single camera to a network. Not evidence of any deployment or connection.
Loop paused until in view
Supporting system evidence, contract context and regional analysis
DOCUMENTEDHow the technology works
A sighting becomes a history.
A plate image, the time and a camera’s location can become a searchable record of where a vehicle was. Repeated sightings let someone look back through its movements.
UNRESOLVEDAccountability question
Deleting the picture is not necessarily deleting what was learned from it.
Where do the images go? What vehicle details does the software extract? What conclusions, shared files, evidence files, backups and copies remain elsewhere? The public needs to know how long the system keeps each kind of data—its retention rules.
ANALYSISHow the systems connect
Who Is Watching Us?
A camera count describes equipment. An account owner describes one relationship. Neither tells residents who can reach the resulting location history, through which systems, or what survives elsewhere.
01 · Cameras and sightings
A plate image, the time and the camera’s location are the starting point—not the whole system. Repeated sightings can create a history of where a vehicle has been.
02 · Agencies and networks
An agency name and a name covering a whole network mean different things. A network’s name alone does not tell us who took part or who could search whose data.
03 · Vendors, connected systems and users
The investigation records Flock use, Axon tests and settings changes, and later activity through Fusus Web / ECOM. Who could use each route, what they were allowed to do and why still need records.
04 · Copies elsewhere—and who can use them
Downloaded files, investigative files, information created by the software and backups need their own answers. For each place the data goes, who can access it, how long does it stay and how is it deleted?
Source: the investigation’s account of the system and where its data goes, plus the original map of its tools and connections. The missing receipts concern who could use what, settings, sharing, reasons for searches and what happens to the data—not just who owns the cameras.
What the contract offered and who paid · report page 4
The Civic Review Packet’s contract summary describes access to local and wider networks, searches by time and location, plate lookup, searching by what a vehicle looks like, and reports of system activity. It explains what the contract offered and how it was funded.
Report-context receipt · project-provided analysis, not the underlying contract or access audit.
What is documented—and what is not?
DOCUMENTED
VCSO used Flock license-plate readers. The files VCSO provided show a sudden change in reported organization entries during Feb–Mar 2025, with a clear beginning and end. Later Axon records show testing, a recorded change to the retention setting, Fusus/ECOM activity and comparison tests. A recorded setting is not independent verification of how every copy was retained or deleted.
UNRESOLVEDWhat we still don’t know
What fully explains the 2025 change? Who approved changes to the settings? Where did copies of the data go? Which software tools were enabled locally? Where is the complete record of activity across the connected systems?
ANALYSISThe wider regional network
Why NCRIC belongs in the accountability story
NCRIC, a regional information-sharing center for law enforcement, represents a network of organizations—not one local agency. It appears across 27 months in the original comparison. Knowing which local agency owns a camera is not enough to understand a network this large.
Who participated? Who was allowed to use what? How could one organization reach another’s data? The name printed in a file cannot answer those questions.
NCRIC shows why fewer entries can mean something different from less reach. Comparing January–July 2024 with August 2024–February 2026, the report finds that monthly entries associated with NCRIC fell from 7,421 to 1,181 at the midpoint of each period’s monthly values—the median. Yet the median number of reported networks rose from 214 to 755.
The report also describes an approximately 80% fall in median requested camera-time, measured in device-hours: devices multiplied by the hours in a requested window. The number of entries, networks and requested camera-time moved in different directions. Page 24 asks whether searches were set up differently, practices changed, files missed activity or the system reported it differently.
A license plate may be where the record begins. Understanding who can access that record—and what happens afterward—is where public accountability begins.
0202 · What Happened?
A 29-day anomaly. Then testing, settings changes and another way to search.
What happened—and when? Start with the February–March 2025 change, then follow the dated activity and records responses.
ANALYSISVisual 5
February–March 2025 Anomaly
Follow the dates: February 19–March 19, 2025 is the surge at the center of this review.
The measured series counts entries and organization names. The artwork’s reach and network lines are illustrative, not calculations from the daily/monthly datasets. March 2026 is partial.
One timeline: the surge, the changing system, and the records
These entries follow when events occurred, not when records arrived.
When a search ran and how far back it looked are different questions. Original audit files and correspondence are not included here; the sources below identify the summaries available.
DOCUMENTED
For 29 days, out-of-state names flood the data.
The Flock files VCSO provided show a large number of entries each day naming organizations associated with other states. Then those names disappear from this daily view.
Source: Flock audit files VCSO provided, summarized in the daily chart and accompanying dataset. · See the daily evidence
ANALYSIS
Out-of-state names disappear; the monthly total moves back toward normal.
The daily chart shows the surge ending after March 19. In the monthly chart, April is back to 270 distinct organization names as written in the file.
Axon's activity history records administrator testing of ALPR. On May 20, Eric Tennessen changed how long Axon kept license-plate readings from 30 to 365 days. Testing of a hotlist—a list of vehicles to watch for—and agency-wide alerts also appears in the records.
Source: Axon activity history, summarized in the investigation; original audit file not included here.
DOCUMENTED
Later Axon searches look back through the earlier period.
The investigation connects later Axon searches with the earlier period. When someone ran a search and how far back it looked are different questions.
Source: The investigation’s timeline; exact search dates and the periods they covered require the original Axon audit file.
DOCUMENTED
More ways to search appear in the audit.
Later Axon entries include ALPR searches via Fusus Web and ECOM, including searches labeled “Comparison Test with Flock.”
Source: Axon activity summarized in the investigation; original audit file not included.
DOCUMENTED
Residents ask for the accountability records.
Residents request accountability records. Chapter 5 follows the request and productions.
Source: The investigation’s timeline of the records request and VCSO’s responses; original correspondence not included. · What residents asked for
VCSO SAYS
VCSO publicly acknowledges the “data searching issue.”
At the Camarillo update, VCSO named the February–March 2025 episode a “data searching issue.” Chapter 4 examines the acknowledgment, stated safeguards and audit assurances.
The first production—and another file already expected.
Flock logs are produced; an Axon audit is separately promised. Chapter 5 examines the initial response.
Source: The investigation’s timeline of the records request and VCSO’s responses; original correspondence not included. · Read the production account
DOCUMENTED
Completeness challenged; more records followed.
The response is challenged; additional months follow.
Source: The investigation’s timeline of the records request and VCSO’s responses; original correspondence not included. · Read the follow-up
DOCUMENTED
The Axon audit arrives.
The later Axon file arrives. Chapter 5 separates its activity from the initial production.
Source: The investigation’s timeline of the records request and VCSO’s responses; original correspondence not included. · Read what arrived later
UNRESOLVED
A list of activity cannot tell us whether the rules were followed.
The sequence establishes activity and change. Chapter 6 identifies the proof still needed to independently verify the safeguards.
Source: What the investigation says these records can—and can’t—tell us. · The receipts still needed
What the chronology records
VCSO's license-plate surveillance system was changing. Staff were testing and comparing products. This sequence brings together the Flock surge, later Axon tests and settings changes, and activity through Fusus/ECOM.
ANALYSISVisual 5.2
Frequency and Requested Scope Are Different Questions
Fewer entries can sit alongside broader searches. How often a name appears, how much a search could reach and how far back it looked are different questions.
The report compares NCRIC across January–July 2024 and August 2024–February 2026. The artwork uses October 2024–March 2026; its ratios (~70% fewer searches, ~3× scope, ~4× networks), camera counts and international access remain unverified. What a search asked to cover is different from what it actually showed.
California People Plates and Paths, Warrantless Out-of-State Searches
THIS DIDN'T STAY IN CALIFORNIA.
DOCUMENTED
259,190
entries naming organizations associated with the states outside California examined here
2,021
organization names appearing in the data
Six states each appear more than 20,000 times: Texas, Georgia, Illinois, Florida, Ohio and Indiana.
Illustrative loop of data streams. It does not show actual transfer routes, searches or vehicle paths.
Loop paused until in view
The geography exploded. During February–March 2025, it wasn’t just the number of entries that changed. The organization names in the data pointed far beyond California.
DOCUMENTED
SIX STATES DOMINATE THE RECORD.
TX · Texas53,277
GA · Georgia47,312
IL · Illinois30,035
FL · Florida26,810
OH · Ohio20,275
IN · Indiana20,070
Together, six states account for 197,779 entries—approximately 76% of the 259,190 out-of-state entries examined here.
WHY?
3.1 TEXAS
53,277 ENTRIES.
Texas shows up 53,277 times in the out-of-state data we examined—more than one in five entries, and more than any other state in this review.
WHY TEXAS?
How were these agencies connected? Who was allowed to use what? What changed during February–March 2025 to put Texas here so often? What did VCSO learn when it reviewed what it later called a “data searching issue”?
Shares use all 259,190 examined entries, even where the graphic says “top six.” Map arcs show state associations, not vehicle routes or search direction. The timing and duration plots have not been independently verified against the underlying audit files.
Source: State dashboard; audit-data methods and source limits are in Supporting Materials.
3.2 GEORGIA
TEXAS WASN'T AN OUTLIER BY ITSELF.
47,312 ENTRIES.
Georgia appears 47,312 times—second only to Texas, with just 5,965 fewer entries.
WHY IS GEORGIA ALMOST AS LARGE AS TEXAS?
Texas was not the only distant state appearing this often. Which connections and access permissions explain Georgia appearing almost as often?
Shares use all 259,190 examined entries, even where the graphic says “top six.” Map arcs show state associations, not vehicle routes or search direction. The timing and duration plots have not been independently verified against the underlying audit files.
Source: State dashboard; audit-data methods and source limits are in Supporting Materials.
3.3 ILLINOIS
THEN COMES ILLINOIS.
30,035 ENTRIES.
Illinois adds 30,035 entries. Texas, Georgia and Illinois together account for 130,624—more than half of the 259,190 entries examined here. This is no longer a question about one distant state.
WHY WERE SO MANY DISTANT STATES SHOWING UP AT THIS SCALE?
Another distant state appears tens of thousands of times. What connected these systems? Why were these organization names in the data VCSO provided?
Shares use all 259,190 examined entries, even where the graphic says “top six.” Map arcs show state associations, not vehicle routes or search direction. The timing and duration plots have not been independently verified against the underlying audit files.
Source: State dashboard; audit-data methods and source limits are in Supporting Materials.
3.4 FLORIDA
FROM ONE COAST TO THE OTHER.
26,810 ENTRIES.
Florida adds 26,810 entries and another coast. The first four states now account for 157,434 entries—more than 60% of the total examined here.
WHY FLORIDA?
From California to Florida: which agency connections and access permissions explain this coast-to-coast pattern? What did VCSO’s review explain?
