Flock Cameras Face Senate Hearing After Woman Spent 13 Days in Jail When AI Wrongly Matched Her Car

Flock cameras came under scrutiny at a US Senate hearing after Florida woman Lindsey Isaacs spent 13 days in jail following an investigation in which Flock camera data was used to identify her vehicle in connection with a fatal crash.

Isaacs told senators that she was the subject of a false match, while prosecutors later dropped the charges against her.

The hearing, held on 23 September, was chaired by Republican Senator Josh Hawley as part of the Senate Judiciary Subcommittee on Crime and Counterterrorism's examination of Flock's nationwide network of automated licence-plate readers.

Hawley's office said the hearing focused on privacy, national security and the risks posed by AI-enabled surveillance. Flock's chief executive and the CEOs of Axon, Motorola Solutions and Verkada were invited to testify but declined.

Flock Cameras Put Under the Senate Spotlight

Hawley said Flock operates about 120,000 cameras across 49 states and generates roughly 20 billion vehicle images or scans each month. He argued that linking the cameras through searchable AI systems could allow authorities to reconstruct people's movements over extended periods.

Alasdair Whitney, legislative counsel for the Institute for Justice, told the hearing that the interconnected system was qualitatively different from older licence-plate reader technology because officers could potentially reconstruct vehicle movements over weeks or months.

'You are not exaggerating. That is exactly right,' Whitney told Hawley when asked whether the technology could be used to track people's movements over long periods.

Whitney's comments were an assessment of the technology's capabilities, rather than a finding that authorities routinely use Flock systems in that way.

Flock has also faced questions over the security of its systems. Cybersecurity engineer Benn Jordan disputed the company's claim that Flock had never been hacked, telling senators he had found instances of unauthorised access.

Jordan said some Flock devices he tested were running an old version of Android, while newer devices used a modified Android kernel that he described as falling short of government-grade security.

He also testified that he had found an administrative interface online that, he claimed, allowed him to access live camera systems and delete footage. Those statements were made during the hearing and should not be treated as an independent finding that Flock's entire network has been compromised.

Flock has previously disputed claims about its systems being hacked and has said some reported security incidents involved configuration problems that were subsequently addressed.

Isaacs Details the 13-Day Jail Ordeal

Isaacs' experience provided the most personal account of the potential consequences of an incorrect vehicle match.

Hawley showed the Senate a photograph of Isaacs with her Dodge Durango and asked about its colour. She confirmed that her vehicle was black.

The vehicle investigators were looking for after a fatal crash had been described as maroon or red. Hawley said Isaacs' vehicle was returned as a false positive during an AI-enabled Flock search based on a general description.

Isaacs agreed when Hawley said the result had led to her spending 13 days in jail, including time in solitary confinement, and had put her livelihood at risk.

The details of the investigation are more complicated than a simple AI match.

Reporting on the case indicates investigators used Flock data to identify Isaacs' black Durango during their investigation into the fatal crash. Her vehicle was later found not to have been involved, and prosecutors dropped the charges against her.

The available evidence supports saying Flock data played a role in the investigation that led to Isaacs' arrest, but it does not establish that an AI system alone caused the arrest or that Flock was solely responsible for the subsequent legal consequences.

Hawley nevertheless presented the case as an example of what he considers the risks of relying on AI-enabled camera networks without stronger safeguards.

'The Constitution is as valuable and as valid today as it was when it was first written,' he said, arguing that modern surveillance technology made protections for Americans' rights increasingly important.

Flock did not send its CEO to the hearing but provided Hawley with a written response.

The company agreed that Congress should establish a federal framework governing automated licence-plate reader technology and acknowledged that it had introduced additional safeguards following scrutiny.

Those measures include tighter controls around searches and greater restrictions on how the technology can be used. However, they do not resolve the questions raised at the hearing over how long vehicle data should be retained, who can access it and how AI-generated matches should be checked before they contribute to an arrest.