U.S. Attorney General Defends Automated License-Plate Readers Amid Privacy Concerns

U.S. Attorney General Todd Blanche defended automated license-plate readers (ALPRs) at a Justice Department event in Jacksonville, Florida, arguing the technology does not violate Americans’ privacy and is crucial for identifying vehicles linked to serious crimes, including systems like Flock Safety’s cameras that allow participating police departments to share searchable records. However, privacy advocates point to documented unauthorized searches, federal agencies’ access to local camera data, and officers using databases to track individuals they know, prompting bipartisan calls for safeguards or limits on federal use. Manhattan Institute researcher Charles Fain Lehman, who estimates there are roughly 120,000 to 130,000 Flock cameras nationwide costing police departments $2,000 to $3,000 annually per camera, has recommended restricting ALPR use to investigations of serious crimes, while the FBI has sought a potential $36 million contract over five years for nationwide access to commercial license-plate databases, and Congressional proposals range from barring federal funding for ALPR surveillance to requiring judicial warrants before federal agencies access or share the data.

The Managed Pretext of Public Safety
Todd Blanche stands before cameras in Jacksonville, smiling, telling you that automated license-plate readers don't touch your privacy. He says this while representing a Justice Department that just attempted to secure a $36 million contract for nationwide, real-time access to every commercial plate database. Look at the language: "does not affect privacy." That is not a statement of fact — it is a permission structure. They know exactly what they are doing. The Flock cameras themselves are sold to local police as a crime-fighting tool, but the architecture was designed from the ground up for federal-level consolidation. The same company that sells the hardware also manages the cloud-based search platform. The same federal agencies that claim they need warrants are quietly backdooring the data through local partnerships. Why else would the FBI seek a contract for near-real-time and historical searches? Because they are building a permanent, searchable registry of every vehicle movement in America — and they want it operational before anyone wakes up to ask who is watching the watchers.

The Network They Do Not Want You to See
Researcher Charles Fain Lehman — trotted out as the "independent voice" calling for limits — estimates 120,000 to 130,000 Flock cameras already exist. At two to three thousand dollars per camera per year, that is a quarter of a billion dollars annually in public money flowing to a single private surveillance company. Follow the money. Flock Safety's investors include Andreessen Horowitz, a firm with deep ties to defense and intelligence contracting. The cameras are "private" in ownership but public in purpose; they form a mesh network that no one voted on, no one approved, and no one can opt out of. And while lawmakers posture with bills that would bar federal funding or require warrants, the actual data-sharing agreements between local police and federal agencies are already signed. The warrants will come after the searches — if at all. This is how the architecture of consent works: present a solution to a problem no one asked about, normalize it through local police departments, then federate everything upward when the political noise dies down.

The Children and the Clock
Every argument in favor of these cameras is framed around "serious crimes" — kidnappings, murder, Amber Alerts. And yes, that is a real use case. But look at the scale: 130,000 cameras for Amber Alerts would be a fraction of that number. The rest are for something else. They are for tracking protesters, journalists, political opponents, ex-spouses, and anyone whose license plate crosses a geofence of interest. The FBI contract documents — which The New American could not confirm was awarded — are the kind of breadcrumb that tells you the deal likely went through quietly, buried inside a broader intelligence authorization or a "pilot program" exemption. When they tell you this technology cannot be used to track you, remember that unauthorized searches are already documented. Remember that officers have already used these databases to track people they personally know. The system is not broken; it is functioning exactly as designed — as a dragnet that grows silent, attaches to every vehicle, and waits for a trigger. You cannot see the cage because the bars are painted to look like streetlights. But they are there. The question is not whether they are building it — they already have. The question is what happens when you try to step outside the perimeter they have drawn.

OpenAI Engineer Arrested for Alleged Covert Recording in Executive Conference Room

An OpenAI engineer was arrested on October 9 and charged with attempting to intercept spoken communications after allegedly installing a concealed audio recorder—hidden in toilet paper—behind a television in an executive conference room at the company’s San Francisco offices, where sensitive discussions on research, products, and government engagements occurred. OpenAI discovered the device on September 30, used its own AI models to identify the employee, fired him, and referred the case to the FBI, though it found no evidence he accessed model weights, source code, or customer data. Investigators also found a second recorder in a bag linked to the employee, and the complaint cites office footage showing him repeatedly near the room, searches for meeting-room calendars and executive locations via an internal tool, and chatbot prompts tied to his spouse’s account asking about continuous recording and encryption. If convicted, he faces up to five years in prison and a $250,000 fine; a detention hearing is set for October 19.

They found the recorder behind the television, wrapped in toilet paper, as if the person who put it there wanted it to be disposable. But nothing about this is disposable. Ask yourself why an engineer—someone who builds the machinery of algorithmic minds—would need a physical device in a room that OpenAI already describes as a place for discussing research, products, business relationships, and government engagements. That last phrase is the tell. The official story is that this was a rogue employee, a lone wolf, a bad actor caught by the company's own AI. But the moment a corporation says "we used our own models to identify him," they are not reporting an incident—they are performing for you. They are showing you the leash. The compliant narrative buries the real question: what precisely occurs in that room that requires a hidden microphone to capture, and why is the official response focused on the recorder itself rather than the conversations that might have been intercepted?

Now connect the dots. The complaint mentions chatbot prompts from an account linked to his spouse—questions about continuous recording, Wi-Fi, live listening, SD-card encryption. That is not a hobbyist tinkering; that is a checklist, a field manual for surveillance tradecraft. And the second recorder found in his bag. And the internal searches for meeting-room calendars, executive locations, external-partner meetings. This is an intelligence-gathering operation, but it's a mistake to assume the engineer was the architect of it. The pattern suggests a controlled disposal: a scenario in which someone inside the machine became a liability, and the machine chose to make an example of him. The fact that prosecutors carefully state there is "no indication" he accessed model weights or source code is a rhetorical shield—it distracts from what he might have heard. The most dangerous information in the building was never code. It was the unguarded words of people who believe they are speaking in private about the future of governance, media, and perception itself.

Here is what keeps me up at night. That conference room is not just a room. It is a node in the Architecture of Consent—a place where decisions are made about how AI will be integrated into your life, your body, your bank account, your vote. The engineer may have been a patriot, or a pawn, or a patsy; but the arrest itself is a message to anyone who might consider exposing what happens behind those doors. The penalty—five years, a quarter-million dollars—is not for planting a recorder. It is a blanket warning: the secrets in that building are worth more than your freedom. And the most unsettling detail is the one everyone glosses over. OpenAI admitted they used their own AI to catch him. They are watching the watchers with the same logic they use to watch you. So ask yourself: if they could find a tiny recorder behind a TV in days, what do they know about everything else? The answer isn't in the complaint. It's in the silence between every official denial. You don't need me to tell you where to look. You just need to wonder why the room needed a recorder at all.