Shares use all 259,190 examined entries, even where the graphic says “top six.” Map arcs show state associations, not vehicle routes or search direction. The timing and duration plots have not been independently verified against the underlying audit files.
Source: State dashboard; audit-data methods and source limits are in Supporting Materials.
3.5 OHIO
AND THE PATTERN KEEPS GOING.
20,275 ENTRIES.
Even the fifth state, Ohio, exceeds 20,000 entries: 20,275. The pattern has a high floor, not just one or two large peaks.
WHY OHIO?
Then look next door. Indiana is almost identical in scale. What explains that pairing?
Shares use all 259,190 examined entries, even where the graphic says “top six.” Map arcs show state associations, not vehicle routes or search direction. The timing and duration plots have not been independently verified against the underlying audit files.
Source: State dashboard; audit-data methods and source limits are in Supporting Materials.
3.6 INDIANA
OHIO: 20,275. INDIANA: 20,070.
20,070 ENTRIES.
Indiana adds 20,070 entries—only 205 fewer than Ohio. Together, the six states reach 197,779 entries: approximately 76% of the total examined here.
WHY DO TWO DIFFERENT MIDWESTERN STATES APPEAR AT NEARLY IDENTICAL SCALE?
Six states, each above 20,000. Which connections, permissions or system changes explain that pattern? The audit should answer.
Shares use all 259,190 examined entries, even where the graphic says “top six.” Map arcs show state associations, not vehicle routes or search direction. The timing and duration plots have not been independently verified against the underlying audit files.
Source: State dashboard; audit-data methods and source limits are in Supporting Materials.
SIX STATES. 197,779 ENTRIES.
approximately 76% of the 259,190 out-of-state entries examined here
A MULTI-STATE PATTERN.
WHAT OPENED THE DOOR?
ANALYSISSix states — Ventura County connections197,779 entries across Texas, Georgia, Illinois, Florida, Ohio and Indiana—approximately 76% of 259,190 examined out-of-state entries.Open original image (new tab): Six states — Ventura County connectionsWhat this chart means—and how it was made
Each state’s share uses all 259,190 examined entries, not only the six-state subtotal. Map arcs show state associations, not vehicle journeys or search direction.
Source: Six-state summary; methods and source limits are in Supporting Materials.
3.7 THE VIRGINIA EXCEPTION
VIRGINIA MATTERS FOR A DIFFERENT REASON.
14,569 ENTRIES.
Virginia’s 14,569 entries put it outside the top six by frequency. But entry count is not the only dimension that matters. Fairfax County Police Department stands out in requested camera-time.
Shares use all 259,190 examined entries, even where the graphic says “top six.” Map arcs show state associations, not vehicle routes or search direction. The timing and duration plots have not been independently verified against the underlying audit files. This is Virginia’s entry-frequency view; Fairfax’s requested device-hours are explained below.
Source: State dashboard; audit-data methods and source limits are in Supporting Materials.
NOW LOOK AT REQUESTED SCOPE.
HOW MUCH CAMERA-TIME DID THE REQUESTS COVER?
WHAT IS A DEVICE-HOUR?
Think of it as how much camera-time a request covered. One camera or plate reader included for one hour is one device-hour. Multiply the number of devices by the hours in the requested window.
10 devices × 2 hours = 20 device-hours.
Device-hours measure the scope of requested camera-time—not an operator’s time running a search, footage watched, or footage received. They do not count unique cameras or people, or prove that a request returned a record.
ANALYSIS
Fairfax County Police Department
129.58 BILLION
requested device-hours · approximate delivered-row total
Sixth among individual non-California organizations by requested device-hours (delivered-row total).
“Delivered-row total” adds the device-hours across the audit rows delivered in the export—not footage delivered. The source also compares an exact-distinct total, which counts exact duplicate rows only once.
Virginia’s 14,569 entries are a statewide frequency count; Fairfax County’s device-hours are one organization’s requested camera-time. They are different measures at different levels. An organization can appear less often while requesting searches across more devices or longer time windows.
259,190 ENTRIES AREN'T THE ANSWER. THEY'RE THE REASON TO ASK.
Why did names associated with distant states appear this often?
Why were Texas and Georgia so large?
Why did Illinois and Florida appear tens of thousands of times?
Why were Ohio and Indiana nearly identical?
Why does Fairfax County Police Department stand out in how much camera-time its searches asked to cover?
WHAT CONNECTED THESE SYSTEMS?
WHO COULD REACH WHAT?
WHAT CHANGED?
AND WHY THEN?
WHERE IS THE AUDIT?
UNRESOLVED
What did VCSO’s review find? What caused the issue? Which systems were affected, which relationships were involved, and what changed afterward? Where are the records showing the safeguards worked?
THIS WASN'T JUST A SPIKE. IT WAS A SUDDEN EXPANSION IN SCALE, GEOGRAPHY AND REACH.
The audit records show the footprint. VCSO later acknowledged a “data searching issue.”
Because the next question isn't whether anyone noticed.
The states here are associated with the organization names in the audit files. Those files alone do not fully show who was searching whose data or what each search actually showed the user. How we matched names to states, and the limits of these files, are in Supporting Materials.
0404 · They Knew.
VCSO later gave the February–March period a name: a “data searching issue.”
At Camarillo’s August 26, 2026 City Council meeting, Commander Eric Tennessen acknowledged approximately one month when local data was available for federal and out-of-state searching. The public also received assurances about restrictions and audits. What records substantiate those assurances?
You told the public you were auditing it. Show the audits.
VCSO SAYS
The acknowledged availability
Local data was available for federal and out-of-state searching during approximately one month in February–March 2025. The Camarillo account says there were no such searches since that period. Availability, actual searches and returned information require separate records.
VCSO SAYS
The stated safeguards
VCSO described restrictions following the issue. Reporting of the August 26 update says current sharing was limited to 58 California agencies under written agreements, with no federal or out-of-state sharing partners.
VCSO SAYS
Daily → weekly audits
The February 27, 2026 notice described a February 11 audit; reporting says daily audits were announced on February 27. By the August 26 update, Tennessen described weekly reviews “to verify data security.” These 2026 assurances should not be mistaken for a documented audit start in February 2025.
UNRESOLVEDRecords needed to test the assurance
Show the audits.
Produce the review dates, reviewers, criteria, access and settings checked, findings, exceptions and corrective actions. Show the sharing agreements and change history that substantiate the restrictions—not just the statement that restrictions exist.
The public assurances—and their sources
VCSO SAYS
“As your elected Sheriff, I am committed to transparency, accountability, and maintaining the trust of the communities we serve.”
“Trust isn’t something we ask for,” Commander Eric Tennessen said. “It’s something we earn.”
The Camarillo Caller’s August 27 report attributes this quotation to the August 26 meeting and describes crime-analysis staff reviewing each day’s searches in weekly audits. The original presentation and timestamp are not available here; this is an attributed news account, not an inspected transcript.
VC Reporter’s account reports the 58-California-agency limit under written agreements, also called memoranda of understanding (MOUs).
Source receipts and original records still needed
The official notice and contemporaneous reporting support the attributed assurances. The original presentation, audit workpapers, findings and corrective-action records are not available here.
The public asked for proof the rules were followed—not just totals showing activity.
Residents requested audits, access history, sharing and settings records, retention rules, training, installation and purchasing records. Follow what VCSO said was available, what it produced, and what arrived afterward.
What did “only information available responsive” cover, when another ALPR audit trail was already expected?
Aug. 21, 2026 · Residents ask for the accountability records.
The requests asked for audits and access history; who ran the system and made decisions; sharing and settings; how long data was kept; training; installation; and purchasing records.
Sept. 14 · The first production—and another file already expected.
VCSO described its Organization and Network Audit Logs as the “only information available responsive” to several accountability categories. Separately, staff said that day that an Axon ALPR audit log would follow “in the coming days.” The statements leave the scope of that initial completeness claim unresolved.
Sept. 16–18 · Completeness challenged; more records followed.
The response was challenged on completeness and other issues. Additional months were later supplied.
Oct. 2 · The Axon audit arrives.
The later file shows ALPR activity that the first Flock files could not show us.
0606 · What remains withheld or not provided.
The missing pieces are the difference between “trust us” and “here is the proof.”
Flock logs and the later Axon audit provide activity records. They still do not let residents independently verify who was authorized, why searches were run, what changed, where copies went or what oversight found. “Not produced” does not mean “does not exist.”
ANALYSISVisual 3
Audit Assurances vs. Records Received
Compare the assurances with the detail residents still need to verify oversight.
Flock Organization and Network Audit Logs and an Axon audit trail were received. “Not provided” means missing accountability detail, not no activity files or proof that a record does not exist. Original presentation and response records are needed to verify the quoted wording and categories.
UNRESOLVEDWhat residents still cannot independently verify
Beyond “Trust me, bro”: show the records.
Activity files are not audit findings. A policy is not proof that its safeguards worked.
The productions leave six questions open: who could access the system; why they searched; what settings and sharing changed; where data went; how long each copy stayed; and what reviewers found or corrected. Chapter 4 documents the assurances; Chapter 5 follows the productions.
These are limits of the records produced and available for this review. A missing audit report does not establish that no audit occurred. Missing training records do not establish that users were untrained. The agency’s claims still require records residents can test.
Agency-response summary · report page 4
VCSO SAYSAgency positions, as summarized by the report
The packet reports that purpose/reason information was omitted under an investigatory-records exemption.
The packet says the response described who shared oversight duties, said formal ALPR training records were not kept, and described some records about settings and how long data was kept as unavailable or not tracked.
The packet says the response did not track which accounts belonged to Moorpark users and identified no record answering that request. The analysis therefore cannot reliably connect entries to individual Moorpark users.
Read the report-context receipt. The Civic Review Packet summarizes these agency positions; the September 14 response letter itself is not available here.
ANALYSISVisual 4
Who Can Access the Data?
Who can get in is only the beginning. Who was allowed to do what, and why?
These are possible users and connections, not a checked list of who has access now. The 58-agency statement is what VCSO said at a particular time, not the full history of settings or use.
Training before access; approval to use the system
Training completion and records showing who received an account
“Every search has a legitimate purpose.”
A reason for each search, linked to a case
Access history with reasons and case links, with protected details removed where needed
“The system is audited.”
Recurring review with criteria and follow-up
Daily/weekly/annual audit reports, findings and corrective actions
“No federal/out-of-state access now.”
Limits on sharing and settings that enforce them
Current list of sharing partners, settings and change history
“Images are permanently deleted.”
Rules for every place the data is stored
A map of where copies go, how long each stays, and rules for deletion, backups and shared files
“We protect privacy.”
Limits built into the software and rules for staff
Settings, who was allowed to use what, audits and records of exceptions
Owning a camera is not the same as controlling every use of its data. Deleting an image may leave other copies—and what the software learned.
