Thornton PD Investigated Itself and Found no Wrongdoing
Focusing only on the overall search count, not the substantive issues, Thornton PD cleared itself of wrongdoing.
Thornton PD audited whether one of its officers ran a lot of Flock searches. It did not audit whether those searches were proper. The Northglenn-Thornton Sentinel reports that “Thornton Police determine officer’s 10,318 searches were part of his job” — but the department’s response addresses only the volume of searches, not their substance.
The original complaint by Thornton For All alleged 19,194 searches, based on information from haveibeenflocked.com. Thornton PD acknowledged the complaint, then set the underlying data aside:
When evaluating this third-party website data, it was clear that the issues articulated in the website’s disclaimer were significant. Instead, our audit focused on internal system records of actual usage by this officer.
haveibeenflocked.com cautions against treating its data as authoritative for several reasons:
- Information in Flock audit logs is unreliable.
- When logs are modified — such as through redaction — they may appear as duplicates.
- When external (network) log information is used, not all searches for an agency may be captured.
These limitations cut in multiple directions. Redaction can multiply searches — the same search by officer “A” in one log and by officer “REDACTED” in another is counted twice. At the same time, Thornton has not published its logs, so the only searches available are those that happen to appear in network logs from other agencies. That’s likely resulting in underreporting.
The discrepancy between 19,194 and 10,318 remains unresolved, and Thornton PD has not published the reconciliation — only asserted that its internal number is the correct one.
The numbers, however, are the smaller question. The substantive issues are what Thornton For All named:
… the department’s letter does not provide an explanation for the irregular search activity that occurred outside of normal working hours. Furthermore, the response does not address the targeted, long-term surveillance of a single license plate that was tracked for up to 145 days
The response also does not appear to address nationwide location history retrievals, often covering multiple months, justified in the logs by entries such as:
“Mexico Plate” · “plate” · “Misuse” · “No record” · “Ebb813b” · “See if stolen” · “n” · “no”
The logs as published cannot support a determination that these were legitimate investigations. Whatever else Thornton PD’s audit reviewed, it was not this record.
That single license plate mentioned, YZ6717D, was tracked over nearly 6 months. Flock’s 30-day retention period is marketed as a meaningful privacy guardrail; 145 days of continuous tracking on one plate moots it entirely.
Conducting long-term warrantless surveillance while representing to the public that retention limits are a real protection is the kind of thing that damages valuable community trust.
And on that issue, the Sentinel quotes Police Chief Baird:
When unverified and inaccurate information circulates, this can negatively shape public perception and damage valuable community trust. I am sharing these findings to provide necessary detail and context, as well as to ensure ongoing public discussion is grounded in fact
We could not agree more. Disclosure of accurate, complete information is essential in building community trust.
“This officer’s dedication to public safety objectives, as well as his tenacity in locating vehicles associated with victimizing members of our Thornton community, is clearly evident,” Baird states.
That’s where we will have to disagree. It is not at all clearly evident from searches justified by “no” or “Mexico plate”.
Of course, if Chief Baird wants to make it clearly evident by publishing complete, unredacted search logs — along with the basis on which “no” or “Mexico Plate” were determined to be proper uses of the system — we would be more than happy to publish that information here.