Dear editor,
Liberty’s debate over license plate cameras deserves a public audit that measures what the technology accomplishes, who can use its data, and how the city corrects mistakes.
Bluebonnet News reported that the City Council voted 5-2 on October 13 to renew its $27,000-a-year contract for nine Flock Safety cameras. Supporters cited help recovering stolen vehicles, finding missing juveniles, and investigating crimes. Opponents raised specific questions about access by outside agencies and the handling of residents’ information. Both concerns merit evidence rather than assurances.
I advise organizations on introducing AI into consequential workflows. The first question I ask is who can review and override a tool’s output when something goes wrong. A camera hit can be useful evidence, but it should never become an automatic verdict about the person behind the wheel.
The city should publish an independent quarterly audit that distinguishes the cameras’ useful leads from the cases they actually helped resolve. How many alerts generated investigative leads? How many were dismissed after human review? How many proved inaccurate, and how were those records corrected? Public reporting can use aggregate numbers and protect active investigations.
A second audit should follow access to the data. Council members need a count of searches performed by Liberty personnel and by outside agencies, the purposes those agencies certify, the retention rules, and the procedure for investigating questionable searches. Residents should learn what happens when a government employee or vendor violates the policy. A transparency portal provides a start; independent verification would make it credible.
Third, the city should make vendor commitments enforceable. Before approving a new feature, officials should publish its purpose, whether it adds automated identification or cross-database searches, who approves each use, and what happens to records when the contract ends. The federal AI Risk Management Framework offers a useful voluntary structure for identifying and measuring risks, assigning responsibility, and reviewing the safeguards. It does not decide Liberty’s policy for the council.
As Councilman Ross Ward argued in his October 16 commentary, the possible expansion of surveillance capabilities deserves scrutiny before any additional features arrive. The city need not assume that every capability a company markets is already deployed locally. It should establish what Liberty is actually purchasing and using.
After a defined review period, council members should publicly decide whether documented investigative benefits justify the costs, data access, and remaining risks. If the city cannot produce reliable answers, it should reconsider the contract. If it can, residents deserve to see the evidence.
The public owns the consequences of this decision. It should be able to inspect the reasoning behind it.
Gleb Tsipursky, PhD, a behavioral scientist, CEO of Disaster Avoidance Experts, and author of The Psychology of AI Adoption at Work: From Resistance to Results