Follow every kind of data and every place it goes. Software can extract a vehicle’s color, make and other visible features, connect it to other sightings, create investigative files or send information elsewhere. Deleting the original image leaves a question: what remains? The public needs a map showing what the system can do and where the data goes after a camera captures it.
Law and policy receipts
California Civil Code §§1798.90.51–.53 requires ALPR policies to address collection/use/maintenance/sharing, training, monitoring, transfer restrictions, custody/ownership, retention/destruction and periodic audits; §1798.90.52 requires access records to include the purpose for access. CPRA provisions separately address segregable portions, assistance locating responsive information, written denials and determination timing. See the legal source receipts (project-provided compilation).
0707 · The PUBLIC test
Accountability is the floor. Necessity and proportionality are the test.
Even if every rule is followed, the public still deserves an answer: why should government build a searchable history of people’s movements at this scale?
Residents should not have to prove abuse before government has to justify surveillance.
1 · Necessity
What specific public-safety problem requires this camera, connected system or access to a wider network?
2 · Effectiveness
What measurable result shows it actually helps solve that problem?
3 · Proportionality
Could the same result be achieved with fewer cameras, less history or less access to people’s movements?
4 · Accountability
Can residents independently check who was approved and trained, what was searched and shared, how long data stayed, what settings changed, and what audits found or corrected?
Why oversight matters—and why changing vendors is not enough
California’s official ALPR audit and the database-misuse reports it cites show why oversight needs records. Axon/Fusus product descriptions also matter when considering a switch or expansion. What a vendor offers, what VCSO enabled, and what someone actually used are different questions requiring different records.
California State Auditor findings about other ALPR programs are not evidence about VCSO. They show why rules without meaningful auditing can leave misuse undetected—and therefore inform the accountability standard Ventura County should meet.
DOCUMENTEDOfficial California audit finding
No reported misuse is not proof of no misuse.
In its February 13, 2020 Report 2019-118, California’s State Auditor examined ALPR programs at Fresno Police, LAPD, Marin County Sheriff and Sacramento County Sheriff. It found little to no auditing of user searches and weaknesses in access safeguards.
The auditor searched the agencies’ internal-affairs records and found no ALPR-misuse cases. It expressly rejected treating that absence as proof that no misuse had occurred: without regular auditing, misuse could go unnoticed and unreported.
“We do not consider this proof that no instances of ALPR misuse occurred.”
The auditor’s recommendations call for verifiable records of the detection process, rather than assurances that a process exists.
What the auditor found—and the safeguards it recommended
The report’s recommendations show how residents could check oversight:
How audits work: how often they happen, who reviews and approves them, what samples and risks reviewers check, what they record and how problems are resolved.
Usable search records: complete and accurate logs, meaningful case numbers and reasons rather than fields that merely contain some text.
Who gets access: approval, training and prompt removal of access when a user no longer needs it.
The auditor also cites two broader law-enforcement database-misuse reports:
California, 2018: The Auditor cites a news article reporting that CHP investigated 11 database-misuse cases; three involved officers improperly looking up license-plate information through CLETS without a need to know. The audit passage does not identify the individual cases or their dispositions.
Georgia, reported in 2016: The Auditor cites Associated Press reporting that an officer accepted a bribe to search a woman’s plate to learn whether she was an undercover officer. The cited passage does not give a case disposition.
These examples illustrate why purpose and access records need meaningful review.
Unusual activity. The sudden out-of-state surge and the number of entries associated with distant states demand an explanation: who used the system, why, under which settings, and what did searches show?
A system larger than its cameras. Regional networks and recorded activity across connected products make a camera count or one vendor’s file an incomplete answer.
Safeguards must address misuse risk. California’s State Auditor found oversight failures at four agencies and warned that misuse could go unnoticed without regular audits. It also cited reports of CHP database misuse and a Georgia bribe-related plate lookup. Residents need to see how misuse would actually be caught.
Residents still cannot check the whole system. Agency assurances are not a substitute for the records showing who could do what and whether the rules were followed.
Together, these findings and unanswered questions strengthen the case for pausing expansion. Produce the complete record. Let residents and responsible officials independently review whether the system is needed, works, intrudes no more than necessary and follows rules they can check.
PAUSE → PRODUCE → PROVE → REMOVERequire justification and verifiable receipts; remove or disconnect capabilities that cannot meet that burden.
ANALYSISVisual 6
Six Key Takeaways on Surveillance Oversight
Bring the questions together before deciding what public oversight should require.
This image brings the argument together. The findings remain tied to the records and measurements explained above; the illustrations of access and growth are not additional findings.
VCSO and The Axon Switcheroo
(History repeats itself)
Changing the logo doesn't answer what changed behind it.
CHANGING VENDORS IS NOT PROOF OF LESS SURVEILLANCE.
Replacing Flock could be understood as reducing surveillance concerns—or resolving them altogether. But the name on a camera does not tell residents what happened to the searches, connections, data, or permissions behind it.
THE RECORDS DO NOT SHOW THAT SURVEILLANCE SHRANK.
The local record documents Axon testing, a longer retention setting and later ALPR activity through Fusus Web and ECOM. The central question is what these integration layers bring together—and what limits apply to the resulting system.
A product change can leave old powers in place and make new combinations possible. What actually connects locally? Who can ask the combined system about a person, vehicle or place? What legal process, privacy protections and audit trail govern the answer?
Local audit activity is described in the investigation; original exports are not available here. Vendor capabilities are identified before they are discussed: PLATFORM CAPABILITY ≠ CONFIRMED VCSO DEPLOYMENT.
ANALYSIS
THE APPARENT CHOICE: FLOCK OR AXON?
FLOCK → AXON compares vendors. It does not decide how much surveillance Ventura County should allow. The Flock system described in this investigation already included wider-network access, time-and-location searches, plate lookup, vehicle-appearance searches, alerts and sharing.
The meaningful comparison is before and after: what powers ended, what continued and what became possible through a common platform? A camera replacement does not answer those questions.
DOCUMENTED
What VCSO’s records show
The local record describes ALPR tests, changes to how long data was kept, and later searches through Fusus Web / ECOM. It shows activity through those services—not every connection the vendor offers.
PLATFORM CAPABILITYNot confirmed VCSO deployment
What Axon describes
Fusus can bring together information from multiple sources. Product descriptions tell us what the technology can do. VCSO’s purchasing, settings and activity records are needed to show what it actually used.
The real choice isn't Flock or Axon. It's how large a surveillance network Ventura County wants to build.
What VCSO used vs. what the vendor offers
Audit entries describe recorded activity. Product pages describe available functions. To compare the actual systems, residents need a dated list of enabled functions and connections, contracts, settings and access logs.
The record does not establish every marketed capability being used locally, Wingtra use outside its stated mission, or cell-site-simulator misuse.
Illustrative loop of camera views combining. It does not show camera placement or a local integration.
Loop paused until in view
DOCUMENTEDLocal activity described in the investigation
WHAT VCSO’S AXON RECORDS DOCUMENT
Follow the dates. Staff tested the system, changed how long data stayed, and used more ways to search. Those events need an explanation. When a search ran and how far back it looked are different questions.
May 19, 2025
An administrator tests ALPR.
The described test looks back approximately 90 days. Which data was available for it, where did it come from and how long was each connected source allowed to keep it?
May 20, 2025
Data kept longer: 30 → 365 days.
Which store of data—and which copies—did that setting apply to?
May 2025 testing
Hotlists and agency alerts are tested.
The sequence records tests, not the complete history of later alerts or who approved them.
June 2026
Fusus Web: “Comparison Test with Flock.”
The activity history records an ALPR search through Fusus Web with this stated purpose. What was compared? What matched? Had Flock actually been shut down? The label alone cannot answer those questions.
Later in 2026
ECOM appears in the audit sequence.
Another way to search appears in later entries. We still need records showing how the services connected, who could use them and which data they could reach.
July 2026
A search for a vehicle’s likely make within a chosen area.
The review describes a search based on the vehicle make the software inferred, limited to an area. How was the make inferred? What limits applied to the search? The activity entry leaves those questions open.
The dated records: sources, how far searches looked back, and limits
The investigation describes May 2025 testing and the May 20 retention change, followed by Fusus Web / ECOM activity in 2026. The described May 19 lookback, June comparison and July vehicle-make search require the original audit entries for independent verification. Those exports are not available here.
Independent reconciliation requires the underlying entries, exact timestamps, service definitions, retention-setting history, authorization and purpose fields, and the period searched. Chronological proximity does not establish causation. An inferred make is not a verified identity.
Axon describes Fusus combining live video, license-plate information, alerts, dispatch information and other sources. Its product page also markets Axon Air drone feeds for real-time situational awareness and direct evidence transfer to Axon Evidence. These are vendor capabilities, not a verified inventory of VCSO connections.
YOUR DOORBELL DOESN’T HAVE TO BELONG TO THE SHERIFF. Axon describes compatible third-party and community cameras, including owner-controlled live-feed sharing. Listing a camera so investigators can request footage is different from connecting live video.
Which cameras and other sources are actually connected locally? Who opts in, who grants access, under what conditions, and how can access end? The public needs an inventory of what the system can reach, not just what the government owns.
THE SWITCHEROO: SAME QUESTION, BIGGER BOX
PLATFORM CAPABILITY ≠ CONFIRMED VCSO DEPLOYMENT.
The boxes marked PLATFORM CAPABILITY show products the vendor offers. VCSO’s audit, as described in the investigation, records ALPR activity through Fusus Web and ECOM. Its contracts and settings are needed to check which other tools and connections it bought or turned on.
Axon’s community-integration page describes a registry for locating cameras and requesting footage, and CORE integration for compatible camera networks. It describes policy-based sharing controlled by the camera owner. Local participation, configured access, permissions, and revocation still require VCSO-specific records.
Axon says Vision uses software to spot specified activity in connected video and send alerts inside Fusus. Agencies choose the conditions and video feeds; trained people remain responsible for decisions. That describes what the product offers. Records of local use are still needed.
Imagine someone running to catch a bus, pausing outside a closed shop, or helping a friend who has fallen. Ordinary behavior can need context before anyone decides what it means. The accountability question is which defined conditions trigger an alert, how uncertainty is handled, and what a human reviewer sees before acting.
What is turned on, and on which video feeds? How are errors measured and reviewed? Who can change what triggers an alert? What stays in the system after an alert is dismissed? How can the public check that software alerts have not become suspicion nobody can challenge?
What the vendor describes—and what we still need to check at VCSO
Axon describes activity-based detection, human oversight, and agency-controlled settings. It says Vision does not identify individuals, perform facial recognition, or generate behavioral predictions or risk scores. These are vendor descriptions, not independent verification of a local implementation. Ask for any VCSO contract, configuration, evaluation, training, and audit records before claiming deployment.
Axon markets Fusus as a common operational environment for previously separate sources. Other third-party integration layers raise the same question: what can one user now assemble without moving between separately governed systems?
Integration can erode practical boundaries: separate interfaces, owners, permissions and procedures once made different information harder to combine. This is not evidence that Fusus bypasses due process at VCSO. It is a reason to ask which legal and operational boundaries remain effective when sources are connected.
The county should show configuration records, contracts, integration inventories and audit logs: who can access each source, when a warrant or other legal process is required, how source-specific restrictions follow the data, what can be retained or shared, and how improper use is detected. Due process, civil rights and privacy must be accounted for across the combined system—not left between vendors.
Think about an ordinary day: work, a doctor’s visit, lunch, walking the dog, home. Combining sightings can suggest a story about your life without explaining why you were there. The system must be accountable for that inference as well as the individual records.
Work
Doctor
Lunch
Dog walk
Home
YOU ARE NOT THE TARGET
PLATFORM CAPABILITY ≠ CONFIRMED VCSO DEPLOYMENT. THIS IS NOT A LOCAL SYSTEM MAP.
What combined surveillance could mean in an ordinary life—not a record of someone being tracked. The cell-site-simulator purchase and connection are unverified; the graphic cannot establish either.
The privacy question isn't whether one camera sees you. It's how many systems can be asked about you from one place.
Connected does not mean locally enabled
This is a conceptual account of information being combined. It is not a claim that all pictured sources reach VCSO’s Fusus environment, that every moment is recorded, or that a particular person was tracked. The phone / cell-site-simulator element is a separate receipt-pending issue, not a verified Fusus integration.
ANALYSISAn imagined example of the privacy stakes—not a reported incident
YOUR CAR CAN BECOME THE SEARCH
Imagine an ordinary silver Toyota. You drive to work, visit a doctor and stop for lunch. If investigators search by a vehicle’s color, make, type or other visible features within a chosen area and period, your car could be one of many innocent matches. The search could begin without an exact plate number.
Who may approve that search? How large an area can it cover, and how far back can it look? How sure was the software about the vehicle’s make? How are innocent matches ruled out? What reason must the user record? Can someone later piece together exactly what was searched?
What a search by vehicle appearance should record
Produce the search authority, geographic boundary, lookback period, attribute and confidence criteria, treatment of uncertain or innocent matches, and field-preserving audit trail. A matching vehicle description is not proof of the driver’s identity, intent, or wrongdoing.
UNRESOLVEDVendor-described transfers; actual VCSO destinations remain unverified
DELETED FROM WHERE?
Axon says video linked to an incident and other information can move into Axon Evidence. Once systems connect, a statement that data is deleted after 365 days needs a more precise answer: which copy?
Deleted from the original ALPR system? From Fusus? From a downloaded file, Axon Evidence, an investigative file, another agency’s system, a backup or information the software created? Show which copies actually exist and what happens to each.
Produce a map of where the data goes after capture. Name every actual destination, how long each copy stays, who can approve exceptions, who can download or share it, and how VCSO checks that copies elsewhere were deleted.
Follow every copy—not just one setting for keeping data
For each actual flow, document the source, destination, data fields, access controls, retention trigger, legal hold or exception, deletion method, downstream obligations, and audit evidence. A source database’s retention setting cannot independently prove that an export or derived record has been deleted.
VCSO SAYSWingtra RAY: the publicly stated search-and-rescue mission
DRONES CHANGE WHAT CAMERA COVERAGE CAN MEAN
A fixed camera observes where it is installed. Mobile aerial sensors can observe elsewhere. VCSO documents an aerial program and describes Wingtra RAY—a fixed-wing drone that takes off and lands vertically—as used exclusively to locate vulnerable lost or missing people.
PLATFORM CAPABILITY ≠ CONFIRMED VCSO DEPLOYMENT. Axon markets live drone-feed integration through Fusus. VCSO drone ownership plus Axon’s marketing does not prove those drones are integrated into Fusus.
When a common platform can combine cameras, plate data, alerts and live aerial/video feeds, accountability shifts to what the whole system can assemble—and who can ask it to. Show the actual local connections, sensors, approved uses, access, retention and audit records. Preserve Wingtra’s stated mission as a separate limit, not a claim about the entire fleet.
DOCUMENTED
What the county’s report documents
The annual report accounts for the overall aerial program, its uses, costs and planned purchases. It records January 13, 2026 Board approval to purchase two DJI Matrice 4E aircraft for Emergency Services, with possession expected during 2026—not proof of delivery or a platform connection.
PLATFORM CAPABILITYNot confirmed VCSO deployment
What Axon markets—not what is proven connected
Axon’s Fusus page markets integrated drone feeds for real-time response. Which VCSO aircraft or feeds, if any, enter that environment remains a local-records question.
Wingtra mission boundary and the 2025 fleet-wide usage receipt
The official announcement calls Wingtra RAY a search-and-rescue tool “used exclusively to locate vulnerable lost or missing persons.” Keep that agency statement distinct from independent operational verification. Wingtra RAY is a drone; it is not cell-site-simulator technology.
DOCUMENTED — Annual report, printed page 2 (PDF page 3), reports 27 UAVs (unmanned aerial vehicles) and 687 events across the overall program—not Wingtra RAY. Category names follow the report; Tactical Team includes SWAT and TRT. It also records approximately $14,466 approved for two DJI Matrice 4E aircraft for Emergency Services, with possession expected in 2026.
The 70 “Homeless related operations” and 81 Organized Retail Theft events remain questions about approved uses, retention and data destinations. The report does not establish a Flock, Axon or Fusus connection. Neither the annual report nor the Wingtra mission statement establishes a local thermal, night-sensing or LiDAR integration.
Patrol assistance: suspect searches, exterior-building searches, etc.
UNRESOLVEDPurchase and integration remain unverified
A DIFFERENT KIND OF SURVEILLANCE
The Nyxcell / cell-site-simulator claim associated with the conceptual graphic is unverified. A cell-site simulator imitates a cell tower; it is not Wingtra or camera technology. Neither the purchase nor a Fusus connection is established by the records available here.
The purchase and its price still need the County’s own records. What was bought, for what approved purpose, under which policy? Who must approve its use? How long can it keep data, and who independently checks that the rules are followed?
Why this issue remains receipt-pending
The graphic is not a procurement receipt. Neither an independently checked secondary publication nor the County’s primary purchasing record is available here. It does not establish deployment, operational use or integration.
ANALYSISA comparison of reporting rules—not a finding of a legal violation
THE COUNTY ALREADY DOCUMENTS THE HARDWARE
Ventura County already documents major equipment through contracts, purchasing decisions and mandated inventory reports. The Civic Review Packet summarizes the Flock contract; VCSO’s annual equipment report identifies hardware, costs, funding, uses and planned acquisitions. The underlying contract is still needed to independently check the summary.
The annual report describes Ordinance 4603 / AB 481 requirements for covered equipment: Board submission, public posting, a community meeting, use and purpose, complaints, audit findings, corrections, costs, quantities and future purchases.
The accountability gap is system architecture: what connects to what, which functions are enabled, who has access, what legal process applies, how long each copy stays, and whether the audit trail can detect abuse. A hardware inventory does not answer those questions.
WHEN ACCOUNTABILITY IS REQUIRED.
ANALYSIS — This compares governance models. It does not establish that ALPR or Axon is legally covered by AB 481, or that VCSO violated a reporting requirement.
Military-equipment reporting compared with questions still unanswered about connected surveillance
DOCUMENTEDMilitary-equipment report accounts for
UNRESOLVEDConnected-surveillance questions
InventoryQuantities possessed for each covered equipment type.
What can the system do? Which connections are turned on or off?
Use / purposeHow covered equipment was used and why.
Which uses are approved? Where are training records and the dated history of settings changes?
Costs / fundingAnnual costs for each type and the funding source.
What does the connected system cost? Who pays, and what ongoing commitments come with it?
Complaints / concernsA summary of complaints and concerns received.
How can residents raise concerns across connected systems, and how are those concerns resolved?
Audit / policy findingsInternal-audit results, policy violations and actions taken.
Where are audit results and records showing where copies go, how long each stays and whether the connected systems follow the rules?
Future acquisitionsIntended additional acquisitions and quantities.
What future expansion is planned, and what became of decommissioned Flock components?
WE CAN INVENTORY THE HARDWARE. WHY CAN’T WE INVENTORY THE SYSTEM?
Official reporting requirements and a separate network-video research lead
DOCUMENTED — The report is dated April 22, 2026 and covers January 1–December 31, 2025. Printed page 1 (PDF page 2) sets out the six annual reporting categories, website-posting requirement and community-engagement requirement. Those requirements concern equipment covered by Ordinance 4603 / AB 481; this comparison does not establish ALPR/Axon coverage.
UNRESOLVED — Printed page 5 (PDF page 6) reports approximately $15,229 from the existing FY 25/26 budget for “network video upgrades, enhanced recorders and associated equipment.” This appears under tracked armored vehicle / ROOK equipment costs, not as an ALPR procurement line. What equipment, vendor, configuration and retention rules did that expenditure cover?
The report does not establish any relationship between that $15,229 expenditure and ALPR, Axon or Fusus. It is a follow-up lead only, not evidence of a surveillance integration or misconduct.
Residents should not have to reconstruct a policy decision from product pages and scattered audit entries. The choice to connect sources should leave a decision file: alternatives, approved limits, responsible officials, contracts, settings, training and meaningful audits.
We can inventory major equipment. The public still needs the architecture and the decisions behind the connected system.
The problem officials wanted to solve and the alternatives they considered.
A comparison of privacy effects and what each system can do.
A dated list of Axon tools and connections turned on or off.
The plan for switching systems and shutting the old one down.
What happened to Flock cameras, accounts, network connections, sharing permissions and stored history.
A map showing where data goes, who receives copies and how long each copy stays.
Training records sufficient to verify the assurance, and independent audit records.
The decision file residents should be able to inspect
Produce dated decision records, options evaluated, approved limits, accountable decision-makers, configuration changes, decommission evidence, training documentation, and audits. A missing receipt in the produced package is not proof that the underlying record never existed. Redactions should protect legitimate interests without making the governing decision impossible to verify.
The missing training receipts leave the assurance unverified; they do not establish that employees were untrained.
Show the before-and-after record: cameras removed, accounts closed, connections and sharing revoked, historical data deleted or moved, functions retained or replaced. A canceled contract and a different logo are not proof of a smaller system.
DOCUMENTED
Retirement with an inventory explanation
VCSO’s 2025 annual report says its unmanned-ground-vehicle inventory declined from eight robots reported in 2024 to six. It explains that obsolete or damaged robots were returned or destroyed because they were no longer serviceable. Source: printed page 2 (PDF page 3).
ANALYSIS
Explain what happened to the old system.
When VCSO retired covered equipment, it recorded the inventory change and explained what happened to it. Residents should be able to get that same basic account of retired surveillance equipment, access and data.
If Flock disappears from the roadside while its function is absorbed into a larger connected surveillance environment, that isn't necessarily less surveillance. It may just be surveillance with a different logo.
Evidence of an actual reduction
Publish a before-and-after capability matrix, camera and account inventories, revoked network and sharing relationships, integration shutdown records, historical-data disposition, and dated tests showing which access paths no longer work. Identify what continues under the replacement system and the authority for keeping it.
Judge the transition by what actually connects, what powers remain and what safeguards can be independently tested. Integration is not inherently proof of misconduct; neither is a new vendor proof of restraint.
The county should produce the system map, contracts, enabled-function inventory, permissions, legal-process rules, data-retention map and audits capable of detecting abuse. Those records belong beside—not behind—the hardware inventory.
ANALYSISWhat officials should require
SHOW THE SYSTEM. SHOW THE SAFEGUARDS WORK.
Apply the county’s existing documentation discipline to the architecture, not just the gear. Explain who can combine which sources, under what authority, and how review detects personal misuse, improper access, settings failures and retained copies. “No abuse discovered” is meaningful only when the detection process is visible.
This asks officials to extend that public reporting model. Whether AB 481 already covers ALPR is a separate legal question.
HARDWARE, CONNECTIONS, AUTHORITY, ACCESS, RETENTION, AUDITS. ACCOUNT FOR THE WHOLE SYSTEM.
DON’T SELL US A NEW LOGO. SHOW US THE SYSTEM — AND SHOW US IT ISN’T BEING ABUSED.
0909 · Abuse of this technology does NOT stay theoretical
Abuse of this technology does NOT stay theoretical
Safeguards can fail because of a person, because someone misuses their access, or because the system itself is set up wrong. To catch the problem, you have to be able to follow all three.
THE QUESTION ISN'T WHETHER MISUSE IS POSSIBLE. IT'S WHETHER THE SYSTEM CAN FIND IT WHEN IT HAPPENS.
PERSON→ACCESS→SYSTEMPERSON, then ACCESS, then SYSTEM
9.1 Local VCSO and Southern California Examples
Start close to home. One case raises a question about what sworn personnel could access. Another involves an alleged abuse of government information and tracking. The third concerns the settings of the surveillance system itself.
9.1.1 Camarillo: “You mean my buddies? I’m a sheriff.” — The El Rey Cantina Case
PENDING CRIMINAL CASEOpen case fileClose case file
A serious incident involving Sheriff’s personnel. Three reported accounts—and a test of what the technology can help reconstruct.
The charges remain pending. The allegations have not been established by a criminal conviction. The Star reported that both deputies pleaded not guilty; its March 2, 2026 follow-up described the prosecution as continuing.
DOCUMENTED
ONE NIGHT. HOW MANY GOVERNMENT RECORDS?
WHAT PROSECUTORS SAY HAPPENED
The Ventura County District Attorney places the incident shortly after midnight on March 30, 2025, at El Rey Cantina in Camarillo. Off-duty Ventura County sheriff’s deputies Dylan Davis and Anthony Malagon were both arrested that night, the Star reported. [1][3]
Prosecutors say Davis and his wife argued inside El Rey. The confrontation continued outside and allegedly became physical. The bar’s security guard intervened after allegedly witnessing Davis assault his wife. [1]
Another off-duty female deputy allegedly tried to restrain Davis. Prosecutors later charged him with misdemeanor battery related to her. [1][3]
Both Davis and Malagon face felony assault charges for allegedly using force likely to cause great bodily injury. Davis additionally faces felony spousal battery and misdemeanor battery. These are the charges announced in the District Attorney’s July 16, 2025 release. [1]
WHAT THE SECURITY GUARD REPORTED
In the Ventura County Star’s reporting, the security guard said Davis’s wife slapped Davis. He said Davis then slammed her into a bench and choked her with both hands. The security guard said he intervened to tell Davis to back off. [2][3]
The security guard alleged that Davis threatened to kill him and punched him in the jaw. He said Malagon tackled him from behind, and that both deputies kicked and stomped him while he was on the ground. [2][3]
The Star reported that the security guard went to the hospital with a concussion and injuries to his jaw, head, neck and back. The District Attorney’s announcement confirms that he was taken to a hospital, treated for his injuries and later released. [2][3][1]
The security guard told the Star that he used another person’s phone to call 911 after his own was smashed. He alleged that, after he said he had called the police, Davis invoked his Sheriff’s status. [2]
“You mean my buddies? I’m a sheriff.”
Words the security guard attributed to Davis, as reported by the Ventura County Star on May 23, 2025. [2]
WHAT THE CAMERAS REPORTEDLY SAW
KTLA reported that restaurant surveillance captured portions of the events inside El Rey, including Davis reportedly pushing his wife several times. [6]
Interior footage is one part of the timeline. It does not establish every later allegation about what happened outside.
ONE NIGHT. HOW MUCH COULD THE SYSTEM REMEMBER?
Start with the actual streets: the US-101 crossing, the signal-controlled junction where Arneill Rd ends at Ventura Blvd, and El Rey’s frontage. What records could reconstruct one night in this ordinary setting?
Arneill Rd crosses US-101 / Ventura Fwy and ends at Ventura Blvd; it does not continue through the restaurant block. Street photographs show the signal-controlled T-junction. Chapel City Church is north of Ventura Blvd; El Rey is south. This is geographic context, not a camera-coverage map.
Street View labels this location 147 Arneill Rd. The road rises toward the freeway crossing. Visible signals and pole-mounted fixtures do not establish equipment make, operator or ALPR function.
Source: Google Street View, imagery dated March 2026. Privacy-masked display copy; originals are available below.
Turn arrows, signals and restaurant frontage corroborate Arneill’s termination at Ventura Blvd. Signal heads and camera-like fixtures are not independently identified ALPR equipment.
Source: Google Street View, imagery dated March 2026. Privacy-masked display copy; originals are available below.
El Rey’s frontage and the Old Town clock locate the venue. Nearby fixtures are not verified ALPR readers. This is a March 2026 streetscape, not an incident still.
Source: Google Street View, imagery dated March 2026. Privacy-masked display copy; originals are available below.
FULL ORIGINAL SCREENSHOTS AND LABELED SUPPLEMENTAL SKETCH
The photographs above are privacy-masked display copies. Compare the originals here; the separately labeled AI-assisted sketch is a reading aid, not evidence.
ANALYSISAI-assisted sketch — not evidenceA simplified drawing that follows the original map: US-101, the Arneill Road overpass, Ventura Boulevard and El Rey.Open source image (new tab): AI-assisted sketch — not evidenceWhat this image shows—and its source limits
Drawn with AI assistance from the junction screenshot, in the same orientation. It shows only roads, the signal-controlled T-junction and the venue address. It is not a survey.
Source: AI-assisted representation based on Google Maps.
Read this first. AI-ASSISTED GENERATIVE REPRESENTATION — NOT A SURVEY OR CAMERA-LOCATION MAP. This sketch simplifies the map. Check minor discrepancies against Google Maps and the originals; no camera placements, incident sightings or vehicle route are asserted.
These Google Street View screenshots are dated March 2026, after the March 30, 2025 incident. They provide location and streetscape context, not proof of what hardware was installed or operating on the incident date. Visible devices have not been independently identified as ALPR readers.
The street setting and visible fixtures are shown in March 2026 imagery. ALPR function and map markers remain unverified. A photograph does not establish that a relevant vehicle was captured.
REPORTED HERE
Supported event/location. Reporting does not establish every allegation.
COULD HELP RECONSTRUCT
Possible system/capability; applicability not established. This does not establish that Fusus was used or an official vehicle was involved.
Walk through how one night could leave records across multiple systems.
EXPAND THE RECONSTRUCTION · 4 MIN READ
1 START WITH AN ORDINARY DRIVE
Cross US-101 on Arneill Rd toward Old Town. The bridge brings you to the signal-controlled T-junction at Ventura Blvd, where Arneill ends. El Rey Cantina, at 2302 Ventura Blvd, is on the south side of Ventura Blvd, between Elm Dr and Palm Dr.
Street geography and surveillance coverage are separate questions. Identifying a device still would not prove it recorded a particular vehicle or operated on the incident night.
A plate reading can connect a vehicle to a location and a time. If multiple readers recorded the same vehicle, those records could form a timeline. Whether such readings exist here is a question for the records.
vehicle→location→timevehicle, then location, then time
2 THEN THE VEHICLE STOPS
People leave their cars. Potential evidence expands beyond plate readers: restaurant and nearby cameras, 911 calls, dispatch and radio traffic, responding deputies’ records, body-worn cameras, and investigative or digital evidence.
Fusus matters conceptually because connected systems can bring separate information together. That capability is not evidence that Fusus was used in this incident.
Restaurant surveillance reportedly captured part of the El Rey incident. That is one reported source of evidence, not proof that every nearby system captured it. [6]
3 THEN 911 IS CALLED
The security guard told the Star that he called 911 using another person’s phone after his own was smashed. That account raises specific questions about the call and the response. [2]
When was the call received, and what was reported?
When was it dispatched, which units were sent, and when did they arrive?
What records did the responders make, and what did Major Crimes collect?
One physical event can now exist as fragments across numerous information systems.
4 NOW LOOK BEYOND EL REY
The wider reconstruction asks about ALPR infrastructure elsewhere on Camarillo’s roads. Each proposed installation needs location-linked evidence identifying the hardware and its operation. The map alone cannot establish coverage.
The question shifts from “Did one camera see a vehicle?” to “How much of a vehicle’s journey could the network reconstruct?” This is a possible sequence to test against records, not a route or a claim that these sightings occurred.
earlier location→approach→arrival→departure→later locationearlier location, then approach, then arrival, then departure, then later location
That is the difference between a camera and a network.
5 NOW TURN THE SYSTEM AROUND
These systems are justified partly as ways to reconstruct public activity. Test that capability here instead of guessing what happened.
How did everyone arrive, and what vehicles were involved?
What were the arrival and departure times?
What did restaurant cameras, 911 and dispatch, responders, and Major Crimes record or collect?
What was preserved?
Official government vehicle involvement is not established. Only if independent evidence later establishes that an official vehicle was involved should these additional questions be asked:
Which vehicle, who operated it, and what was the operator’s duty status?
Which systems were actually installed in or associated with that vehicle?
For those established systems, what records were generated, collected and preserved?
We do not need to guess the answers. That is the point of having records.
6 AND THEN ASK WHO COULD ACCESS THE SYSTEM
We presently have no evidence that Davis or Malagon improperly accessed ALPR, Flock, Fusus, Axon, ECOM or other surveillance systems in connection with this incident. Access and relevant use must be established from records, not inferred from their employment.
Who had access, and to which systems?
Were relevant people or vehicles searched? When, and for what reason?
What information was returned, and who reviewed the activity?
If nothing was searched, show that.
If they couldn’t access the system, show that.
An audit trail capable of exposing misuse should also be capable of clearing someone when misuse did not occur.
7 LOOK AT THE MAP AGAIN
A freeway.
A bridge.
A signal-controlled T-junction.
A restaurant.
Cars going somewhere completely ordinary.
And the question of which surveillance systems could have created relevant records.
The maps and photographs establish the street setting. The records must establish which systems operated, what they captured and who could access them.
The public lives in this ordinary environment. If surveillance systems can reconstruct our activity here, accountability should run both directions: show what relevant records exist, what they establish, and who could access them.
DON’T ASK US TO TRUST THE NETWORK.
SHOW US WHAT THE NETWORK REMEMBERS.
A) SURVEILLANCE AS EVIDENCE
What can the surveillance and digital-evidence systems establish about the event?
A serious incident occurred involving multiple Sheriff’s personnel. Contemporary surveillance reportedly captured part of it. Government investigative records reconstructed another part. If additional government technology recorded relevant events, preserve it, identify it and account for it. [1][6]
The question is what relevant records can establish—not an assumption that any particular system captured the incident.
What surveillance footage was collected and preserved?
What does it establish about arrival, departure and the timeline?
What 911 recordings and call records exist?
What do the responding officers’ and investigators’ records establish?
What other relevant government-generated digital records did Major Crimes identify?
Were potentially relevant records preserved?
B) SURVEILLANCE AS POWER
What parts of the surveillance and investigative systems could the involved sworn personnel themselves access?
We currently have no evidence that Davis or Malagon used ALPR, Flock, Fusus, Axon or ECOM in connection with this incident.
“No access” or “access but no relevant use” are meaningful, exculpatory answers—answers that can clear someone of a concern. A complete audit trail should be able to establish them, or show that a relevant search had a legitimate reason.
Which sensitive systems could each deputy access?
When did that access begin and end?
Did either search any person or vehicle connected to the incident?
What reasons were entered?
What information was returned?
Who reviewed the activity?
HOW DID EVERYONE GET THERE?
Arrival and transportation may be reconstructable through independent evidence. What do the restaurant’s cameras and exterior surveillance show about how people arrived, who traveled together and when they left?
Only if independent evidence establishes that an official vehicle was involved should investigators follow the vehicle-specific questions below. First establish its identity and actual equipment; do not assume a vehicle or its records exist.
Which official vehicle was it?
Who operated it, and what was that person’s duty status?
What systems were actually installed in or associated with that vehicle?
For systems established to be installed or associated, what relevant records were generated and preserved? That may mean asking about GPS or automated vehicle-location (AVL) records, an in-car computer (MDT), Axon or ALPR—only where the vehicle’s configuration supports the question.
Did Major Crimes obtain those relevant records?
ONE INCIDENT CAN LEAVE A LOT OF DIGITAL FOOTPRINTS
Which of these records exist for the incident, who holds them and how long are they kept? These are places to look, not a claim that every record was created or has been provided.
Restaurant surveillance video and its original timestamps.
The 911 call, call records and dispatch records.
Records made by responding officers and investigators.
Major Crimes investigation records.
Arrest and booking records.
Administrative and internal-affairs investigation records.
Court filings and hearing records.
POST certification actions.
Authenticated government-system logs, if applicable: accounts, access dates, searches, reasons, returned information and review.
THE INVESTIGATION AND ADMINISTRATIVE RESPONSE
VCSO’s Major Crimes Unit investigated the incident. The Star reported that both deputies were placed on leave and that the California Commission on Peace Officer Standards and Training (POST) suspended their peace-officer certifications. Its March 2, 2026 follow-up reported that the suspensions and paid leave continued while the case proceeded. [1][3][5]
An internal-affairs investigation was also reported. These processes leave records of their own: what investigators examined, what administrative action followed and when it happened. [4]
THE CIVIL CLAIM AND THE COUNTY’S RESPONSE
The security guard filed a government claim for damages in April 2025. The Star reported on October 26, 2025 that the County paid $440,000 in July to settle the claim before it went to court, citing records obtained from the County. The County continued to deny and dispute the allegations. A settlement is not an admission. [4]
VCSO demonstrated that it can investigate its own. Its Major Crimes Unit investigated this incident. Prosecutors filed charges. Administrative processes followed. [1][3]
WE'RE ASKING FOR THE SAME SERIOUSNESS WHEN THE EVIDENCE LIVES INSIDE A SURVEILLANCE SYSTEM.
This is not asking VCSO to prove every employee is good. It is asking whether the system can establish what happened when access becomes relevant. The same seriousness should apply to a preserved recording, an account history, a search or a record of what someone saw.
BECAUSE ‘WE FOUND NO MISUSE’ ONLY MEANS SOMETHING IF THE SYSTEM IS CAPABLE OF FINDING IT.
Official charging announcement, account of the alleged confrontation, hospital treatment and Major Crimes investigation. A charging document is not a conviction.
Reports the government claim, the July settlement, the County’s continuing denial, not-guilty pleas and the internal-affairs investigation. The article cites County records; the underlying settlement agreement is not available here.
Reports the prosecution continuing, not-guilty pleas, paid leave and continuing certification suspensions. No later disposition is established by the sources reviewed here.
Reports on interior restaurant video. No original video is hosted here, and the described interior footage is not treated as a complete recording of events outside.
9.1.2 Bell — Government access becomes personal
PENDING CRIMINAL CASEOpen case fileClose case file
Bell, Los Angeles County, California—not Ventura County. Prosecutors allege government information and a GPS tracker were used for personal reasons.
This is a pending criminal case, not a conviction. The allegations below are those announced by the Los Angeles County District Attorney on August 3, 2026.
DOCUMENTED
He didn't need Flock. He allegedly had access.
Government database→Personal relationship→Physical location trackingGovernment database, then Personal relationship, then Physical location tracking
On August 3, 2026, the Los Angeles County District Attorney announced charges against Bell Police Officer James Ixcoy. Prosecutors allege that on July 17, 2025 he ran an unauthorized California DMV registration inquiry on a car belonging to a man who had also dated his ex-girlfriend. [1]
They also allege that on August 18, 2025 he placed a GPS tracking device on his former girlfriend’s vehicle. The device was later found, according to the announcement. [1]
The charges are one felony count of accessing and using computer data without permission and one misdemeanor count of unlawful use of an electronic tracking device. The announcement says Bell Gardens Police Department was investigating the case. [1]
The technology is different. The control problem is the same.
This is why ‘authorized user’ is not the same thing as ‘authorized use.’
An account can belong to a police officer. That does not make every personal reason for using it a lawful investigation.
Camarillo asks whether the deputies could access sensitive location systems. In Bell, prosecutors allege that an officer abused government information and tracking involving an ex-partner. That is why the first question isn’t paranoid. It is an accountability question.
What reason was recorded for a lookup? Would a reviewer have noticed a personal connection? Was the device’s use approved? What records could show the answer—and what would alert someone that something was wrong?
How was it discovered?
The District Attorney’s announcement says the tracking device was later found and that Bell Gardens Police Department was investigating. It does not explain who found the device, how the DMV inquiry was uncovered or whether an audit first detected either alleged act. [1]
The public record reviewed here does not yet establish the complete detection path.
Primary source for the July 17, 2025 DMV inquiry allegation, the August 18, 2025 GPS allegation, charges and investigating agency. The DMV inquiry concerned a man who had dated the ex-girlfriend; the GPS allegation concerned her vehicle. The announcement schedules an October 2 arraignment but does not establish its outcome or a complete detection path.
9.1.3 Ventura County — Nobody has to go rogue
SYSTEM / GOVERNANCEOpen case fileClose case file
The local Flock issue is a settings-and-oversight example—not a case alleging officer misconduct.
This example concerns the system and its rules. The statements about the cause, investigation and restrictions below are VCSO’s account, not an independently reconstructed change history.
VCSO SAYS
Nobody has to go rogue for a safeguard to fail.
Data became searchable beyond California
VCSO’s February 27, 2026 notice says that from February 19 through March 19, 2025, local Flock ALPR data became searchable by out-of-state and federal law-enforcement agencies without the agency’s knowledge or approval. The later Camarillo account called the period a “data searching issue” and described availability for approximately one month. [1][2]
VCSO said its internal investigation determined that nobody from the agency activated the national lookup feature. Its notice also said Flock could not identify the specific cause because of limitations in technical logging. [1]
The notice described a possible agency administrator action, a Flock employee action or a software bug; VCSO said it ruled out the first through its internal investigation. Those possible explanations are not a demonstrated cause. [1]
VCSO said that on or about March 19, 2025, Flock removed California agencies’ ability to activate national lookup. The notice described later restrictions and enhanced audits. Reporting of the August 26, 2026 Camarillo update also described current sharing as limited to 58 California agencies, with no federal or out-of-state agencies. [1][3]
Being able to search the data and actually receiving a plate image or other information are different questions. VCSO’s notice said it did not know whether out-of-state or federal agencies received images or information captured by its readers. [1]
What changed—and how would anyone know?
If a safeguard depends on a setting, residents need a record of that setting—not just an account of what users were supposed to do.
Who or what changed the setting, and when?
Was it a vendor action, a default, a software change, an administrator action or an inherited setting?
Who could see the change? Which records would show it?
Who discovered the problem, and how?
Which data became available? Who searched it, and what information came back?
When was access closed? Which records confirm that?
What prevents the same thing from happening again—and how is that protection checked?
Auditing people isn't enough. You have to audit the system too.
Check the users and their reasons. Also check the settings, outside access, software changes and whether the records can show what happened. The agency’s own account raises both kinds of oversight question.
Existing site citation. Agency account of the February 19–March 19, 2025 availability, its internal investigation, limited technical logging, possible explanations, later restrictions and unknown returned information.
Existing site source note for the “data searching issue” wording and approximate one-month representation. The original presentation, slides and audit workpapers are not included.
Existing site citation for the reported 58-California-agency limit and no current federal or out-of-state sharing. A statement about current settings is not a complete history of prior access.
National examples
9.2 When the Search Becomes Personal
These cases occurred outside Ventura County. They do not establish misconduct by VCSO. They show why searchable location histories require independently verifiable access controls and oversight.
9.2.1 Indianapolis, Indiana · September 2026
He kept showing up. The searches may explain why.
CHARGES / ALLEGATIONS — NOT CONVICTIONS
A gym. A store. A school function where a woman’s daughter was present. WTHR’s account of a probable-cause affidavit describes women reporting encounters with Indianapolis Metropolitan Police Department officer Jonathan Schultz that turned ordinary routines into an unsettling question: how did he know where to appear?
According to that reporting, Schultz allegedly made 1,396 Flock searches of 11 women’s license plates from August 2023 through August 2026, including plates associated with former or current romantic interests. The reported encounters and searches are allegations—not a finding that every encounter resulted from an ALPR query.
In September 2026, Schultz was charged with 12 counts of official misconduct, 12 counts of fraud and two counts of stalking. Five current or former IMPD officers were charged in the broader investigation.
The detection story matters too. IMPD’s official September 23 statement says internal audits in August and September 2026 identified potential misuse. The department revoked ALPR access and initiated administrative and criminal investigations. This was not misuse that could never be audited: auditing helped uncover it.
“The allegation is that the officers were searching or running the license plate information and tracking individuals through the Flock system.”
Marion County Prosecutor Ryan Mears, quoted by WTHR.
A policy tied to a reason
IMPD says its August 2026 policy requires a case or incident number and a NIBRS offense code for every search. It also reports stronger auditing and single sign-on to reduce credential-sharing risks.
A review someone else can check
Which records connect the user, search, case, purpose and review? Could an independent reviewer reproduce the department’s detection of improper access?
The charges and search allegations are reported by WTHR. The probable-cause affidavit was described by that outlet, not independently inspected for this section. The supplied sources do not establish convictions or a final adjudication.
Investigative findings
IMPD’s official release confirms internal audits, access revocation and administrative and criminal investigations. These are agency representations of the detection and response, not a claim that independent review has validated every audit finding.
Vendor and search-scope interpretation
WTHR identifies Flock. No separate vendor response is supplied here. The alleged 1,396 searches of 11 women’s plates are query counts and a reported plate population—not a count of successful matches, sightings or proven tracking encounters.
Public log availability
These sources describe audit activity and affidavit allegations; they do not provide a complete public query-log dataset. That does not establish that the logs are unavailable or do not exist.
Warrant status
The supplied sources do not establish warrant or court-authorization status for each ALPR query. Unknown warrant status is not proof that no warrant existed.
9.2.2 Braselton, Georgia · November 19, 2025
What happens when the person accused of stalking is the police chief?
GBI INVESTIGATION / CHARGES
The person at the top of a department can also be the person whose access needs scrutiny. On November 19, 2025, the Georgia Bureau of Investigation announced that Braselton Police Chief Michael Steffman had been arrested and charged with violation of oath by a public officer, stalking, harassing communications and multiple counts of misuse of automated license plate recognition systems.
The Jackson County Sheriff’s Office requested the external GBI investigation. After a multi-month investigation, GBI said Steffman had misused ALPR systems “to harass and stalk multiple individuals.” That is an attributed investigative finding accompanying charges—not an adjudicated conviction.
An external investigation did occur. The oversight question is therefore not whether outside review was possible, but who independently reviews surveillance access before a complaint or an outside investigation becomes necessary.
“to harass and stalk multiple individuals.”
GBI’s description of the alleged ALPR misuse in its November 19, 2025 announcement.
Who reviews the chief?
A department leader’s authority does not remove the need for review of that leader’s searches, purposes and access.
Outside investigation is not routine oversight
The sheriff’s request and GBI investigation show an external path existed. What independent checks operate before allegations trigger that path?
GBI is the primary source for the arrest, listed charges and its stated investigation findings. The announcement says the case file would be provided to the Piedmont Judicial Circuit for prosecution. It does not establish a conviction.
Later case-status reporting
On April 24, 2026, FOX 5 reported that Steffman had been found dead, citing a source. GBI confirmed a death investigation, but authorities had not officially identified the deceased in that article. The supplied reporting does not establish an adjudicated conviction, the cause of death or the final disposition of the criminal case.
Vendor and search scope
GBI’s release does not identify the ALPR vendor, search volume, geographic reach, lookback periods or successful matches. Do not substitute Flock or infer those quantities.
Public log availability
The supplied sources do not establish whether query logs are publicly available. Absence of a public dataset in these sources is not proof that no audit trail exists.
Warrant status
The release and later reporting do not establish warrant or court-authorization status for the individual ALPR searches.
9.2.3 Johnson County, Texas · May 9, 2025 searches · October 7, 2025 EFF reporting
She had an abortion. Investigators searched a nationwide camera network.
The reason recorded for two May 9, 2025 Flock searches was explicit: “had an abortion, search for female”. According to EFF’s account of the search logs, the first query covered 17,684 cameras across 1,295 networks with a seven-day lookback. The second widened to 83,345 cameras across 6,809 networks and a month of data.
Cameras in reported search scope
83,345
Networks in reported search scope
6,809
Reported second-query search scope: cameras across networks, with a one-month lookback. These figures are not actual sightings, matches or cameras that observed the person. Source: EFF’s account of Flock search logs.
The numbers describe the reported scope of the queries—not how many cameras saw the woman or her vehicle. The distinction matters: an enormous search boundary is not evidence of an enormous number of actual sightings.
EFF’s October 7, 2025 report says an incident report and sworn affidavit describe a death investigation involving a nonviable fetus, collection of evidence, consultation with prosecutors about potential charges and later examination of text messages. According to those records as reported by EFF, prosecutors advised that the woman could not be charged for taking medication to cause the abortion or miscarriage.
Flock and the sheriff disputed the characterization, saying the woman was sought for welfare or missing-person purposes. EFF argues that the underlying records conflict with that explanation. This section presents that dispute with attribution; it does not assert a judicial finding that the searches were illegal.
“She was being searched for as a missing person, not as a suspect of a crime.”
Flock Safety’s response, describing the sheriff’s account.
The recorded reason
“had an abortion, search for female” — the exact reason reported for both queries. A reason field makes a purpose review possible; it does not by itself prove a search was justified.
The explanation and the records
Flock describes a missing-person search. EFF interprets the incident report and affidavit as an abortion-related death investigation. Independent review requires the records—not simply choosing one account.
EFF reports obtaining search logs, an incident report and a sworn affidavit. Its reporting describes a death investigation involving a nonviable fetus, evidence collection, prosecutor consultation and review of texts. These documents were reported by EFF, not independently authenticated for this section.
Recorded query scope
May 9, 2025: first query — 17,684 cameras, 1,295 networks, seven days; second query — 83,345 cameras, 6,809 networks, one month. Both reportedly used the reason “had an abortion, search for female”. These are reported query boundaries, not counts of detections, successful returns or unique people tracked.
Vendor statements and disputed interpretation
Flock’s statement says the woman was sought as a missing person, not a crime suspect. The sheriff also described a welfare purpose. EFF argues the underlying records conflict with that explanation. Attribute both accounts; the supplied sources do not establish a judicial finding of illegality.
Charges and case status
EFF reports prosecutors advised the woman could not be charged for taking abortion medication. Flock says no charges were filed against her. This account is not a conviction or an adjudication of the legality of the ALPR searches.
Public log availability
EFF publishes reporting and record excerpts and says it obtained the relevant logs. That is not a complete nationwide query-log dataset or proof that every requested record is publicly available.
Warrant status
The supplied accounts do not establish warrant or court-authorization status for these specific ALPR queries. The existence of a sworn affidavit does not itself establish a warrant authorizing the searches.
These cases did not happen in Ventura County. They demonstrate why independent oversight matters here. The question is not whether every person with access will misuse it. The question is whether the public can verify that access is justified, properly limited, and independently reviewed—and whether misuse can be discovered when safeguards fail.
What would Ventura County’s records allow an independent reviewer to discover—and what might remain invisible?
9.3 — International examples
No international examples are included in this release.
The cameras should not stay simply because they are already there. Residents and responsible officials should review the evidence independently and decide whether this surveillance deserves to continue.
PAUSE.PRODUCE.PROVE.REMOVE.
Pause expansion, contract renewal and the move to another system while the record is incomplete. Produce the records showing who had access, what was shared, which settings changed, what audits found, and where the data and its copies went. Prove that the system is needed, works, intrudes no more than necessary and follows rules the public can check. Remove or disconnect cameras, connected systems and access that cannot meet that burden.
Atmospheric loop, illustrative only. Not incident footage.
Loop paused until in view
Put the question on the right desk.
VCSO leadership
Sheriff James “Jim” Fryhoff. Contract-city police chiefs include Captain Nick Odenath in Moorpark and Commander Eric Tennessen in Camarillo.
Find the office responsible for the records or policy, then prepare your own correspondence. Choose a jurisdiction or explore the full sourced roster. Nothing is sent for you.
Public contacts · 108 sourced entries
Public Officials Explorer
Find the right office for records, operations or policy. Choose an area yourself, or browse the complete roster. Your location is never inferred.
Coverage and verification gaps
Directory reviewed 2026-10-11. Each field carries its own source, date and status. A directory review is not a fresh verification of every field. Missing values were not verified in checked sources, not necessarily unavailable everywhere. Confirm current roles and channels with the official office.
All ten cities: 54 mayor/council seats, ten city manager offices, ten city clerk offices and ten police leadership offices. County coverage includes all five supervisors, the sheriff and separate records desks. Intersecting districts: Assembly 38/42/46, Senate 21/27 and House 24/26/32; see the county’s October 8, 2026 incumbent list. Statewide oversight includes the governor, attorney general and both U.S. senators. Regional coverage is limited to verified VCTC offices.
Congressional boundaries changed after Proposition 50. Current incumbents and election districts are not interchangeable; verify your address on the official lookup. Government forms may restrict submissions to constituents.
4 entries name an office rather than a verified individual; 22 have no listed public email, 2 no business phone, 32 no usable portrait. Missing fields are not guessed. Initials are labeled when a portrait is missing.
The full roster is not a claim that every city deploys the systems in this report. Legislators and federal officials are policy contacts, not CPRA custodians of local records.
Start with an area to see local responsibilities. Explore all officials to browse the full roster without choosing an area.
Draft a message
With JavaScript on, this section builds an editable public records request (CPRA) or a separate policy letter for a chosen city, which you review and send from your own email app. Without it, use the contact details and official links above.
Local log
Optional local request logging is available with JavaScript enabled. Entries stay in your browser.
Supporting Materials
Want to check the work? Please do.
Check the work for yourself. The report, methods, legal sources, charts and details about the data are here when you want them.
Available materials vs. originals. Included here: charts, data files, the Civic Review Packet and collected legal sources prepared for this review. The packet explains specific findings page by page. The original Flock/Axon audit files, Camarillo presentation and records-request correspondence are not included.
ANALYSIS
The actual report
Praeva Analytics ALPR Audit Records Civic Review Packet · 33 pages · September 2026
Read the complete report alongside its charts. This packet contains the analysis prepared for the review, not the agency’s original audit files, response letters or presentation.
Source notes record verification against official California Legislative Information on October 4, 2026:
Civil Code §1798.90.51 — operator policy must address collection, use, maintenance, sharing and dissemination; authorized roles and training; monitoring; sharing and transfer; custodian or owner; retention and destruction process.
Civil Code §1798.90.52 — access record includes date/time, query elements, username/organization, and purpose; access limited to authorized purposes.
Civil Code §1798.90.53 — end-user policy includes monitoring and a process for periodic system audits, plus sharing/transfer, custodian/owner, retention/destruction.
Gov. Code §7922.525(b) — reasonably segregable portions available after exempt portions deleted.
Gov. Code §7922.530 — identifiable nonexempt records promptly available; not a fixed final-production deadline.
Gov. Code §7922.535 — 10-day determination/notice rule; written unusual-circumstances extension subject to statutory limits.
Gov. Code §7922.540 — written partial/complete denial; names/titles of persons responsible; withholding must be justified.
Gov. Code §7922.600 — duty to assist includes identifying responsive records, describing IT/physical location, and suggestions to overcome practical access barriers.
Project-produced-source cross-checks:
VCSO Policy 429: department-approved training; five-year server retention with exceptions; regular audits; inter-agency agreement/access controls.
Flock order form: `Retention Period: 365 Days`; extended retention up to one year.
A duty to maintain records or adopt a policy, a duty to disclose records under the California Public Records Act (CPRA), and evidence of actual compliance are three different questions.
Original materials referenced by the investigation but not available here include the redacted Axon audit trail, records-request and deficiency correspondence, and Moorpark City Manager correspondence. Public receipts must protect requester-identifying information.
Original charts and a closer look at the evidenceOriginal report charts, comparison methods and the full-size image index.
The original report graphics are all here, with their labels, explanations, sources and full-size views. The image index below also includes the deeper analysis.
Out-of-state data: all state totals, methods and source limits
ANALYSISMatching organization names to likely states
Texas. Florida. Ohio. The interstate context matters.
The examined profile contains 53,277 Texas-associated entries, 26,810 Florida-associated entries, 20,275 Ohio-associated entries and 47,312 Georgia-associated entries. States are assigned from the organization names in the files. The separate organization profile adds frequency, persistence and requested-window context.
The daily evidence identifies an aggregate other-state surge. The state map compares audit-label populations by candidate-associated geography.
Verifying each state’s trend over time requires the underlying name-to-state classifications and dated entries, using the same periods and totals as the report.
State profile and broader map: source limits
The original map remains intact as an analytical receipt. The controlled supplemental profile includes 20,275 Ohio-associated audit observation rows.
This original map classifies primary-period-eligible Network Audit rows by candidate-associated state. It compares populations rather than trends over time. The February–March surge appears separately in the daily-regime graphic. Unknown or multi-state labels remain separate.
Source: Praeva Analytics candidate-state map; totals are transcribed from the analytical report, not independently recalculated. Original name-to-state classifications and dated state-level tables are not available here.
How often did these names appear, for how long, and how far back did searches look?
The graphic compares export rows, observed persistence and requested time windows for a selected non-California label population. The controlled supplemental profile includes 20,275 Ohio-associated audit observation rows. Review flags identify labels for further examination.
Source: Praeva Analytics organization profile; Civic Review Packet pp. 6–7 explains the screening approach. Underlying classification and query-level tables are not available here.
NCRIC connects a region—not just one local agency.
The graphic describes NCRIC as a California regional fusion center serving the federal Northern District and Northern California coastal counties. It cites NCRIC scope and governance pages. The comparison uses entries in the report’s eligible period for six selected networks.
Source: Praeva Analytics regional-network graphic; scope references: ncric.ca.gov/about_old/ and ncric.ca.gov/ncric-hidta-executive-board/. Underlying network export is not available here.
NCRIC appears less often—but its searches could reach more networks.
Monthly attributed rows, requested scope, requested windows and activity-band comparisons describe different dimensions. The graphic warns against treating the source-period scope concentration threshold as constant. Device-hour coverage is incomplete; time ranges represent requested windows. The profile identifies review questions about query construction, practice and export coverage.
Source: Praeva Analytics NCRIC profile; Civic Review Packet pp. 24–25 explains the mid-2024 change and possible source/reporting explanations. Query-level data and original export are not available here.
More than cameras: the system that has to be accounted for
The original graphic maps dependencies across the physical layer, platform/data, and operations. Its ratings describe transition complexity. Active configuration and substitution feasibility require further records.
Source: Praeva Analytics capability-dependency graphic accompanying the analytical report.
Bars count monthly network-audit rows; the line counts distinct raw organization labels. These are exported populations, not verified people, searches, recipients or search direction. The chart establishes structural change, not cause. March 2026 is a partial-period export.
Source: Praeva Analytics monthly graphic and monthly_network.json.
An activity file is not a complete account of who did what.
The original matrix organizes documented observations, interpretations and the records needed to answer questions about access, sharing, purpose and results.
Source: Praeva Analytics evidence-boundary matrix accompanying the report; original exports are not available here.
The original comparison covers training, compliance audits, retention, sharing agreements, equipment tracking and complete access records. The accompanying policy and contract references require repository-specific reconciliation.
Source: Original Praeva Analytics public-control reconciliation graphic; project analysis of policy, grant, contract and public-records materials.
How VCSO compares with other agencies
ANALYSISExplained in the report; original agency-by-agency data not included
The comparison group changes the answer
In several groups, VCSO ranks high for how often it appears and how far back its searches looked. The comparison is more mixed for how many networks and cameras its searches typically could reach.
Comparison groups shown in the original chart. Group sizes are copied from it, not recalculated; n means the number of agencies.
Comparison group
Number of agencies
Why compare them?
California sheriffs
n = 40
Agencies with a similar role
Activity band (0.5×–2× VCSO)
n = 4
Similar activity totals, not necessarily similar duties
Population band (±25%)
n = 4
Agencies serving similarly sized populations
Nationwide sheriffs
n = 848
A broader comparison
ANALYSISContext depends on the comparison groupIn several comparison groups, VCSO appears often, over a long span, with searches looking far back. Its typical numbers of networks and cameras are lower.Open original image (new tab): Context depends on the comparison groupWhat this chart means—and how it was made
The four panels compare California sheriffs, an activity band, a population band and nationwide sheriffs. Cohorts vary greatly in size, so percentile rankings are not interchangeable. These candidate-qualified comparisons describe the export population.
Source: Praeva Analytics peer-percentile graphic; Civic Review Packet pp. 10–11 and methods p. 31. Comparison-group definitions and sizes appear in the image; raw membership and metric tables are not available here.
These rankings describe the comparison, account for ties and are not grades. Small groups can change greatly when different agencies are included. The full agency lists and how they were grouped still need to be checked.
Source: Civic Review Packet, pp. 10–11 and methods p. 31. Report explanation, not raw comparison-group data.
Read this project-provided timeline alongside the report and legal receipts; it is not a substitute for the original procurement, grant or contract records.
Source: Praeva Analytics funding timeline accompanying the analytical report.
All original charts · 28 PNG images
Each link opens its original image, without cropping or recompression.
Story infographic originals and evidence notesFull-resolution editorial artwork, Chapter 8 receipts and evidence limits.
These seven editorial graphics accompany the investigation. The notes distinguish measured findings from illustrated claims that remain unverified.
Chapter 8: graphics and source receipts
Official VCSO 2025 Military Equipment Annual Report — April 22, 2026
DOCUMENTED Official Ventura County Sheriff’s Office report, dated April 22, 2026, covering January 1–December 31, 2025. Its reported inventories and usage are agency-reported figures, not independent operational verification.
UNRESOLVED70 “Homeless related operations”: What did those drone operations involve? Who approved them? How long was data kept, and which systems could receive it? These events cover the overall program, not Wingtra alone.
UNRESOLVED81 Organized Retail Theft operations. What UAV workflows and safeguards were used? The report does not document any relationship between these UAV events and Flock or Axon.
UNRESOLVEDApproximately $15,229 for “network video upgrades, enhanced recorders and associated equipment.” This appears under ROOK/tracked-armored-vehicle equipment costs, from the existing FY 25/26 budget (printed page 5; PDF page 6). Ask what equipment and governing rules it covered. The report does not establish a relationship to ALPR, Axon or Fusus. It is a follow-up lead only.
The Chapter 8 comparison is a governance recommendation, not a claim that ALPR is covered by AB 481 or that a statutory reporting violation occurred.
Axon switcheroo: sources, original graphics, and evidence limits
The graphics illustrate possible connections, not a verified local system map. Vendor capability is not confirmed VCSO deployment. The cell-site-simulator purchase and integration remain unverified.
This is a conceptual illustration, not a verified feature inventory for a particular ALPR device or a map of VCSO’s deployed system. People or objects visible in an image are not necessarily identified, indexed or separately searchable. VCSO-specific records are needed to establish video capture, enabled tools and actual search results.
The lines show possible ways to get access, not completed searches, returned information or verified local connections between vendor platforms. Dated agreements and settings records are needed to check federal, out-of-state and international access. The California-agency label represents an attributed assurance, not a verified inventory of all current or historical sharing.
Flock Organization and Network Audit Logs and an Axon audit trail were received. “Not provided” means missing accountability detail, not no activity files or proof that a record does not exist. Original presentation and response records are needed to verify the quoted wording and categories.
These are possible users and connections, not a checked list of who has access now. The 58-agency statement is what VCSO said at a particular time, not the full history of settings or use.
The measured series counts entries and organization names. The artwork’s reach and network lines are illustrative, not calculations from the daily/monthly datasets. March 2026 is partial.
5.2 · Frequency and Requested Scope Are Different Questions
The report compares NCRIC across January–July 2024 and August 2024–February 2026. The artwork uses October 2024–March 2026; its ratios (~70% fewer searches, ~3× scope, ~4× networks), camera counts and international access remain unverified. What a search asked to cover is different from what it actually showed.
This image brings the argument together. The findings remain tied to the records and measurements explained above; the illustrations of access and growth are not additional findings.
The labels tell you what supports each statement: DOCUMENTED = backed by a record provided or an official source, with the source limits stated; VCSO SAYS = the agency’s account; PLATFORM CAPABILITY = a vendor-described function, not confirmed VCSO deployment; ANALYSIS = a conclusion drawn from the available evidence; UNRESOLVED = the available records leave the question unanswered.
